profiles = new Array(
    "Teaching the How of Cognitive Science: The Inquiry Website...||Tutorials Track Teaching the How of Cognitive Science: The Inquiry Website Adele Abrahamsen , Center for Research in Language, University of California adele@crl.ucsd.edu William Bechtel , Department of Philosophy, University of California bill@mechanism.ucsd.edu Peter Bradley , Philosophy-Neuroscience-Psychology Program, Washington University pbradley@artsci.wustl.edu Carl F. Craver , Philosophy-Neuroscience-Psychology Program, Washington University ccraver@artsci.wustl.edu Cognitive scientists are pretty adept at coordinating multiple methods and perspectives; it is what we do. Teaching other people to do it is harder, especially when the other people are whole classes of undergraduates in one of the new cognitive science majors or specializations. The challenge is to provide students with (a) a sufficient understanding of some of the methods used in the contributing fields, (b) the strengths and weaknesses of these methods, and (c) how they can be coordinated in interdisciplinary research to achieve new understanding. Inquiry is a web-based curriculum for introducing students to the range of research methods employed in cognitive science. This tutorial provides an opportunity to hear about Inquiry and the ideas it embodies, to interact with it in a group setting, and to help and be helped to improve undergraduate education in cognitive science. To promote active understanding of the research methods that are introduced, the course materials are interactive. For example, instead of just providing a definition of cognitive science, students are guided to construct their own characterization after identifying and classifying a variety of phenomena as cognitive or not. They are then challenged to test the adequacy of their characterization in light of other phenomena and characterizations advanced by other students. Such devices as animation, pop-up windows, and a dynamic menu system also increase students' engagement. The materials are organized into semi-independent modules that can be selected and recombined to meet the objectives of particular courses. To provide integration to the different methods, research on memory provides a common theme, but examples from a variety of other domains are offered as well.|../mac/prof1.html",
    "Development and Cognitive Architecture...||Symposium Track Development and Cognitive Architecture Robert Hadley , School of Computing Science, Simon Fraser University hadley@cs.sfu.ca Gary Marcus , Department of Psychology, New York University gary@psych.nyu.edu Robert Jacobs , Department of Brain and Cognitive Sciences, University of Rochester robbie@bcs.rochester.edu This symposium examines interrelationships be- tween fundamental aspects of neural/cerebral devel- opment and later cognitive architecture. Topics will include: the role of DNA in plasticity and organi- zation; the bene ts of innate, domain-speci c con- straints; as well as a defense of distributed functional modularity. Theoretical positions on these issues are advanced and defended. Recent arguments con- cerning the implications of brain-imaging studies for functional modularity are also examined.|../mac/prof10.html",
    "Perception of e-mail personality at zero-acquaintance:Extraversion takes ca...||Paper Track Perception of e-mail personality at zero-acquaintance:Extraversion takes care of itself; Neuroticism is a worry Alastair J. Gill , School of Informatics, University of Edinburgh agill@cogsci.ed.ac.uk Jon Oberlander , School of Informatics, University of Edinburgh J.Oberlander@ed.ac.uk We investigate the impact of computer-mediated in- teraction on person perception. In particular, we study how traits important for socialisation and collaboration Extraversion and Neuroticism can be detected from the text of an e-mail communi- cation. We have previously shown how Extraver- sion influences people's language production in elec- tronic communication, in broadly intuitive ways. Here, we briefly outline the ways in which Neuroti- cism is expressed more through the high-level prop- erties of a text. By their nature, these properties are less accessible to intuition. In subjective ratings of the texts for personality, we demonstrate that author Extraversion can be accurately perceived, given the limited cues, and that judges also exhibit relatively high agreement with each other for this trait. Neuroticism, however, appears more di cult. This result is consistent with previous ndings, but suggests that e-mail heightens the difficulty.|../mac/prof100.html",
    "The Distribution of People to Resources in a Networked Multi-player Environ...||Paper Track The Distribution of People to Resources in a Networked Multi-player Environment Robert L. Goldstone , Cognitive Science Program Indiana University rgoldsto@indiana.edu Benjamin C. Ashpole , Cognitive Science Program Indiana University bashpole@indiana.edu We have developed an experimental platform that allows a large number of human participants to interact in real-time within a common virtual world. Within this environment, human participants foraged for resources distributed in two pools. In addition to varying the relative replenishment rates for the two resources (50-50, 65-35, 80-20), we manipulated whether participants could see each other and the entire food distribution, or had their vision restricted to food at their own location. Two empirical violations of optimal distributions of participants were found. First, there was a systematic underutilization of the more preponderant resource. Second, there were oscillations in the harvesting rate of the resources in a pool across time, revealed by a Fourier analysis with prominent power in the range of 50 seconds per cycle. These suboptimalities were more prevalent when participants had no knowledge of other participants or the complete food distribution. Individual participant knowledge thus affects the efficiency with which a population of participants harvests resources.|../mac/prof101.html",
    "Literary Evidence for the Cultural Development of a Theory of Mind...||Paper Track Literary Evidence for the Cultural Development of a Theory of Mind Andrew S. Gordon , Institute for Creative Technologies, University of Southern California gordon@ict.usc.edu Anish Nair , Institute for Creative Technologies, University of Southern California nair@ict.usc.edu The term Theory of Mind is used within the cognitive sciences to refer to the abilities that people have to perceive and reason about their own mental states and the mental states of others. Although there continues to be debate concerning whether these abilities are grounded in inferential theories or mental simulations, an orthogonal question is whether these abilities are culturally acquired or innate to our species. This paper outlines the argument that Theory of Mind abilities are based on representational models of reasoning that are culturally acquired. To support this thesis, we present evidence gathered from the large-scale automated analysis of text corpora. We show that the Freudian conception of a subconscious desire is a relatively modern addition to our culturally shared Theory of Mind, as evidenced by a shift in the way these ideas appeared in 19th and 20th century English language novels.|../mac/prof102.html",
    "Why/AutoTutor: A Test of Learning Gains from a Physics Tutor with Natural L...||Paper Track Why/AutoTutor: A Test of Learning Gains from a Physics Tutor with Natural Language Dialog A.C. Graesser, G.T. Jackson, E.C. Mathews, H.H. Mitchell, A. Olney, M. Ventura, P. Chipman, D. Franceschetti, X. Hu, M.M. Louwerse , Institute for Intelligent Systems, University of Memphis a-graesser@memphis.edu G.T. Jackson , Rhodes College gtjacksn@memphis.edu E.C. Mathews , Institute for Intelligent Systems, University of Memphis emathews@memphis.edu H.H. Mitchell , Institute for Intelligent Systems, University of Memphis hmitchll@memphis.edu A. Olney , Institute for Intelligent Systems, University of Memphis aolney@memphis.edu M. Ventura , Institute for Intelligent Systems, University of Memphis mventura@memphis.edu P. Chipman , Institute for Intelligent Systems, University of Memphis pchipman@memphis.edu D. Franceschetti , Institute for Intelligent Systems, University of Memphis dfrncsch X. Hu , Institute for Intelligent Systems, University of Memphis xhu@memphis.edu M.M. Louwerse , Institute for Intelligent Systems, University of Memphis mlouwers@memphis.edu N.K. Person , Institute for Intelligent Systems, University of Memphis person@rhodes.edu Why/AutoTutor is a tutoring system that helps students construct answers to qualitative physics problems by holding a conversation in natural language. Why/AutoTutor provides feedback to the student on what the student types in (positive, neutral, negative feedback), pumps the student for more information, prompts the student to fill in missing words, gives hints, fills in missing information with assertions, identifies and corrects bad answers and misconceptions, answers students’ questions, and summarizes answers. In essence, constructivist learning is implemented in a mixed-initiative dialog. Why/AutoTutor delivers its dialog moves with an animated conversational agent whereas students type in their answers via keyboard. We conducted an experiment that compared Why/AutoTutor with two control conditions (Read textbook, nothing) in assessments of learning gains. The tutoring system performed significantly better than the two control conditions on a test similar to the Force Concept Inventory.|../mac/prof103.html",
    "Probability, algorithmic complexity, and subjective randomness...||Paper Track Probability, algorithmic complexity, and subjective randomness Thomas L. Griffiths , Department of Psychology, Stanford University gruffydd@psych.stanford.edu Joshua B. Tenenbaum , Brain and Cognitive Sciences Department, MIT Institute for Intelligent Systems, University of Memphis We present a statistical account of human randomness judgments that uses the idea of algorithmic complexity. We show that an existing measure of the randomness of a sequence corresponds to the assumption that non-random sequences are generated by a particular probabilistic nite state automaton, and use this as the basis for an account that evaluates randomness in terms of the length of programs for machines at di erent levels of the Chomsky hierarchy. This approach results in a model that predicts human judgments better than the responses of other participants in the same experiment. that contributes to our inferences. The structures that people nd simple form a strict (and exible) subset of those easily expressed in a computer program. For example, the sequence of heads and tails TTHTTTHHTH appears quite complex to us, even though, as the parity of the rst 10 digits of , it is easily generated by a computer. Identifying the kinds of regularities that contribute to our sense of simplicity will be an important part of any cognitive|../mac/prof104.html",
    "Developing Problem Solving Competence: A Distributed Model and a New Class ...||Paper Track Developing Problem Solving Competence: A Distributed Model and a New Class of Strategies With the Tower of Hanoi Task Bernadette Guimberteau , Cognition and Development, School of Education, University of California Bern0@mac.com The ability to focus on the largest disk of a pyramid at the outset and to define largest-disk subgoals constitute two essential aspects in most known strategies and models of problem solving with the Tower of Hanoi. Yet, those abilities are typically assumed by existing accounts. This paper presents a distributed model, which learns to focus on the largest disk of a pyramid and set subgoals to move largest disks. The model exhibits a capacity to solve 4- and 5-disk Tower of Hanoi versions optimally. Moreover, the fit between this analysis and the data from Anzai & Simon (1979) is excellent. The present model provides a new interpretation of those data. It is suggested that the subject’s optimal performance in her second problem solving attempt is due to the acquisition of task-specific afffordances and of difference-reduction strategies that are affordance-driven. The above analysis defines a new class of Tower of Hanoi strategies – based on a problem solver’s capacity to define and use task-specific affordances. The mechanisms proposed by the model can be used to examine the distributed nature of learning and problem solving in other tasks as well.|../mac/prof105.html",
    "The Proper Treatment of Semantic Systematicity...||Paper Track The Proper Treatment of Semantic Systematicity Robert F. Hadley , School of Computing Science and Cognitive Science Program Simon Fraser University hadley@cs.sfu.ca Connectionist-minded philosophers, including Clark and van Gelder, have espoused the merits of viewing hiddenlayer, context-sensitive representations as possessing semantic content, where this content is partially revealed via the representations' position in vector space. In recent work, Bodén and Niklasson have incorporated a variant of this view within their conception of semantic systematicity. Moreover, Bodén and Niklasson contend that they have produced experimental results which not only satisfy a kind of context-based, semantic systematicity, but which, to the degree that reality permits, effectively deals with challenges posed by Fodor and Pylyshyn (1988), and Hadley (1994a). This paper examines the claims of Bodén and Niklasson. It is argued that their case fatally involves a fallacy of equivocation. In addition, it is argued that their ultimate construal of context sensitive semantics employs lax, incorrect standards.|../mac/prof106.html",
    "Metrical Categories in Infancy and Adulthood...||Paper Track Metrical Categories in Infancy and Adulthood Erin E. Hannon , Department of Psychology, 211 Uris Hall eeh5@cornell.edu Sandra E. Trehub , Department of Psychology, University of Toronto at Mississauga sandra.trehub@utoronto.ca Adults tend to perceive and produce rhythmic structures with simple duration ratios and implied isochrony. If these biases result from long-term exposure to Western metrical structure, they should be evident in North American adults but not in infants. Adult similarity judgments were obtained for alterations of folk melodies that maintained or matched the original metrical structure, whether simple or complex, and for alterations that violated or mismatched the original metrical structure. Adults rated mismatching alterations as less similar than matching alterations, for simple meter patterns, but their ratings of matching and mismatching alterations did not differ for complex meter patterns. Infants, who were tested with a familiarization preference procedure, responded differentially to matching and mismatching alterations for complex as well as simple meter patterns. These findings imply that adult biases reflect long-term exposure to culture-specific metrical structure rather than intrinsic preference for simple meters.|../mac/prof107.html",
    "Naïve Sampling and Format Dependence in Subjective Probability Calibration...||Paper Track Naïve Sampling and Format Dependence in Subjective Probability Calibration Patrik Hansson , Department of Psychology, Umeå University patrik.hansson@psy.umu.se Peter Juslin , Department of Psychology, Umeå University peter.juslin@psy.umu.se Anders Winman , Department of Pschology, Uppsala University anders.winman@psyk.uu.se|../mac/prof108.html",
    "Structural Priming in Sentence Comprehension...||Paper Track Structural Priming in Sentence Comprehension Michael Harrington , Linguistics Program, EMSAH, University of Queensland m.harrington@uq.edu.au Simon Dennis , Institute of Cognitive Science, University of Colorado Simon.Dennis@colorado.edu This study examines structural priming in an on-line reading task that compares reading performance by first and second language readers of English. Robust effects for priming were evident for the second, but not first, language readers, suggesting that priming may occur where there are insufficient sentence traces in memory to support sentence interpretation. The empirical results are simulated using the Syntagmatic Paradigmatic (SP) model, a distributed, instance-based account of sentence comprehension.|../mac/prof109.html",
    "Representation and Activity...||Symposium Track Representation and Activity Participants: Timo Honkela, Helsinki University of Technology Jonna Kangasoja, University of Helsinki Tarja Knuuttila, University of Helsinki A crisis of representation has undoubtedly swept over philosophy, social sciences and humanities lately. One of its main targets has been representationalism according to which to know is to have an accurate reproduction of the object of knowledge in the mind (in the form of suitable representations). Instead of mental representation we focus on representation-in-action: we approach representation as goal-directed collective activity that uses material external means. The symposium is interdisciplinary and multiplex due to our di erent disciplinary backgrounds (philosophy, arti cial intelligence and education). The participants' topics to be discussed advance from the more abstract to the more concrete. The symposium starts with a philosophical criticism of the more or less received notion of representation, which is conceived of as a dyadic relation between external things (real entities and processes) and their mental, abstractor theoretical pictures. External representations, according to this view, are treated implicitly as transparent replicas of either external objects or internal representations. Their use and usefulness is not questioned or explicated. In contradistinction to this view we propose that to know is to construct and use external, semiotic devices of the phenomena in interest, in which process the object of knowledge merges with its representation. The next presentation takes these general philosophical themes to the level of cognitive and computational modeling. Attempts to create models of activity, e.g., as societies of arti cial agents, bring in the intersubjective or cultural level. If the agents are to communicate meaningfully with each other one needs to assume a Platonian worldview somewhat automatically shared by all the agents, or then the system of a collection of agents has cultural level of meaning negotiations and emergent conceptual systems.|../mac/prof11.html",
    "Why Believability Cannot Explain Belief Revision...||Paper Track Why Believability Cannot Explain Belief Revision Uri Hasson , Psychology Department, Princeton University uhasson@princeton.edu Philip N. Johnson-Laird , Psychology Department, Princeton University phil@princeton.edu A common view in epistemology is that some beliefs are more entrenched than others. This view is plausible, but we show that it fails to explain which statements individuals tend to doubt when an incontrovertible fact is inconsistent with the relevant set of statements. We report three studies that each show that the believability of statements is influenced by context. Given a conditional of the form If P then Q and a categorical statement P, individuals tend to judge the categorical as more believable than the conditional. But, when the same statements are followed by an incontrovertible fact, not-Q, that is inconsistent with them, individuals tend to judge the conditional as more believable than the categorical. The theory of mental models accounts in part for these and other results of the experiments, including a study of the believability of exclusive disjunctions and categoricals.|../mac/prof110.html",
    "Why Does Similarity Correlate With Inductive Strength?...||Paper Track Why Does Similarity Correlate With Inductive Strength? Uri Hasson , Psychology Department, Princeton University uhasson@princeton.edu Geoffrey P. Goodwin , Psychology Department, Princeton University ggoodwin@princeton.edu It has been repeatedly demonstrated that a robust predictor of the strength of an inductive argument is the similarity between the categories that are the focus of the induction. In this paper we evaluate why similarity increases the strength of such arguments. One on view, category similarity makes an argument strong because it affords similarities that provide reasons to think that the conclusion is justified. On another view, similarity may afford few differences, so that there are not many reasons to think that the conclusion is unjustified. We evaluate this issue by examining how engagement in inductive reasoning affects the perceived similarity between categories. Our findings suggest that people attempt to find reasons to disbelieve arguments. They consider differences when evaluating affirmative inductions, and consider similarities when evaluating negative inductions. This is done independent of whether one is evaluating the truth or falsity of the induction.|../mac/prof111.html",
    "Creativity and Constraints: The Creation of Novel Sentences...||Paper Track Creativity and Constraints: The Creation of Novel Sentences Catrinel Haught , Department of Psychology, Green Hall, Princeton University chaught@princeton.edu Philip N. Johnson-Laird , Department of Psychology, Green Hall, Princeton University phil@princeton.edu Two experiments explored how people create novel sentences referring to given entities presented either in line drawings or in nouns. The line drawings yielded more creative sentences than the words, both as rated by judges and objectively in Shannon’s measure of the amount of information that the sentences conveyed. A hypothesis about the cognitive processes of creation predicted this result: creativity depends on constraints. Line drawings of entities present more information about them than nouns denoting the same entities, and so the pictures provide more constraints than the nouns. Hence, line drawings yield more creative sentences than words.|../mac/prof112.html",
    "Co-operation and Co-ordination in the Production of Noun Phrases...||Paper Track Co-operation and Co-ordination in the Production of Noun Phrases Sarah Haywood , Psychology department, University of Edinburgh Sarah.Haywood@ed.ac.uk Martin J. Pickering , Psychology department, University of Edinburgh Martin.Pickering@ed.ac.uk Holly P. Branigan , Psychology department, University of Edinburgh Holly.Branigan@ed.ac.uk We investigate the influence of two processes on noun phrase production during a referential communication task: priming (co-ordination) and audience design (co-operation). Participants played a collaborative communication game that involved describing picture cards (Experiment 1) or wooden blocks (Experiment 2). They interacted with a confederate who was scripted to produce particular kinds of noun phrases. As in previous studies, participants tended to be primed by the structure of the confederate’s preceding description. However, they also showed evidence of tailoring their descriptions to make it easier for their addressee to carry out the task at hand.|../mac/prof113.html",
    "Experimenting with Clarification in Dialogue...||Paper Track Experimenting with Clarification in Dialogue Patrick G.T. Healey , Department of Computer Science; Queen Mary, University of London ph@dcs.qmul.ac.uk Matthew Purver , Department of Computer Science; Kings College London matthew.purver@kcl.ac.uk James King , Department of Computer Science; Queen Mary, University of London jking@dcs.qmul.ac.uk Jonathan Ginzburg , Department of Computer Science; Kings College London jonathan.ginzburg@kcl.ac.uk Greg J. Mills , Department of Computer Science; Queen Mary, University of London gj@dcs.qmul.ac.uk A new technique for integrating experimental manipulations into text-based, synchronous, dialogue is introduced. This method supports the systematic transformation of conversational turns and the use ‘artificial’ probe turns and turn sequences. These manipulations can be made in a way that is sensitive to aspects of the local linguistic and conversational context. This technique is used in an experimental investigation of the e ect of word category and level of grounding on the interpretation of reprise clari- fications. The results show that both factors influence the type and likelihood of response to reprise fragment clarifications.|../mac/prof114.html",
    "Cognitive Design Principles for Visualizations: Revealing and Instantiating...||Paper Track Cognitive Design Principles for Visualizations: Revealing and Instantiating Julie Heiser , Department of Psychology, Stanford University jheiser@psych.stanford.edu Barbara Tversky , Department of Psychology, Stanford University bt@psych.stanford.edu Maneesh Agrawala , Microsoft Research maneesh@graphics.stanford.edu Pat Hanrahan , Department of Computer Science, Stanford University hanrahan@graphics.stanford.edu Visualizations can be improved by cognitive research. Effective visualizations convey the desired cognitions clearly and accessibly. Cognitive experiments can provide design principles that can be instantiated into computer algorithms. We describe a collaboration between psychology and computer science to uncover design principles for assembly instructions and capture them in algorithms. Good visualizations for assembly show step-by-step action diagrams with 3-D views that best show the assembly steps. These principles extend to visualizations showing how to operate devices and how systems work. The techniques of extracting design principles extend to other domains.|../mac/prof115.html",
    "Animal Models of Expertise Development...||Paper Track Animal Models of Expertise Development William S. Helton , Department of Psychology, Wilmington Deak_Helton@Wilmington.Edu There is a continuing debate in the psychological literature between those who lean more towards learning theories of expertise development and those leaning more towards talent theories. However, the development of human expertise has not been open to direct experimental methods and will probably continue to elude experimentalists in the future. A promising alternative is to employ non-human animal models. Expertise researchers have seemingly overlooked this possibility. However, there are studies in the animal behavior literature that address the development of non-human animal expertise without specifically referring to the topic as expertise. I will discuss two non-human animal examples of expertise development that have been researched by ethologists. Non-human animal expertise development, unlike human expertise development, is subject to direct experimentation. Hence, I recommend initiating expertise research with non-human animals.|../mac/prof116.html",
    "Proximity in agreement errors...||Paper Track Proximity in agreement errors Barbara Hemforth , Laboratoire Parole et Langage, Univ. de Aix en Provence barbara.hemforth@lpl.univ-aix.fr Lars Konieczny , Center for Cognitive Science, IIG, Univ. Freiburg lars@cognition.iig.uni-freiburg.de Across languages subject-verb agreement errors have been established when subjects are producing complex (NP PP) noun-phrases (see Bock, 1995, for an overview). Very recently, Haskell & MacDonalds (2002) proposed a locality-based principle, the principle of proximity, to explain a variety of agreement errors in production. They base they argument partly on preferences in verb number marking in sentences with disjunctive noun phrases (e.g., the hat and the gloves vs. the gloves and the hat), where they found a preference for number marking that matched the local noun. We will present a series of five written production experiments on German constructions with disjunctive Subjects, NP PP Subjects and Subject- Object-verb subordinate clauses. We will show that - although comparable effects can be established for German sentences with disjunctive Subjects - a proximity based principle fails to cover major portions of the results. We propose an account that highlights the dynamics of plural-feature activation and percolation.|../mac/prof117.html",
    "Semantic Illusion in Sentence Processing: a Right-Hemisphere Mechanism?...||Paper Track Semantic Illusion in Sentence Processing: a Right-Hemisphere Mechanism? John C. J. Hoeks , Behavioral and Cognitive Neurosciences hoeks@let.rug.nl Ingeborg Prinsen , Behavioral and Cognitive Neurosciences prinsen@let.rug.nl Laurie A. Stowe , Behavioral and Cognitive Neurosciences stowe@let.rug.nl Recent research on sentence processing using ERPs (Event Related brain potentials) has shown that there are situations in which the semantic relationships between words in a sentence are so strong that they can block the semantic interpretation that is actually prescribed by the syntactic structure of that sentence (Hoeks, Stowe, & Doedens, 2003; Kolk, Chwilla, van Herten, & Oor, in press). As syntactic processing is the assumed province of the left hemisphere (LH), it was hypothesized that this so-called 'semantic illusion' might result from a transient but apparently rather influential non-syntactic sentence representation formed in the right hemisphere (RH). Two reaction time experiments using the Divided Visual Field paradigm only partially supported this hypothesis, as they showed that it is the LH that is most sensitive to semantic illusion.|../mac/prof118.html",
    "Framework for Modeling Partial Conceptual Autonomy of Adaptive and Communic...||Paper Track Framework for Modeling Partial Conceptual Autonomy of Adaptive and Communicating Agents Timo Honkela , Laboratory of Computer and Information Science, Helsinki University of Technology timo.honkela@hut.fi Kevin Ilmari Hynna , Laboratory of Computer and Information Science, Helsinki University of Technology Tarja Knuuttila , Center for Activity Theory and Developmental Work Research, University of Helsinki tarja.knuuttila@helsinki.fi We consider the idea of conceptual autonomy of natural and arti cial agents. We claim that agents that have explicit, propositional representations are limited in their conceptual autonomy. We discuss how partial conceptual autonomy is obtained through an self-organization process. The input for the agents consists of perceptions on the context, expressions communicated by other agents as well as recognized identities of the other agents.|../mac/prof119.html",
    "Redefining Collaboration to Make it Work in Classrooms...||Symposium Track Redefining Collaboration to Make it Work in Classrooms Naomi Miyake and Hajime Shirouzu , School of Computer and Cognitive Sciences, Chukyo University nmiyake@sccs.chukyo-u.ac.jp Janet L. Kolodner , College of Computing, Georgia Institute of Technology janet.kolodner@cc.gatech.edu , Margaret Jacks Professor of Education, Stanford University This symposium tries to bridge the newly emerging learning sciences and the cognitive sciences, by focusing on the definition of collaboration. It also raises the possibility of a need for a new science. In the last 20 years, cognitive studies on collaboration have yielded implications to raise the quality of learning in real classrooms. Studies on classroom learning have in turn taught us about real processes of learning that are often more complex and sustainable than what has been observed and studied in experimental settings. This trend of mutual stimulation between basic cognitive studies and their real world implementation has recently evolved into the birth of the International Society of the Learning Sciences. This symposium is one trial of us to strengthen chances for mutual learning for the cognitive science researchers.|../mac/prof12.html",
    "Predicting Cognitive Strategies and Eye Movements in Hierarchical Visual Se...||Paper Track Predicting Cognitive Strategies and Eye Movements in Hierarchical Visual Search Anthony J. Hornof , Department of Computer and Information Science, University of Oregon hornof@cs.uoregon.edu Tim Halverson , Department of Computer and Information Science, University of Oregon thalvers@cs.uoregon.edu This article advances computational cognitive modeling of visual search, and the synergistic relationship between cognitive modeling and eye tracking. The paper presents cognitive models of the perceptual, cognitive, and motor processing involved in the visual search of a hierarchical layout. Two types of visual layouts are searched: unlabeled layouts in which words are arranged in groups but with no hierarchical organization, and labeled layouts in which each group is given a heading that guides the search. The two types of layouts motivate fundamentally different search strategies. The models are post hoc explanatory models of the search time data and a priori predictive models of the eye movement data. The models are evaluated here based on the eye movement data. The research demonstrates a methodology and provides guidance for predictive cognitive modeling of visual search.|../mac/prof120.html",
    "A Split Model to Deal with Semantic Anomalies in the Task of Word Predictio...||Paper Track A Split Model to Deal with Semantic Anomalies in the Task of Word Prediction Janet Hui-wen Hsiao , Division of Informatics, University of Edinburgh h.hsiao@sms.ed.ac.uk On the task of predicting the range of possible next words in a sentence, many networks (e.g. Elman, 1990, 1993) that have been proposed are capable of displaying a certain degree of systematicity, but fail in recognizing grammatically correct but semantically anomalous sentences. Based on an expansion of Hadley’s model (Hadley et al, 2001), we present a competitive network, which employs two sub-networks that discern coarsegrained and fine-grained categories respectively, by being trained via different parameter settings. Hence, one of the sub-networks will have a greater capacity for recognizing the syntactic structure of the preceding words, while the other will have a greater capacity for recognizing the semantic structure. This corresponds to the recent suggestion about specialization of the two hemispheres in the human brain (Beeman, 1998). Also, we employ a mechanism to switch attention between the predictions from the two sub-networks, in order to make the global network more closely approximate human behavior. The results show that the network is able to deal with grammatically correct but semantically anomalous sentences.|../mac/prof121.html",
    "LSA: First dimension and dimensional weighting...||Paper Track LSA: First dimension and dimensional weighting X. Hu, Department of Psychology, The University of Memphis Z. Cai, Department of Psychology, The University of Memphis D. Franceschetti, Department of Physics, The University of Memphis P. Penumatsa, Department of Psychology, The University of , University of Memphis University of Memphis Z. Cai , University of Memphis D. Franceschetti , University of Memphis P. Penumatsa , University of Memphis A. C. Graesser , University of Memphis M.M. Louwerse , University of Memphis D. S. McNamara , University of Memphis We made two discoveries on Latent Semantic Analysis (LSA). First, we observed the special properties of the ¤rst dimension of the LSA space. Second, we observed that dimensional weighting plays an important role in LSA analysis. Based on the ¤rst discovery, we examined the cosine matches without ¤rst dimension. Based on the second discovery, we explored di¤erent dimensional weighting schemes. Based on these observations, we recommend a new algorithm for LSA cosine computation such that LSA becomes more sensitive to relevant similarities and differences.|../mac/prof122.html",
    "A Race Model of Perceptual Forced Choice Reaction Time...||Paper Track A Race Model of Perceptual Forced Choice Reaction Time David E. Huber , Department of Psychology, 1147 Biology/Psychology Building dhuber@psych.colorado.edu Denis Cousineau , Département de psychologie, University Montreal Denis.Cousineau@UMontreal.CA We present a race model for forced-choice data that provides a unified account of both latency and accuracy. The model is applied in the domain of short-term priming, but could characterize many other response tasks. A series of perceptual identification experiments found systematic bias changes as a function of prime duration. Notably, reaction times (RTs) were observed to change along with response bias. Furthermore, correct RTs changed in an opposite manner to error RTs. These results are explained by assuming a race between choice alternatives. The theory provides an alternative to signal detection theory, with faster finish times, rather than greater signal strength, determining both accuracy and RT.|../mac/prof123.html",
    "Comparative Search Reveals the Tradeoff between Eye Movements and Working M...||Paper Track Comparative Search Reveals the Tradeoff between Eye Movements and Working Memory Use in Visual Tasks Shantanu Inamdar , Department of Computer Science, University of Massachusetts Boston inamdar@cs.umb.edu Marc Pomplun , Department of Computer Science, University of Massachusetts Boston marc@cs.umb.edu The experiments reported here provide an insight into how the use of working memory is influenced when eye movements become ‘costlier’ in a visual task. In our comparative search paradigm, each half of the screen contains a column of simple geometrical objects of three sizes and forms. Participants have to detect whether the two halves of the screen are exactly identical or contain a difference. The eye movement data recorded from two experiments allows us to investigate working memory adaptation to varying costs of eye movements.|../mac/prof124.html",
    "On the Use of Intelligent Agents as Partners in Training Systems for Comple...||Paper Track On the Use of Intelligent Agents as Partners in Training Systems for Complex Tasks Thomas R. Ioerger , Department of Computer Science, Texas A&M University fioerger@cs.tamu.edu Joseph Sims , Department of Computer Science, Texas A&M University jms3627@cs.tamu.edu Richard A. Volz , Department of Computer Science, Texas A&M University volz@cs.tamu.edu Judson Workman , Department of Psychology, Wright State University judwwork@yahoo.com Wayne L. Shebilske , Department of Psychology, Wright State University wayne.shebilske@wright.edu Training protocols that involve working with a human partner have been shown to be beneficial for learning complex tasks. In this paper, we explore emulating the function of the partner with an intelligent agent. Given a cognitive task analysis, the task can be decomposed into cognitive components, and these behaviors can be independently automated using agent-programming techniques. Then a trainee and the agent can work together to solve practice problems, each taking responsibility for a different function. We argue that it is desirable not only for the agent to produce correct and consistent behavior (e.g. demonstrating the optimal strategy), but also to appear realistic (human-like, including errors), and we show how this can be achieved by introducing randomness in an agent’s decisions. We implemented a Partner Agent for Space Fortress, a laboratory task designed to be representative of complex tasks, and found that trainees who swapped roles with this agent during training achieved significantly higher performance scores asymptotically than those who trained using a standard (whole-task) training protocol. We also simulated 3 different levels of expertise and found that trainees who worked with an “expert-level” agent received the most benefit.|../mac/prof125.html",
    "Mundane Comparisons Can Facilitate Relational Understanding...||Paper Track Mundane Comparisons Can Facilitate Relational Understanding Dedre Gentner , Department of Psychology, Northwestern University Dedre Gentner , Department of Psychology, Northwestern University This paper addresses the role of comparison in highlighting relational structure. Specifically, we investigated whether even literal similarity comparisons (where both objects and relations match) can lead to a focus on common relations over equally suitable object matches. Whereas previous studies have demonstrated the highlighting of relations during analogical comparisons (where only relations match), the study presented in this paper suggests that even literal similarity comparisons can promote relational focus. In addition, we explore the role of types of comparison tasks (listing commonalities or listing differences). This extension follows naturally from structure-mapping theory, which predicts that in certain cases, listing differences between two things can actually lead to a heightened focus on the structure common to both.|../mac/prof126.html",
    "Statistical Detection of Local Coherence Relations in Narrative Recall and ...||Paper Track Statistical Detection of Local Coherence Relations in Narrative Recall and Summarization Data Cynthia M. Jaynes , School of Human Development, GR4.1 University of Texas at Dallas jaynes@utdallas.edu Richard M. Golden , School of Human Development, GR4.1 University of Texas at Dallas golden@utdallas.edu A new categorical time-series analysis method (which has a connectionist model interpretation) called KDC (Knowledge Digraph Contribution) analysis was used to investigate differences in recall and summarization production data as a function of reproductive and semantic coherence relations. The results provided support for the hypothesis that reproductive memory contributions play a dominant role in characterizing differences between recall and summarization. Moreover, the methodology and results described here illustrate the usage and application of KDC analysis.|../mac/prof127.html",
    "Automatic Contexonym Organizing Model...||Paper Track Automatic Contexonym Organizing Model Hyungsuk Ji H. Ji, Institut National Polytechnique de Grenoble jihs@isc.cnrs.fr Sabine Ploux , University of Lyon I, Institut des Sciences Cognitives ploux@isc.cnrs.fr Normal language user's word association intuition (like drunken stagger) raises questions on mental lexicon organization and its implementation for natural language processing tasks. We present an automatic contextually related words (contexonym) organizing model that re ects this intuition, giving one of the possible answers to this question. The model, trained by large corpus, (1) selects contexonyms and (2) classi es these groups of related words by geometric representation. Test on the synonyms with di erent connotation like error, blunder, slip, lapse (examples discussed in Near-Synonymy and Lexical Choice [Edmonds and Hirst, 2002]) shows that the model provides valuable contexonyms that re ect di erent usage and nuance of each word. Furthermore, test on some polysemous words showed that the model could classify contexonyms according to the di erent senses of the target words. The model could serve for both theoretical lexicon related research and practical NLP research besides its own referencial functionality.|../mac/prof128.html",
    "Event Reasoning as a Function of Working Memory Capacity and Long Term Work...||Paper Track Event Reasoning as a Function of Working Memory Capacity and Long Term Working Memory Skill Mark T. Jodlowski , Department of Psychology, Mississippi State mtj27@ra.msstate.edu Stephanie M. Doane, Ph.D. , Department of Psychology, Mississippi State sdoane@ra.msstate.edu The present research examined the relationship of working memory (WM) capacity and long term working memory (LTWM) skill in complex task performance as a function of expertise. Individual differences in WM capacity and LTWM skills of sixty-two aviation pilots (Expert =25, Novice=37) were related to their performance on a task designed to measure components of flight situation awareness (SA). LTWM skill and WM capacity were not correlated, suggesting they are distinct constructs. Experts have higher LTWM skill compared to novices, and LTWM skill acts as a predictor of Expert SA task performance. Implications of these results are discussed. There is currently a debate in the literature regarding the LTWM is a theory of a memory process that explains how individuals can extend WM capacity well beyond the proposed seven plus or minus two chunks (Miller, 1956). In LTWM theory, domain-specific knowledge and meaningful experiences are thought to increase individual ability to efficiently encode information into long-term memory (LTM), and create easily accessible retrieval structures. Ericsson and Kintsch (1995) define a retrieval structure as an organization of meaningful data into a stable structure that can be used to rapidly encode information into and retrieve information from LTM. LTWM uses these retrieval structures as indices to information stored in LTM. These|../mac/prof129.html",
    "Representational, Textual, and Experimental Practices:The Case of Michael F...||Symposium Track Representational, Textual, and Experimental Practices:The Case of Michael Faraday Session Organizer: Ryan D. Tweney (tweney@bgnet.bgsu.edu) Department of Psychology, Bowling Green State University Ronald Anderson (ronald.anderson@bc.edu) Department of Philosophy, Boston College Elizabeth Cavicchi (ecavicch@mit.edu) The expansion of cognitive-historical methods and theory within cognitive science in recent years (e.g., Gooding; 1990; Gorman, 1997; Klahr & Simon, 1999; Nersessian, 1999; Tweney, 2001) has relied in part on the application of existing cognitive scientific methods to important historical cases of scientific research. Michael Faraday (1791-1867) has served as a case study in many of these efforts, partly because the extensive diary and notebook record of his work is so large, and also because substantial amounts have been published (e.g., Martin, 1931-36). In this symposium we present three different uses of the vast case material available to the student of Faraday, with a view toward understanding and extending the accounts of Faraday’s representational practices. The complexity of Faraday’s work requires that the larger social and material context of his work be approached from a variety of points of view, each of which can be grounded in cognitive scientific principles. Each of three speakers will discuss recent research using primary source materials (Faraday’s diaries and notebooks, surviving specimens from his laboratory, and experimental replications of his work), in addition to Faraday’s published writings (e.g., Faraday, 1839- 1855), with a special focus on understanding the representational dynamics of his scientific practices.|../mac/prof13.html",
    "The Additive Judge: On the Abstraction of Explicit Knowledge of Cue-Criteri...||Paper Track The Additive Judge: On the Abstraction of Explicit Knowledge of Cue-Criterion Relations Linnea Karlsson , Department of Psychology, Umeå University linnea.karlsson@psy.umu.se Peter Juslin , Department of Psychology, Umeå University peter.juslin@psy.umu.se Henrik Olsson , Department of Psychology, Uppsala University henrik.olsson@psyk.uu.se An experiment with a multiple-cue judgment task tested the hypothesis that humans can only abstract explicit representations of cue-criterion relations when the cues are related to the criterion by an additive function. It is proposed that the sequential and capacity-constrained nature of controlled, explicit thought can only induce and execute linear additive cue integration; non-additive environments require exemplar memory. The results showed that an additive task induced processes of cue abstraction and cue integration, while a multiplicative task induced exemplar processes. The results suggest flexible interplay between distinct representation-levels, a preference to abstract explicit “rules” whenever possible, although this capacity is constrained to additive cue-criterion relations.|../mac/prof130.html",
    "Distributed Tutorial Strategies...||Paper Track Distributed Tutorial Strategies Sandra Katz , Learning Research and Development Center, University of Pittsburgh Katz+@Pitt.Edu In one-on-one tutoring sessions, lessons are sometimes distributed across several interaction episodes. This paper presents a framework for describing the goal structure of tutorial dialogues that are often distributed between physics problem-solving sessions and post-solution, reflective discussions. These dialogues generalize from the current problem or attempt to build the student’s repertoire of methods for achieving particular problem-solving goals. We observed recurring patterns in the way that tutors parcel dialogue sub-goals that achieve these functions. We call these patterns “distributed tutorial strategies.”|../mac/prof131.html",
    "A Probabilisitic Parser as a Model of Global Processing Difficulty...||Paper Track A Probabilisitic Parser as a Model of Global Processing Difficulty Frank Keller , School of Informatics, University of Edinburgh keller@inf.ed.ac.uk We present a model of global processing difficulty in human parsing. This model is based on a probabilistic context-free grammar and is trained on a realistic corpus sample. It achieves broad coverage and high parsing accuracy on unseen text, and its predictions are significantly correlated with experimental data on word order preferences in German. The model makes predictions about the differential behavior of verb final and verb initial sentences and provides evidence for the importance of lexical information in sentence processing.|../mac/prof132.html",
    "A Psychophysical Law for Linguistic Judgments...||Paper Track A Psychophysical Law for Linguistic Judgments Frank Keller , School of Informatics, University of Edinburgh keller@inf.ed.ac.uk It has been argued that linguistic acceptability can be estimated using the psychophysical technique of magnitude estimation, in the same way as physical continua such as brightness and loudness (Bard, Robertson, & Sorace, 1996; Cowart, 1997). For physical continua, plotting the perceived stimulus magnitude against the actual physical magnitude results in a power relationship, the Psychophysical Law (Stevens, 1957). We show that a power law of the same kind can be derived by plotting estimated linguistic acceptability against the number of linguistic constraints violated in the stimulus. 1 5 25 brightness loudness|../mac/prof133.html",
    "Theory-Based Induction...||Paper Track Theory-Based Induction Charles Kemp , Department of Brain and Cognitive Sciences Massachusetts Institute of Technology ckemp@mit.edu Joshua B. Tenenbaum jbt@mit.edu We show how an abstract domain theory can be in- corporated into a rational statistical model of induc- tion. In particular, we describe a Bayesian model of category-based induction, and generate the prior distribution for our model using a formal theory of the distribution of biological properties across classes of biological kinds. Our theory-based model is both more principled than previous approaches and better able to account for human ratings of ar- gument strength. Philosophers since Hume have struggled with the logical problem of induction, but children solve an even more di±cult task the practical problem of induction. Children somehow manage to learn con- cepts, categories, and word meanings, and all on the basis of a set of examples that seems hopelessly in- adequate. The practical problem of induction does not disappear with adolescence: adults face it ev- ery day whenever they make any attempt to predict an uncertain outcome. Inductive inference is a fun- damental part of everyday life, and a fundamental phenomenon in need of a psychological explanation. Two important questions can be asked about in- ductive generalization: what resources does a per inspires our work is to develop a model with an in- ductive bias that is well motivated by a theory of the domain under consideration. Many previous models have taken similarity judgments as their representation of prior knowl- edge (Nosofsky, 1986; Osherson et al., 1990). This approach has been dominant within the tradition of category-based induction, and Osherson et al.'s (1990) similarity-coverage model will be one stan- dard against which our new model will be compared. Using similarity data to represent prior knowledge is a reasonable ¯rst attempt, but similarity judg- ments are less than ideal as a starting point for a model of inductive inference. As Goodman (1972) has pointed out, similarity is a vague and elusive notion. It is meaningless to say that two objects are similar unless a respect for similarity has been speci¯ed. Any model based on similarity alone is therefore a model without a secure foundation. Instead of relying on similarity, the model devel- oped in this paper is founded on a simple theory of a particular domain of reasoning: kinds of ani- mals and their properties. The theory consists of|../mac/prof134.html",
    "The Fallacy of Single-Source Explanations:The Multiple Difficulties of the ...||Paper Track The Fallacy of Single-Source Explanations:The Multiple Difficulties of the Nine-Dot Problem Trina C. Kershaw , Department of Psychology, University of Illinois tkersh1@uic.edu Stellan Ohlsson , Department of Psychology, University of Illinois stellan@uic.edu Colleen Coyne , Department of Psychology, University of Illinois Single-source explanations of problem difficulty are common throughout the insight literature (cf. Dominowski & Dallob, 1995; Isaak & Just, 1995; Weisberg & Alba, 1981). However, many insight problems contain multiple difficulties. We propose that the nine-dot problem (Maier, 1930) is difficult because of the interaction between perceptual, knowledge, and process factors. Two experiments show how different types of training can have a statistically significant effect on the overall solution rate, yet produce a small effect size. A third experiment shows how the multiple difficulties of the nine-dot problem can be simultaneously addressed.|../mac/prof135.html",
    "The effect of cultural differences in fear of isolation on dialectical reas...||Paper Track The effect of cultural differences in fear of isolation on dialectical reasoning Kyungil Kim , Department of Psychology, University of Texas kyungil@psy.utexas.edu Arthur B. Markman , Department of Psychology, University of Texas markman@psy.utexas.edu Previous research suggests that members of East Asian cultures show a greater preference for dialectical (vs. nondialectical) thinking than do members of Western Cultures. We suggest this difference is rooted in a greater chronic fear of isolation (FOI) in East Asians than in Westerners. To explain this hypothesis, we manipulated fear of isolation in a group of Western college students and assessed their preference for dialectical and nondialectical proverbs. Consistent with our proposal, High FOI students showed a greater relative preference for dialectical proverbs than did Low FOI students.|../mac/prof136.html",
    "Differences in How English and German Speakers Talk and Reason about CAUSE...||Paper Track Differences in How English and German Speakers Talk and Reason about CAUSE Bianca Klettke , Department of Psychology, Memphis bklettke@memphis.edu Phillip Wolff , Department of Psychology, Memphis wolff@memphis.edu This research examined the possibility that causal reasoning might be influenced by language. In Experiments 1 and 2, English and German speakers described 3D animations of complex events using CAUSE verbs (cause, get) and ENABLE verbs (let, enable). As predicted, English speakers used CAUSE verbs to describe a wider range of events than German speakers. In Experiment 3, English and German speakers viewed 3D animations of CAUSE and ENABLE events and then estimated the likelihood of the effect (E) in the presence of the affector (A), p(E A), in hypothetical situations similar to the one they just saw. Given the results of Experiments 1 and 2, we predicted that German speakers’ estimates of p(E A) would be higher than English speakers’ estimates of p(E A) for ENABLE events, but not necessarily for CAUSE events. The results were as predicted. The findings suggest that English and German speakers differ in the range of situations they describe as causal and that these differences in linguistic coding can lead to differences in causal reasoning.|../mac/prof137.html",
    "Actions and Roles: using depicted events for disambiguation and reinterpret...||Paper Track Actions and Roles: using depicted events for disambiguation and reinterpretation in German and English Pia Knoeferle , Saarland University, Department of Computational Linguistics knoeferle@coli.uni-sb.de Matthew W. Crocker , Saarland University, Department of Computational Linguistics crocker@coli.uni-sb.de Christoph Scheepers , University of Dundee chsc@coli.uni-sb.de Martin J. Pickering , The University of Edinburgh Martin.Pickering@ed.ac.uk Can depicted actions help in rapidly establishing role relations between participants in an event and trigger a rapid disambiguation in the comprehension of temporarily ambiguous spoken sentences? Verb information has been shown to influence visual referential processing (e.g. Altmann & Kamide, 1999; Kako & Trueswell, 2000). Moreover, results from priming studies (e.g. Ferretti, McRae & Hatherell, 2001) suggest that verbs activate typical agents, patients and instruments via event knowledge. Thus, verbs can make complex knowledge structures accessible. However, verb knowledge may sometimes not be sufficient to interpret role relations correctly. Do action verbs and visual information about a character’s role-relation to other characters in an event interact in sentence comprehension? Three experiments present evidence that visual event information – instruments and role properties – can lead to a rapid disambiguation of an initial ambiguity which verb knowledge alone could not have disambiguated. In addition we found effects of reinterpretation processes. Findings have been replicated cross-linguistically for two different sentence structures (German SVO/OVS and English MV/RR clause).|../mac/prof138.html",
    "Toward a mutual adaptive interface: An interface induces a user’s adaptatio...||Paper Track Toward a mutual adaptive interface: An interface induces a user’s adaptation and utilizes this induced adaptation, and vice versa Takanori Komatsu (komatsu@fun.ac.jp) Department of Media Architecture, Future University-Hakodate Atsushi Utsunomiya (au@cs.c.u-tokyo.ac.jp) Kentaro Suzuki (suzuki@cs.c.u-tokyo.ac.jp) Kazuhiro Ueda (ueda@gregorio.c.u-tokyo.ac.jp) K , Future University-Hakodate komatsu@fun.ac.jp Atsushi Utsunomiya , The University of Tokyo au@cs.c.u-tokyo.ac.jp Kentaro Suzuki , The University of Tokyo suzuki@cs.c.u-tokyo.ac.jp Kazuhiro Ueda , The University of Tokyo ueda@gregorio.c.u-tokyo.ac.jp Kazuo Hiraki , The University of Tokyo khiraki@idea.c.u-tokyo.ac.jp Natsuki Oka , Matsushita Electric Industrial Co., Ltd. oka@mrit.mei.co.jp The purpose of this paper is to construct a meaning acquisition model as a basic technology for a mutual adaptive speech interface which can communicate smoothly with an everyday user. We then constructed the meaning acquisition model in consideration of the following assumptions: (1). the model needs to induce a user’s adaptation and to utilize this induced adaptation for the meaning acquisition processes, (2). the model focuses on the prosodic information, rather than phoneme information on which most past interface studies focused. As a result, we could confirm that this model could recognize the intentions/ meanings of users’ verbal commands by inducing users’ adaptations and utilizing these for the meaning acquisition process when appropriate instructions were given to them from an experimenter. This result would complement the interface studies that focused only on phoneme information, and contribute to a customization or personalization technology for a speech interface system.|../mac/prof139.html",
    "Structure-Mapping: Twenty Years After...||Symposium Track Structure-Mapping: Twenty Years After Phillip Wolff, Ken Forbus, Arthur Markman, Kenneth Kurtz, Dedre Gentner Discussant: Mike Tomasello It is twenty years since the publication of Gentner’s (1983) paper on structure-mapping theory. This theory was one influence in creating the subfield of analogical learning and reasoning within Cognitive Science. Over the ensuing two decades, analogy has become a central focus in Cognitive Science, both in psychological research and in computational modeling. Great strides have been made in delineating the basic processes of analogy and similarity and their roles in learning and reasoning. In this panel we discuss research delineating the basic processes of analogy and similarity: the process of comparison and similarity-based memory retrieval. Then we present research showing that the principles of structural alignment and mapping theory apply broadly throughout cognition — to categorization, decision-making, and learning and transfer. This symposium presents a range of work both testing basic predictions of Structure-Mapping Theory and applying it to cognition in the large.|../mac/prof14.html",
    "The Role of Causal Models in Reasoning Under Uncertainty...||Paper Track The Role of Causal Models in Reasoning Under Uncertainty Tevya R. Krynski , Department of Brain & Cognitive Sciences, Massachusetts Institute of Technology tevya@mit.edu Joshua B. Tenenbaum , Department of Brain & Cognitive Sciences, Massachusetts Institute of Technology jbt@mit.edu When making judgments based on statistical data, people have been shown to be poor at probabilistic reasoning, specifically Bayesian inference. While recent research shows people perform better when information is provided in a natural frequency format, we find this result’s explanatory reach limited, both for explaining people’s judgment failures and as a theory of human reasoning under uncertainty. Most prior studies demonstrating probabilistic reasoning deficits gave their subjects probabilistic inference problems that did not explain the causal mechanisms behind the provided statistics. Our research shows that when questions are posed that explain the causal structure of the domain, subjects perform significantly better. Specifically, base rate neglect can be made to virtually disappear when the content of a question reflects the true causal structure of the domain. We propose that causality is essential to probabilistic reasoning, and that without the opportunity to incorporate statistical data into a consistent theory of the causal structure of a domain, the typical person will have trouble performing normatively correct Bayesian inference.|../mac/prof140.html",
    "Inferring Hidden Causes...||Paper Track Inferring Hidden Causes Tamar Kushnir , Department of Psychology, Tolman Hall, University of California tkushnir@socrates.berkeley.edu Alison Gopnik , Department of Psychology, Tolman Hall, University of California gopnik@socrates.berkeley.edu Laura Schulz , Department of Psychology, Tolman Hall, University of California laurasch@socrates.berkeley.edu David Danks , Carnegie Mellon University ddanks@cmu.edu One of the important aspects of human causal reasoning is that from the time we are young children we reason about unobserved causes. How can we learn about unobserved causes from information about observed events? Causal Bayes nets provide a formal account of how causal structure is learned from a combination of associations and interventions. This formalism makes specific predictions about the conditions under which learners postulate hidden causes. In this study adult learners were shown a pattern of associations and interventions on a novel causal system. We found that they were able to infer hidden causes as predicted by the Bayes net formalism, and were able to distinguish between one hidden common cause and two hidden independent causes of the observed events.|../mac/prof141.html",
    "Choice and Learning under Uncertainty: A Case Study in Baseball Batting...||Paper Track Choice and Learning under Uncertainty: A Case Study in Baseball Batting Christian Lebiere , Human Computer Interaction Institute Carnegie Mellon University cl@cmu.edu Robert Gray , Applied Psychology Department Arizona State University robgray@asu.edu Dario Salvucci , Department of Computer Science Drexel University salvucci@cs.drexel.edu Robert West , Departments of Cognitive Science and Psychology Carleton University robert_west@carleton.ca This paper describes the modeling of human performance in a real-world, embodied, stochastic task: baseball batting. Experimental results were gathered in a virtual reality setup and a Markov model of performance, especially errors, was developed. The focus of this paper is on a model of the task developed in the ACT-R cognitive architecture, most specifically the critical subtask of generating an expectation for the next pitch. The model has no free parameter and provides an account for the results based on the architectural constraints of declarative memory. The Markov and ACT-R models are briefly compared. The broader relevance of the task is discussed and possible applications are suggested.|../mac/prof142.html",
    "A Connectionist Account of Analogical Development...||Paper Track A Connectionist Account of Analogical Development Robert Leech , School of Psychology, Birkbeck, University of London r.leech@psychology.bbk.ac.uk Denis Mareschal , School of Psychology, Birkbeck, University of London d.mareschal@bbk.ac.uk Richard Cooper , School of Psychology, Birkbeck, University of London r.cooper@bbk.ac.uk We present a connectionist model that provides a mechanistic account of the development of simple relational analogy completion. Drawing analogies arises as a bi-product of pattern completion. in a network that learns input/output pairings representing relational information. Analogy is achieved by an initial example of a relation priming the network such that the subsequent presentation of an input produces the correct analogical response. The results show that the model successfully solves simple A:B::C:D analogies and that its developmental trajectory closely parallels that of children. Finally, the model makes two strong empirical predictions.|../mac/prof143.html",
    "The Effect of Semantic Relatedness and Typicality upon Visual Detection of ...||Paper Track The Effect of Semantic Relatedness and Typicality upon Visual Detection of a Target Laure Léger , Laboratoire Cognition & Usages, Université Paris laure.leger@free.fr Denis Chêne , France Télécom Recherche et Developpement, DIH/UCE/RCE denis.chene@francetelecom.com Thierry Baccino , Laboratoire de Psychologie Expérimentale et Quantitative, Université Nice Sophia-Antipolis Thierry.Baccino@unice.fr Charles Tijus , Laboratoire Cognition & Usages, Université Paris tijus@idf.ext.jussieu.fr In line with previous work by Treisman, many studies have demonstrated effects of perceptual discrimination between visual targets and distractors. Few studies however have investigated semantic discrimination effects. We report the findings of two experiments that were carried out in order to demonstrate the influence of semantic relatedness between a target word and distracter words in interaction with perceptual discrimination. The results demonstrate perceptual discrimination effects within a semantic task (eg. from a group of words finding a word which means a type of fish), the role of degree of target typicality and, importantly, the effect of semantic relatedness between a target and its distracter context.|../mac/prof144.html",
    "Reconsidering the notion of Dynamic Systems Theory Resources as a Conceptua...||Paper Track Reconsidering the notion of Dynamic Systems Theory Resources as a Conceptual Framework Daniel Hsi-wen Liu , Division of Humanities, Providence University hwliu@pu.edu.tw Taking DST (dynamic systems theory) resources as a conceptual framework for thinking about dynamic systems transforms numerical complexities into conceptual deliberations, and consequently greatly facilitates embodied cognitive science (Clark 1997, 1999, 2001). In fact, some dynamicists, such as Thelen & Smith (1994), Thelen (1995) and van Gelder (1998), worked at this conceptual level. However, the management of those conceptual resources may risk floating away from the anchor of the DST-based equation-governed modeling. This may, at least, incur three mis-implications, as this paper aims to uncover. Firstly, conceiving of cognitive systems on the grounds of DST resources, surprisingly, does not warrant a DSTbased underlying mechanism. Secondly, inference across the DST resources may be a mistake. Lastly, the entitlement of a ‘dynamic system’ on grounds of DST resources may direct our attention to various troubling ambiguities of that term. Therefore, before all those risks are avoided, DST resources would not be safely useful as tools.|../mac/prof145.html",
    "Transposition and Generalization on an Artificial Dimension...||Paper Track Transposition and Generalization on an Artificial Dimension J.E.M. Locke , Department of Experimental Psychology M.B. Suret , Department of Experimental Psychology I.P.L. McLaren , Department of Experimental Psychology iplm2@cus.cam.ac.uk Abstract In two experiments we demonstrate that a model based on generalization gives a good fit to data obtained when subjects have t o transfer learning based on one discrimination to another on the same dimension. The standard associative account of this phenomenon appeals to the notion of generalization, making it a promising preparation for our purposes. Spence’s (1936) theory of discrimination learning in animals assumes that if a response to a particular stimulus is followed by reward,|../mac/prof146.html",
    "Evaluating the Causal Role of Unobserved Variables...||Paper Track Evaluating the Causal Role of Unobserved Variables Christian C. Luhmann , Department of Psychology, Vanderbilt University christian.luhmann@vanderbilt.edu Woo-kyoung Ahn , Department of Psychology, Vanderbilt University woo-kyoung.ahn@yale.edu Current psychological models of causal induction assume that causal relationships are inferred based on observations about whether the cause and effect are present or absent. The current study investigated how people infer the causal roles of unobserved events. In Experiment 1 we demonstrate that participants are indeed willing to evaluate the causal roles of unobserved events. We then suggest that the basis for these judgments may be situations in which effect occur in the absence of observed causes. Experiment 2 provides evidence that such information does influence participants’ judgments about unobserved causes.|../mac/prof147.html",
    "Did, Made, Had, Said: Capturing Quasi-Regularity in Exceptions...||Paper Track Did, Made, Had, Said: Capturing Quasi-Regularity in Exceptions Gary Lupyan , Center for Neural Basis of Cognition glupyan@cnbc.cmu.edu James L. McClelland , Center for Neural Basis of Cognition jlm@cnbc.cmu.edu The English past tense is a quasi-regular system, in that many of the irregular verbs share characteris- tics with regular items. Among high-frequency ex- ceptions, in particular, several have the regular /d/ or /t/ ending but with either a reduction of the vowel (did, said) or a deletion of a stem consonant (had, made). Such forms suggest that many so-called irreg- ular verbs re°ect a joint in°uence of the systematic past-tense pattern captured in fully regular items to- gether with a pressure to be short or simple. We adapt familiar neural network formalisms to show how such forms can arise if the phonological content of word forms are constrained (a) to support accurate communication of the word's meaning and (b) to be simple. languages are functional (Bybee, Perkins, & Pagliuca, 1994), but the prevalence of irregularity and quasi- regularity begs for an explanation. The present work combines the lexical phonology approach of Burzio (2002) with the framework of minimum-description length (Rissanen & Ristad, 1994; Zemel % Hinton, 1994) to study how phonology changes over time to maintain regularity and produce morphological excep- tions in response to various pressures on communica- tion. Although many of our examples are from the English past-tense, our aim is more general: to eluci- date the emergence of irregularity and quasi-regularity|../mac/prof148.html",
    "The Role of Knowledge Support in Creating Noun-Noun Compounds...||Paper Track The Role of Knowledge Support in Creating Noun-Noun Compounds Dermot Lynott , Department of Computer Science, University College Dublin dermot.lynott@ucd.ie Mark Keane , Department of Computer Science, University College Dublin mark.keane@ucd.ie The production of novel nominal compounds has long been neglected in the empirical and computational literature. In this paper, we advance a new paradigm for examining the creation of novel noun-noun compounds. Two experiments are reported that show that compound creation is affected by available world knowledge and syntax. To conclude, we discuss some of the issues arising out of this new work and outline a computational model to account for compound creation.|../mac/prof149.html",
    "Beyond Switch Cost as a Measure of Cognitive Control...||Publication-Based Talk Track Beyond Switch Cost as a Measure of Cognitive Control Erik M. Altmann , Department of Psychology, Michigan State University ema@msu.edu An area of cognitive research that has drawn increasing attention over the past 10 years is task switching, the question being how the cognitive system sets itself to perform one task as opposed to another when more than one task is possible. In a typical task-switching paradigm, there might be two simple tasks, for example classifying a digit either as even or odd or as higher or lower than some criterion (usually 5). A trial stimulus in this scenario would be a single digit, and a response one of two key-presses, with the same two keys typically used for both tasks, to increase the burden on the cognitive system in terms of keeping straight which task to perform on the current trial. There might be hundreds or thousands of trials presented in series, with the environment periodically cueing the participant as to which task to perform on the upcoming trial or trials. The conventional dependent measure is switch cost — an increase in response time (and often error) on a trial on which the task switched relative to the previous trial. Research on task switching is important because it addresses basic questions about goal-directed behavior, and because it lies at the intersection of various domains, including selective attention, executive function, and shortterm or working memory. And yet, empirical taskswitching research has had tunnel vision in its focus on switch cost (Altmann, in press). Theorizing is still largely in terms of archaic metaphors in which a homunculus scurries about, “reconnecting and reconfiguring the various modules in our brains” (Monsell & Driver, 2000) when the environment cues a new task. The preoccupation with switch cost and the switching homunculus has offered little incentive to raise integrative questions about how task switching might be affected by constraints that we should be familiar with from other domains. This presentation will identify several new task-switching phenomena, and will relate them to one another and to existing theoretical constraints from the memory domain. Among the new effects that will be discussed are within-run slowing and within-run error increase (Altmann, 2002; Altmann & Gray, 2002); an unconventional, full-run switch cost evident in errors but not latencies; and a task-cue encoding cost incurred whenever a task cue appears, regardless of whether the cue switches the task. This diverse collection may seem difficult to unify when interpreted in terms of a switching homunculus, but has a straightforward interpretation in terms of standard memory constructs like activation, and processes like priming, decay, and proactive interference. A theoretical framework will be presented that integrates these phenomena functionally, and that emphasizes a systems view of cognitive control over the narrow perspective that currently dominates the taskswitching literature.|../mac/prof15.html",
    "The Benefits of Epistemic Action Outweigh the Costs...||Paper Track The Benefits of Epistemic Action Outweigh the Costs Paul P. Maglio , IBM Almaden Research Center pmaglio@almaden.ibm.com Michael J. Wenger , Department of Psychology University of Notre Dame mwenger1@nd.edu Angelina M. Copeland , Department of Psychology University of Notre Dame ahill1@nd.edu Epistemic actions are physical actions people take to simplify internal problem solving rather than to move closer to an external goal. When playing the video game Tetris, for instance, experts routinely rotate falling shapes more than is strictly needed to place the shapes. Maglio and Kirsh (Kirsh & Maglio, 1994; Maglio, 1995) proposed that such ac- tions might serve the purpose of priming memory by external means, reducing the need for internal computation (e.g., by way of mental rotation). This proposal requires that information provided by epis- temic actions (e.g., additional views of the shape) serve the same function as memory primes, and that the bene t of such priming exceed the costs of per- forming the epistemic action. To calculate bene- t, we used a novel statistical method for mapping reaction-time data onto an estimate of the increase in individual processing capacity a orded by seeing shapes in multiple orientations. To calculate cost, we used an empirical estimate of time needed to take action in a Tetris game. We found that the bene ts of extra previews far outweigh the costs of taking extra action.|../mac/prof150.html",
    "Distributed Cognition and Joint Activity in Collaborative Problem Solving...||Paper Track Distributed Cognition and Joint Activity in Collaborative Problem Solving Paul P. Maglio , IBM Almaden Research Center pmaglio@almaden.ibm.com Eser Kandogan , IBM Almaden Research Center eser@almaden.ibm.com Eben Haber , IBM Almaden Research Center ehaber@almaden.ibm.com Troubleshooting large software systems is often highly collaborative. Because such systems consist of complex infrastructures with many interdependent components, expertise is spread across different people and organizations. Those who administer such systems are faced with cognitive and social challenges, including the maintenance of trust and management of attention, as they troubleshoot in collaboration with peers, technical support, and software application developers. We take a distributed cognition approach to interpreting a specific instance of problemdiagnosis in administering a web-based system, examining the flow of representational state across media in a single system administrator's environment. We also apply the idea of language use as joint activity to understand how discourse attributes affect what is accomplished collaboratively. Our analysis focuses on how trust shifts among participants and other information sources, and how this affects what information is attended to, transmitted, and used.|../mac/prof151.html",
    "Perceiving the Infinite and the Infinitesimal World: Unveiling and Optical ...||Paper Track Perceiving the Infinite and the Infinitesimal World: Unveiling and Optical Diagrams in the Construction of Mathematical Concepts Lorenzo Magnani , University of Pavia lmagnani@cc.gatech.edu Riccardo Dossena , University of Pavia rdossena@pure-reason.unipv.it any important concepts of the calculus are dif- cult to grasp, and they may appear epistemologi- cally unjusti ed. For example, how does a real func- tion appear in \\small neighborhoods of its points? How does it appear at in nity? Diagrams allow us to overcome the di culty in constructing repre- sentations of mathematical critical situations and objects. For example they actually reveal the be- havior of a real function not \\close to a point (as in the standard limit theory) but \\in the point. We are interested in our research in the diagrams which play an optical role microscopes and \\microscopes within microscopes, telescopes, windows, a mirror role (to externalize rough mental models), and an unveiling role (to help create new and interesting mathematical concepts, theories, and structures). In this paper we describe some examples of optical diagrams as a particular kind of epistemic media- tor able to perform the explanatory abductive task of providing a better understanding of the calculus, through a non-standard model of analysis. We also maintain they can be used in many other di erent epistemological and cognitive situations.|../mac/prof152.html",
    "Culture and the Subversion of Cognition...||Paper Track Culture and the Subversion of Cognition Kareen Ror Malone , Department of Psychology, State University of West Georgia kmalone@westga.edu This paper seeks to conceptualize some of the issues at stake when cognitive science seriously takes up the question of culture. A number of approaches have begun to consider cultural context as more than yet another variable, the complexity of which will simply require more sophisticated computational models. Rather these approaches have undertaken to research the very form of cognition differently, as situated, distributed, and as a social practice. In these new endeavors, what is the implicit model of culture that is being used? This paper raises this essential theoretical question and takes Michael Tomasello’s work on cognition, language, and culture as an example of how recent approaches in cognitive science can raise foundational questions on the relationship of culture to cognition.|../mac/prof153.html",
    "Interpretation of Ambiguous Information in Causal Induction...||Paper Track Interpretation of Ambiguous Information in Causal Induction Jessecae K. Marsh , Department of Psychology, Vanderbilt University Jessecae.Marsh@Yale.Edu Woo-kyoung Ahn , Department of Psychology, Vanderbilt University Woo-kyoung.Ahn@Yale.Edu The current study investigates how people incorporate ambiguous information into judgments of causal relations. We presented participants with information that was not easily classified into the presence or absence of a candidate cause, breaking a traditional requirement of models of causal induction. We found that people were willing to incorporate this ambiguous information into their collected evidence, instead of ignoring the information as uninformative. Furthermore, people interpreted ambiguous stimuli as the type of evidence most consistent with their prevailing causal hypothesis. These results give an idea of how people begin to determine what can function as a candidate cause in a causal induction problem.|../mac/prof154.html",
    "Segmenting Ambiguous Events...||Paper Track Segmenting Ambiguous Events Bridgette A. Martin , Department of Psychology, Bldg. 420 Jordan Hall martin@psych.stanford.edu Barbara Tversky , Department of Psychology, Bldg. 420 Jordan Hall bt@psych.stanford.edu Everyday events, such as making a bed, are segmented hierarchically, with the coarse level punctuated by objects or object parts and the fine level by articulated actions on objects. Here we extend event cognition to events involving abstract, ambiguous motion paths of several geometric figures, viewed once or five times. Segmentation was hierarchical for both; however after one viewing, events were interpreted as movements whereas after five viewings, they were interpreted as intentional actions. Fewer (but the same) segments were identified after five viewings. Experience did not affect segment boundaries but did affect segment interpretation, shifting from bottom-up to top-down.|../mac/prof155.html",
    "Encoding of Elements and Relations of Object Arrangements by Young Children...||Paper Track Encoding of Elements and Relations of Object Arrangements by Young Children Leslee J. Martin , Department of Psychology & Center for Cognitive Science, Ohio State University martin.1103@osu.edu Vladimir M. Sloutsky , Center for Cognitive Science, Ohio State University sloutsky.1@osu.edu Two experiments investigate the ability of four-year-old children to spontaneously process relations as well as elements in an immediate recognition task. The experiments also test predictions of a model proposed to account for differential processing of elements and relations. Both experiments used a two-item forced-choice task. In each experiment, children accurately recognized the target, regardless of whether distracter items included different elements or different relations. The results of these experiments suggest that young children do spontaneously process relations as well as elements. These findings indicate that previous findings of privileged processing for elements occurs during recoding of items into a long term memory code, or during a retrieval stage, but not during initial encoding.|../mac/prof156.html",
    "The experiential basis of meaning...||Paper Track The experiential basis of meaning Teenie Matlock , Department of Psychology, Stanford University tmatlock@psych.stanford.edu Michael Ramscar , Department of Psychology, Stanford University michael@psych.stanford.edu Lera Boroditsky , Department of Brain and Cognitive Sciences, MIT lera@mit.edu How are abstract ideas acquired and structured? One idea is that people’s understanding of abstract domains is constructed using more basic, experiential knowledge that is acquired directly. For instance, a series of studies (Boroditsky 2000, Boroditsky & Ramscar, 2002) has shown that people’s understanding of time supervenes on their physical conceptions of space, to the extent that manipulations of people’s spatial knowledge have predictable affects on their temporal reasoning. In this paper we explore just how widespread this phenomenon is. To examine whether basing abstract knowledge on concrete knowledge is a pervasive aspect of cognition, we investigate whether thought about an abstract, non-literal type of motion called fictive motion (Matlock, 2003a; Talmy, 1996) can influence the way people reason about time. Our results suggest that, contrary to previous claims (Jackendoff, 2002), abstract, metaphorical knowledge about motion involves the same structures used in understanding literal motion, and that the activation of these “literal” aspects of fictive motion serve to influence temporal reasoning. These results provide further evidence of the intimate connection between abstract and concrete knowledge.|../mac/prof157.html",
    "Incremental Nonmonotonic Parsing through Semantic Self-Organization...||Paper Track Incremental Nonmonotonic Parsing through Semantic Self-Organization Marshall R. Mayberry, III , Department of Computer Sciences The University of Texas martym@cs.utexas.edu Risto Miikkulainen , Department of Computer Sciences The University of Texas risto@cs.utexas.edu Subsymbolic systems have been successfully used to model several aspects of human language processing. Subsymbolic parsers are appealing because they allow combining syntactic, semantic, and thematic constraints in sentence interpretation and revising that interpretation as each word is read in. These parsers are also cognitively plausible: processing is robust and multiple interpretations are simultaneously activated when the input is ambiguous. Yet, it has been very difficult to scale them up to realistic language. They have limited memory capacity, training takes a long time, and it is difficult to represent linguistic structure. In this study, we propose to scale up the subsymbolic approach by utilizing semantic self-organization. The resulting architecture, INSOMNET, was trained on semantic representations of the newly-released LINGO Redwoods HPSG Treebank of annotated sentences from the VerbMobil project. The results show that INSOMNET is able to accurately represent the semantic dependencies while demonstrating expectations and defaults, coactivation of multiple interpretations, and robust parsing of noisy input.|../mac/prof158.html",
    "The Importance of Long-term Memory in Infant Perceptual Categorization...||Paper Track The Importance of Long-term Memory in Infant Perceptual Categorization Martial Mermillod , Quantitative Psychology and Cognitive Science mmermillod@ulg.ac.be Robert M. French , Quantitative Psychology and Cognitive Science rfrench@ulg.ac.be Paul C. Quinn , Psychology, University of Delaware pquinn@psych.udel.edu Denis Mareschal , Psychology, Birkbeck College d.mareschal@bbk.ac.uk Quinn & Eimas (1998) reported that young infants include non-human animals (i.e., cats, horses, and fish) in their category representation for humans. To account for this surprising result, it was proposed that the representation of humans by infants functions as an attractor for non-human animals and is based on infants’ previous experience with humans. We report three simulations that provide a computational basis for this proposal. These simulations show that that a “dualnetwork” connectionist model that incorporates both bottom-up (i.e., short-term memory) and top-down (i.e., long-term memory) processing is sufficient to account for the empirical results obtained with the infants.|../mac/prof159.html",
    "Simulating Activities: Relating Motives, Deliberation, and Attentive Coordi...||Publication-Based Talk Track Simulating Activities: Relating Motives, Deliberation, and Attentive Coordination William J. Clancey , NASA-Ames Research Center, Computational Sciences Division william.j.clancey@nasa.gov Activities are located behaviors, taking time, conceived as socially meaningful, and usually involving interaction with tools and the environment. In modeling human cognition as a form of problem solving (goal-directed search and operator sequencing), cognitive science researchers have not adequately studied “off-task” activities (e.g., waiting), non-intellectual motives (e.g., hunger), sustaining a goal state (e.g., playful interaction), and coupled perceptual-motor dynamics (e.g., following someone). These aspects of human behavior have been considered in bits and pieces in past research, identified as scripts, human factors, behavior settings, ensemble, flow experience, and situated action. More broadly, activity theory provides a comprehensive framework relating motives, goals, and operations. This paper ties these ideas together, using examples from work life in a Canadian High Arctic research station. The emphasis is on simulating human behavior as it naturally occurs, such that “working” is understood as an aspect of living. The result is a synthesis of previously unrelated analytic perspectives and a broader appreciation of the nature of human cognition. Simulating activities in this comprehensive way is useful for understanding work practice, promoting learning, and designing better tools, including human-robot systems.|../mac/prof16.html",
    "Inequality between the classes: Phonological and distributional typicality ...||Paper Track Inequality between the classes: Phonological and distributional typicality as predictors of lexical processing Padraic Monaghan , Department of Psychology, University of Warwick Padraic.Monaghan@warwick.ac.uk Nick Chater , Department of Psychology, University of Warwick N.Chater@warwick.ac.uk Morten H. Christiansen , Department of Psychology, Cornell University mhc27@cornell.edu Information about the syntactic category of a word can be derived from a number of complementary sources. We focus here on phonological and distributional cues for distinguishing nouns and verbs that have been proposed as useful for language acquisition. We assessed the extent to which the phonological and distributional typicality of nouns and verbs contribute towards accounting for variance in response times for naming, lexical decision, and a noun/verb decision task. We found that words which were typical of their syntactic category in terms of their phonological properties were responded to more quickly in both the naming and the lexical decision task. Distributional typicality related only to the noun/verb decision task.|../mac/prof160.html",
    "Two wrongs make a right: Learnability and word order consistency...||Paper Track Two wrongs make a right: Learnability and word order consistency Padraic Monaghan , Department of Psychology, University of Warwick Padraic.Monaghan@warwick.ac.uk Markus Gonitzke , Department of Psychology, University of Warwick Markus.Gonitzke@warwick.ac.uk Nick Chater , Department of Psychology, University of Warwick N.Chater@warwick.ac.uk Languages often demonstrate word order inconsistencies, and such inconsistencies ought to make languages harder to acquire. We present an integrative approach exploring the relationship between learnability and word order, incorporating syntactic theory, corpus analyses and computational modelling. We focus on comparisons between English and German, and conclude that inconsistencies may be preserved in the language due to the interaction between several syntactic structures.|../mac/prof161.html",
    "The Influence of Affect on Risky Behavior: From the Lab to Real World Finan...||Paper Track The Influence of Affect on Risky Behavior: From the Lab to Real World Financial Behavior Simon C Moore , Decision Technology Research Group, Department of Psychology, Warwick University simon.moore@warwick.ac.uk Nick Chater , Decision Technology Research Group, Department of Psychology, Warwick University n.chater@warwick.ac.uk There is some evidence indicating a relationship between variations in affect and risk aversion: under certain conditions the behavior observed suggests less risk aversion the more positive the affective state. The research presented in this paper examined how variations in everyday affective states influenced risk taking behavior in the laboratory using simple gambling tasks and then sought to corroborate findings in the laboratory using data on real world financial decision making. We observed a significant relationship between affect and risky behavior in the laboratory that we replicated using structural equation modeling. It is argued that cognitive theories of affect and decision making might have real economic consequences.|../mac/prof162.html",
    "Mechanisms of long-term repetition priming and skill refinement:A probabili...||Paper Track Mechanisms of long-term repetition priming and skill refinement:A probabilistic pathway model Michael C. Mozer , Department of Computer Science, Institute of Cognitive Science mozer@colorado.edu Michael D. Colagrosso , Department of Computer Science, Institute of Cognitive Science colagrom@colorado.edu David E. Huber , Department of Psychology, Institute of Cognitive Science dhuber@colorado.edu We address an omnipresent and pervasive form of human learning—skill refinement, the improvement in performance of a cognitive or motor skill with practice. A simple and well studied example of skill refinement is the psychological phenomenon of long-term repetition priming: Participants asked to identify briefly presented words are more accurate if they recently viewed the word. We simulate various phenomena of repetition priming using a probabilistic model that characterizes the time course of information transmission through processing pathways. The model suggests two distinct mechanisms of adaptation with experience, one that updates prior probabilities of pathway outputs, and one that increases the instantaneous probability of information transmission through a pathway. These two mechanisms loosely correspond to bias and sensitivity effects that have been observed in experimental studies of priming. The mechanisms are extremely sensible from a rational perspective, and can also explain phenomena of skill learning, such as the power law of practice. Although other models have been proposed of these phenomena, we argue for the probabilistic pathway model on grounds of parsimony and the elegant computational perspective it offers.|../mac/prof163.html",
    "Policy Shift Through Numerically-Driven Inferencing:An EPIC Experiment Abou...||Paper Track Policy Shift Through Numerically-Driven Inferencing:An EPIC Experiment About When Base Rates Matter Edward Munnich , University of California, Graduate School of Education munnich@uclink.berkeley.edu Michael Ranney , University of California, Graduate School of Education ranney@cogsci.berkeley.edu Janek Nelson , University of California, Department of Psychology jamin@socrates.berkeley.edu Jenny Garcia de Osuna , University of California, Graduate School of Education jmgdo@socrates.berkeley.edu Noli Brazil , University of California, Graduate School of Education n_brazil@uclink.berkeley.edu Building on literature on estimation, judgment and decision making, and conceptual change, we present findings from a paradigm known as Numerically-Driven Inferencing (NDI), in which we present germane, base rate information relevant to policy issues, and find that this can affect people’s belief systems. College students estimated base rate quantities relevant to policy issues for which people tend to hold strong opinions (e.g., the average SAT-I percentile of students admitted to their university), and state their preferences for these numbers. Next, we give them the actual numbers (from reliable sources), and ask them again what their preferences are. In this way, we can quantify the relationship between participants’ understandings of base rates and their policy stances. Echoing past results in the literature, we find people not particularly good at estimating base rates. Going beyond past literature, however, we also find that people readily revise their numerical preferences in response to receiving base rate feedback. This shift is beyond would be expected if people were merely rescaling their answers in proportion to the feedback, suggesting a conceptual change that gives rise to belief revision. As a secondary finding, we discuss an apparent framing effect: When numbers relevant to a particular issue are framed in different ways, they affect policies differently. Finally, we discuss a new secondary school curriculum that engages younger people in NDI, which we believe will benefit them in life, and will allow us to consider how numerical understanding and preference develop longitudinally.|../mac/prof164.html",
    "Applying Text Comprehension and Active Reading Principles to Adaptive Hyper...||Paper Track Applying Text Comprehension and Active Reading Principles to Adaptive Hyperbooks Tom Murray , Hampshire College School of Cognitive Science tmurray@hampshire.edu We show how cognitive and educational research and theory from the areas of text comprehension and active reading strategies can be applied to the domain of hypermedia textbooks (hyperbooks). We describe our work on the MetaLinks system, which is an authoring tool and web-based server for adaptive hyperbooks. In this paper we describe how findings from text comprehension research are incorporated into MetaLinks design features, and summarize formative studies of students using MetaLinks hyperbooks.|../mac/prof165.html",
    "Flexible attention and modality preference in young children...||Paper Track Flexible attention and modality preference in young children Amanda C. Napolitano , Center for Cognitive Science & Department of Psychology, Ohio State University napolitano.7@osu.edu Vladimir M. Sloutsky , Center for Cognitive Science, Ohio State University sloutsky.1@osu.edu It has been previously established (Sloutsky & Napolitano, in press) that when presented simultaneously with visual and auditory stimuli equated for discriminability and familiarity, 4-year-olds exhibited strong preference for auditory stimuli, failing to process visual stimuli. At the same time, they had no difficulty processing visual stimuli when these were presented without auditory stimuli. The current study examines the possibility that a flexible attentional mechanism underlies modality preference in young children. We specifically examine under which conditions young children are more likely to process auditory stimuli, and under which conditions they are more likely to process visual stimuli, when both stimuli are presented simultaneously. Results indicate that when visual stimuli are well familiar, 4-year-olds are likely to attend to visual stimuli, whereas when neither visual nor auditory stimuli are familiar, they are likely to attend to auditory stimuli.|../mac/prof166.html",
    "Global Model Analysis by Landscaping...||Paper Track Global Model Analysis by Landscaping Daniel J. Navarro , Department of Psychology Ohio State University navarro.20@osu.edu In Jae Myung , Department of Psychology Ohio State University myung.1@osu.edu Mark A. Pitt , Department of Psychology Ohio State University pitt.2@osu.edu Woojae Kim , Department of Psychology Ohio State University kim.1124@osu.edu How well do you know your favorite computational model of cognition? Most likely your knowledge of its behavior has accrued from tests of its ability to mimic human data, what we call local analy- ses because performance is assessed in a speciÞc testing situation. Global model analysis by land- scaping is an approach that ?sketches? out the per- formance of a model at all of its parameter val- ues, creating a landscape of how the relative per- formance abilities of the model and a competing model. We demonstrate the usefulness of landscap- ing by comparing two models of information inte- gration (Fuzzy Logic Model of Perception and the Linear Integration Model). The results show that model distinguishability is akin to power, and may be improved by increasing the sample size, using better statistics, or redesigning the experiment. We show how landscaping can be used to measure this improvement.|../mac/prof167.html",
    "Research Laboratories as Evolving Distributed Cognitive Systems...||Paper Track Research Laboratories as Evolving Distributed Cognitive Systems Nancy J. Nersessian , College of Computing, Program in Cognitive Science, Georgia Institute of Technology nancyn@cc.gatech.edu Elke Kurz-Milcke , College of Computing, Georgia Institute of Technology kurzmi@cc.gatech.edu Wendy C. Newstetter , Department of Biomedical Engineering, Georgia Institute of Technology wendy@bme.gatech.edu Jim Davies , College of Computing, Georgia Institute of Technology jim@jimdavies.org We are carrying out a research project aimed at understanding reasoning and representational practices employed in problem solving in biomedical engineering (BME) laboratories. These laboratories are best construed as evolving distributed cognitive systems: the laboratory is not simply a physical space, but a problem space, the components of which change over time; cognition is distributed among people and artifacts; and the cognitive partnerships between the technological artifacts and the researchers in the system evolve. To investigate this evolving cognitive system we use both ethnography and cognitive-historical analysis. Understanding practices in innovative research laboratories requires in-depth observation of the lab as it presently exists, as well as research into the histories of the experimental devices used in it. We are aiming here for relational accounts (Ô biographiesÕ ) of the distributed cognitive systems within the lab as they change in time. In this we find that one cannot divorce research from learning in the context of the laboratory, where learning involves building relationships with artifacts.|../mac/prof168.html",
    "Representations at Work...||Paper Track Representations at Work Martin M. Nielsen , Department of Information and Media Studies, Helsingforsgade 14, 8200 Aarhus N, Denmark mmn@imv.au.dk This paper explores relations between distributed cognition and organizational theory. A representational ontology is carried forth and utilized in an analysis of the work flow at a post office. Finally, the merits of this approach are evaluated in comparison to that of actor-network theory while spelling out further implications for organizational theory.|../mac/prof169.html",
    "A New Vision of Language...||Publication-Based Talk Track A New Vision of Language Shimon Edelman , Department of Psychology, Cornell University se37@cornell.edu A metaphor that has dominated linguistics for the entire duration of its existence as a discipline views sentences as edifices consisting of Lego-like building blocks. It is assumed that each sentence is constructed (and, on the receiving end, parsed) ab novo, starting (ending) with atomic constituents, to logical semantic specifications, in a recursive process governed by a few precise algebraic rules. The assumptions underlying the Lego metaphor, as it is expressed in generative grammar theories, are: (1) perfect regularity of what Saussure called langue, (2) infinite potential recursivity of syntactic structures, (3) unlimited human capacity for linguistic creativity, (4) the impossibility of acquiring structural knowledge from examples, and (5) the impossibility of such knowledge being stored in a memory-intensive form (ensembles of exemplars). Although these assumptions still reign in Lego-land (Lasnik, 2002; Hauser et al., 2002; Jackendoff, 2002), cognitive psychologists contend that the empirical evidence for the psychological reality of generative grammar is patchy at best, while the mathematical apparatus it postulates is overly complicated (viz. “Mathematics versus Psychology”; Tomasello, 1998, p.ix). The drive for psychological realism should not, however, preclude a theory of language from being mathematically rigorous. Indeed, the theoretical framework envisaged by Langacker (1987), which calls for a reexamination of the classical generative stance and sees grammar as a potentially large collection of units of varying complexity, or constructions (Croft, 2001), does allow for computationally sophisticated modeling of acquisition and processing (Solan et al., 2003). The new computational model gives up the logicism of generative grammar in favor of information-theoretic learning of distributed construction patterns. The structure and the meaning of a sentence (which can be thought of as the proverbial elephant groped by the blind men) are thus represented by the chorus of responses of construction detectors, which can be further processed by methods that are being worked out for another cognitive domain with similar computational needs: vision (Edelman and Intrator, 2003).|../mac/prof17.html",
    "From Prototypes to Exemplars: Representational Shifts in a Probability Judg...||Paper Track From Prototypes to Exemplars: Representational Shifts in a Probability Judgment Task Håkan Nilsson , Department of Psychology, Umeå University hakan.nilsson@psy.umu.se Peter Juslin , Department of Psychology, Umeå University peter.juslin@psy.umu.se Henrik Olsson , Department of Psychology, Uppsala University henrik.olsson@psyk.uu.se What information is used to make subjective probability judgments? In this study we test a hypothesis, grounded both in research on categorization and developmental psychology, proposing that when first confronted with an environment people create prototypes and as a function of learning they start to store concrete exemplars. The hypothesis is tested with a design that has previously proven effective at separating the two models (Juslin, Nilsson, & Olsson, 2001). The results support the hypothesized representational shift with novices relying on the representativeness heuristic but experts relying on exemplar memory when making probability judgments.|../mac/prof170.html",
    "The Meaning(s) of “If”: Conditional Probabilities and Mental Models...||Paper Track The Meaning(s) of “If”: Conditional Probabilities and Mental Models Klaus Oberauer , Allgemeine Psychologie I, University of Potsdam ko@rz.uni-potsdam.de Oliver Wilhelm , Department of Psychology, Humboldt-University Berlin oliver.wilhelm@rz.hu-berlin.de Three experiments with a probabilistic truth-table evaluation task suggest that most people interpret conditionals as asserting a high conditional probability of the consequent, given the antecedent. A minority seems to endorse an interpretation in terms of a single explicit mental model (Johnson-Laird & Byrne, 1991). There was no evidence that a substantive number of people interpret conditionals as material implications. We propose a revision of the theory of mental models that can accommodate both prevalent interpretations as two levels of elaboration of model-based representations.|../mac/prof171.html",
    "Multiple-Cue Judgment in Individual and Dyadic Learning...||Paper Track Multiple-Cue Judgment in Individual and Dyadic Learning Anna-Carin Olsson , Department of Psychology, Umeå University anna-carin.olsson@psy.umu.se Peter Juslin , Department of Psychology, Umeå University peter.juslin@psy.umu.se Henrik Olsson , Department of Psychology, Uppsala University henrik.olsson@psyk.uu.se Most studies of multiple-cue judgment focus on learning by individuals. In a multiple-cue judgment task we examined if people acquire rule or exemplar knowledge as a function of learning the task alone or in dyads. The expectation was that learning in pairs should promote explicit rule-based thinking as a consequence of increased verbalization (a social abstraction effect) and produce a larger joint exemplar knowledge base (an exemplar pooling effect). The results suggest more accurate judgments by dyads, an exemplar pooling effect, but no evidence for a social abstraction effect. In contrast to previous research, the social interaction had beneficial effects that allowed participants working in dyads to surpass their combined individual performance.|../mac/prof172.html",
    "Reduction of Uncertainty in Human Sequential Learning: Evidence from Artif...||Paper Track Reduction of Uncertainty in Human Sequential Learning: Evidence from Artificial Grammar Learning Luca Onnis , Department of Psychology, University of Warwick, l.onnis@warwick.ac.uk Morten H. Christiansen , Department of Psychology, Cornell University, mhc27@cornell.edu Nick Chater , Institute for Applied Cognitive Science and Department of Psychology, University of Warwick nick.chater@ warwick.ac.uk Rebecca Gómez , Department of Psychology, University of Arizona rgomez@u.arizona.edu Research on statistical learning in adults and infants has shown that humans are particularly sensitive to statistical properties of the input. Early experiments in artificial grammar learning, for instance, show a sensitivity for transitional n-gram probabilities. It has been argued, however, that this source of information may not help in detecting nonadjacent dependencies, in the presence of substantial variability of the intervening material, thus suggesting a different focus of attention involving change versus non-change (Gómez, 2002). In this paper we contend that alternative sources of information may be attended to simultaneously by learners, in an attempt to reduce uncertainty. With several potential cues in competition, performance crucially depends on which cue is strong enough to be relied upon. By carefully manipulating the statistical environment it is possible to weigh the contribution of each cue. Several implications for the field of statistical learning and language development are drawn.|../mac/prof173.html",
    "Grounding Functions of Instrument Plays in Dialogue: a Case-Study of Piano ...||Paper Track Grounding Functions of Instrument Plays in Dialogue: a Case-Study of Piano Duos in Joint Practice Chika Oshima , School of Knowledge Science Japan Advanced Institute of Science and Technology cooshi@jaist.ac.jp Atsushi Shimojima , School of Knowledge Science Japan Advanced Institute of Science and Technology, ATR Media Information Science Labs ashimoji@jaist.ac.jp This paper reports an exploratory study of the grounding functions of instrument plays in dialogue. Grounding theories of dialogue (Clark & Schaefer 1989, Clark 1996, Traum 1994) have been mainly modeled on exchanges in spoken language, and the contributions of non-verbal media to the grounding process are an largely untouched area. We used the Clark-Schaefer contribution model as a core theory, and analyzed six dialogues conducted in joint practice session of a piano duo play. We found that the two general functions of external representations, namely, a cognitive function for personal manipulation of information and a communicative function for interpersonal presentation of information, were co-present and integrated in piano plays in our data.|../mac/prof174.html",
    "Helping Middle Schoolers Use Cases to Reason: The CASE INTERPRETATION TOOL...||Paper Track Helping Middle Schoolers Use Cases to Reason: The CASE INTERPRETATION TOOL Jakita N. Owensby , College of Computing, Georgia Institute of Technology jowensby@cc.gatech.edu Janet L. Kolodner , College of Computing, Georgia Institute of Technology jlk@cc.gatech.edu This paper’s focus is on how we can enhance the way students in middle school use second-hand expert cases to reason. We examine the issue in the context of Learning By DesignTM (Kolodner, et. al, 1998, 2003) classrooms where students learn science in the context of addressing interesting design challenges. We have previously identified the roles cases can play in helping students learn science better, several difficulties students have with interpreting and applying cases, and the roles teachers can play in guiding students as they grapple with those difficulties (Owensby & Kolodner, 2002). We have also developed a software tool, the Case Application Suite (Owensby & Kolodner, 2002) to help students through some of the difficulties. Its design is based on suggestions from the transfer and case-based reasoning literatures and the approach to education called cognitive apprenticeship. In this paper, we discuss the design of the software, present data from an early analysis of its effectiveness, and attempt to draw some preliminary conclusions about the feasibility of dividing up responsibilities for scaffolding student learning between teacher, peers, and software tool. We address these issues in the context of project-based inquiry science, an approach to science education that has students learning science content and scientific reasoning skills in the context of addressing interesting challenges or driving questions. We have identified several affordances that cases have as resources along with some of the difficulties students have with interpreting and applying them and roles the teacher can productively play in guiding students as they grapple with those difficulties (Owensby & Kolodner, 2002). Our software suite, the Case Application Suite, is designed to scaffold students as they apply expert cases to addressing a project challenge. It includes three tools -- Case Interpretation, Case Application, and Solution Assessment. We’ve piloted it with sixth and eighth grade students in Learning by Design (Kolodner et al., 1998, 2003a, 2003b) classes as they learn geology in the context of designing a tunnel. Expert cases presented in the unit are|../mac/prof175.html",
    "Qualitative Modeling and Similarity in Back of the Envelope Reasoning...||Paper Track Qualitative Modeling and Similarity in Back of the Envelope Reasoning Praveen K. Paritosh , Qualitative Reasoning Group, Department of Computer Science, Northwestern University paritosh@cs.northwestern.edu Kenneth D. Forbus , Qualitative Reasoning Group, Department of Computer Science, Northwestern University forbus@northwestern.edu Back of the envelope reasoning involves generating quantitative answers in situations where exact data and models are unavailable and where available data is often incomplete and/or inconsistent. A rough estimate generated quickly is more valuable and useful than a detailed analysis, which might be unnecessary, impractical, or impossible because the situation does not provide enough time, information, or other resources to perform one. Such reasoning is a key component of commonsense reasoning about everyday physical situations. This paper presents a similarity-based approach to such reasoning. In a new scenario or problem, retrieving a similar example from experience, sets the stage for solving the new problem by borrowing relevant modeling assumptions and reasonable values for parameters. We believe that this tight interweaving of qualitative and analogical reasoning is characteristic of common sense reasoning more broadly. Understanding the feel for magnitudes is another crucial aspect of such reasoning, and incorporating effects of quantitative dimensions in similarity judgments and generalizations, hitherto unexplored, raises very interesting questions.|../mac/prof176.html",
    "The Role of Coherence in Category-Based Explanation ...||Paper Track The Role of Coherence in Category-Based Explanation Andrea L. Patalano , Department of Psychology, Wesleyan University apatalano@wesleyan.edu Seth Chin-Parker , Department of Psychology, University of Illinois chinpark@s.psych.uiuc.edu Brian H. Ross , Beckman Institute, University of Illinois bross@s.psych.uiuc.edu Category coherence refers to the extent to which a category is perceived to be a meaningful whole (Patalano & Ross, 2002; Ross & Patalano, 2002). We tested the hypothesis that category coherence influences the extent to which a category is used in the generation of category-based causal explanations of social behavior and preferences. In Experiments 1a and b, participants were told that members of a category shared a particular preference (e.g., sky divers prefer fiction to non-fiction), and were asked to generate the most plausible explanation for the preference. Explanations generated for high coherence categories were more plausible than those generated for low coherence categories. In Experiment 2, high and low coherence categories were contrasted in the context of a single problem. Participants were told that members of two categories (e.g., people who are both sky divers and pianists) shared a particular preference and were again asked for the most plausible explanation of the preference. References to the high coherence category occurred more often than those to the low coherence category. It is concluded that coherence influences both category selection and quality of category-based causal explanation. Implications of this work and future research directions are discussed.|../mac/prof177.html",
    "The role of space in socially distributed cognition: some issues for cognit...||Paper Track The role of space in socially distributed cognition: some issues for cognitive engineering Mark Perry , DISC, Brunel University mark.perry@brunel.ac.uk Kenton O’Hara , The Appliance Studio kenton@appliancestudio.com Gabriella Spinelli , DISC, Brunel University gabriella.spinelli@brunel.ac.uk Bill Sharpe , The Appliance Studio bill@appliancestudio.com This paper explores and identifies cognitive issues that develop out of the use of representational media by collaborating groups of people involved in problem solving. We take the analytic perspective of distributed cognition to examine the role that these artifacts have on information processing activity in augmenting human action and in transforming the problem space. The analysis is further used in identifying issues for cognitive engineering in the design of spatial, augmentative resources to support collaborative problem solving.|../mac/prof178.html",
    "Additive or Multiplicative Perceptual Noise? Two Equivalent Forms of the A...||Paper Track Additive or Multiplicative Perceptual Noise? Two Equivalent Forms of the ANCHOR Model Alexander A. Petrov , Department of Cognitive Sciences, University of California apetrov@uci.edu ANCHOR is an integrated memory-based scaling model that accounts for a wide range of phenomena in category rating and absolute identification. The model uses anchors stored in memory that serve as prototypes for each response category. The stimuli are represented by magnitudes. Two alternative formulations of the magnitude variability are considered: additive noise, which leads to logarithmic scales, and multiplicative noise, which leads to power scales. Both formulations are consistent with Weber’s and Stevens’s laws. Four variants of the ANCHOR framework systematically explore these alternative formulations. The performance of the models is evaluated against experimental data. The results show that the form of the perceptual equation is not critical for the operation of the model. Thus, the power vs. logarithmic controversy does not a®ect ANCHOR’s central claim that human scaling performance is memory-based.|../mac/prof179.html",
    "Computation and Agency in Scientific Cognition...||Publication-Based Talk Track Computation and Agency in Scientific Cognition Ronald N. Giere , Department of Philosophy & Center for Philosophy of Science, University of Minnesota giere@umn.edu I begin with a representative example of a contemporary scientific activity, observations using the Hubble Space Telescope, and ask what approaches within the cognitive sciences seem most fruitful as aids in developing an overall account of this sort of scientific activity. After presenting the Hubble Space Telescope System and a recent result, I consider applying a standard computational paradigm to this system. I find difficulties in identifying an appropriate cognitive agent and in making a suitable place for the instrumentation that constitutes such a large part of the whole system. I next consider applying the notion of distributed cognition as developed by Hutchins (1995), and then return to the question whether The Hubble System, understood as a distributed cognitive system, should be regarded as a computational system. I find a large computational component, but also an important part, the Hubble Telescope itself, that seems better characterized as a dynamic system than as a computational system. Moreover, the group of scientists interpreting the images produced by the system seem best thought of as a human/cultural system along the lines advocated by those developing a cognitive (Lakoff, 1987) or usage-based (Tomasello, 2003) approach to language acquisition and language use. I argue next that, while cognition may be theorized as distributed among both humans and instruments, there is no need to introduce into cognitive science a notion of distributed knowledge beyond simple collective knowledge. Even less is there any need to introduce notions of distributed mind or distributed consciousness. The result is that the agency involved in distributed cognitive systems remains simply human agency as ordinarily conceived. I conclude that distributed cognitive systems like The Hubble System are hybrid systems composed partly of dynamic physical systems, partly of computational systems, and partly of human cultural systems.|../mac/prof18.html",
    "Can Quirks of Grammar Affect the Way You Think? Grammatical Gender and Obje...||Paper Track Can Quirks of Grammar Affect the Way You Think? Grammatical Gender and Object Concepts Webb Phillips , MIT NE20-457 webb@mit.edu Lera Boroditsky , MIT NE20-456 lera@mit.edu an the language you speak affect the way you think? Unlike English, many languages have a grammatical gender system whereby all nouns are assigned a gender. Does talking about inanimate objects as if they were masculine or feminine actually lead people to think of inanimate objects as having a gender? A series of studies found effects of grammatical gender on people?s perceptions of similarity between objects and people. This was true even though the tasks were performed in English (a language devoid of grammatical gender), even when the tasks were non-linguistic (e.g., rating similarities between unlabeled pictures), and even while subjects were engaged in a verbal interference task. Finally, results showed that cross-linguistic differences in thought can be produced just by grammatical differences and in the absence of other cultural factors.|../mac/prof180.html",
    "Automatic Landing Technique Assessment using Latent Problem Solving Analysi...||Paper Track Automatic Landing Technique Assessment using Latent Problem Solving Analysis José Quesada , Institute of Cognitive Science, University of Colorado quesadaj@psych.colorado.edu Walter Kintsch , Institute of Cognitive Science, University of Colorado wkintsch@psych.colorado.edu Emilio Gomez , Department of Experimental Psychology University of Granada egomez@ugr.es Latent Problem Solving Analysis is applied to model the decision processes of expert instructors judging professional pilots’ landing technique in a B747 flying simulator, showing that that a memory-based model can do well in the absence of more conscious, logical processes.|../mac/prof181.html",
    "Latent Problem Solving Analysis as an explanation of expertise effects in a...||Paper Track Latent Problem Solving Analysis as an explanation of expertise effects in a complex, dynamic task José Quesada , Institute of Cognitive Science, University of Colorado quesadaj@psych.colorado.edu Walter Kintsch , Institute of Cognitive Science, University of Colorado wkintsch@psych.colorado.edu Emilio Gomez , Department of Experimental Psychology University of Granada egomez@ugr.es Latent Problem Solving Analysis (LPSA) is a theory of knowledge representation in complex problem solving that argues that problem spaces can be represented as multidimensional spaces and expertise consist on the construction of those spaces from immense amounts of experience. The model was applied using a dataset from a longitudinal experiment on control of thermodynamic systems. When the system is trained with expert-level amounts of experience (3 years), it can predict the end of a trial using the first three quarters with an accuracy of .9. If the system is prepared to mimic a novice (6 months) the prediction accuracy falls to .2. If the system is trained with 3 years of practice in an environment with no constraints, performance is similar to the novice baseline.|../mac/prof182.html",
    "The use of “that” in the Production and Comprehension of Object Relative Cl...||Paper Track The use of “that” in the Production and Comprehension of Object Relative Clauses David S. Race , Department of Psychology, University of Wisconsin-Madison drace@lcnl.wisc.edu Maryellen C. MacDonald , Department of Psychology, University of Wisconsin-Madison mcmacdonald@.wisc.edu We explore the interplay between production and comprehension by investigating why producers insert or omit the function word “that” in Object Relative Clauses, and how this choice affects comprehension. We present data from three experiments which suggests that producers insert “that” to alleviate production difficulty and in doing so create a distributional pattern of “that” use. Comprehenders are shown to be sensitive to these patterns. Implications for the interaction of comprehension and production processes are discussed.|../mac/prof183.html",
    "Why children sometimes say “mice-eater”...||Paper Track Why children sometimes say “mice-eater” Michael Ramscar , Department of Psychology, Stanford University michael@psych.stanford.edu Ivan Pearson , Division of Informatics, University of Edinburgh, Shamuna Ali S. Ali, Division of Informatics, University of Edinburgh Do the production patterns of plural forms in noun-noun compounds reveal the workings of innate constraints that govern morphological processing? Gordon (1985) claims that the fact that children in elicitation tasks produce “rat-eater but “mice-eater” to describe monsters that eat either rats or mice is evidence for a structural constraint on word formation that prevents regularly inflected forms from entering compounds. This paper re-examines the claim that adults and children ordinarily describe an eater of mice as a mice-eater. Contrary to the nativist claim that people say mice-eater because they lack an innate constraint to prevent them from doing so, it is found that once lexical priming is controlled for, even children describe a monster that eats mice as a mouseeater, which is in keeping with the kind of forms that they encounter in the input.|../mac/prof184.html",
    "Thinking Graphically: Extracting Local and Global Information...||Paper Track Thinking Graphically: Extracting Local and Global Information Raj M. Ratwani , George Mason University rratwani@gmu.edu J. Gregory Trafton , NRL Code 5513 trafton@itd.nrl.navy.mil Deborah A. Boehm-Davis , George Mason University dbdavis@gmu.edu This study investigates how information is extracted from a graph when different types of questions are asked. Although the process for extracting local information from simple graphs is understood quite well, the processes used to extract global information from more complex graphs are not as clear. In a series of two studies using verbal protocols and eye tracking, we compared responses to local and global questions. We replicated previous research on local questions, and show that people extract global information using a different set of cognitive processes.|../mac/prof185.html",
    "Constraint Satisfaction Processes in Social Reasoning...||Paper Track Constraint Satisfaction Processes in Social Reasoning Stephen J. Read , Department of Psychology, University of Southern California, read@usc.edu Chadwick J. Snow , Department of Psychology, University of Southern California, csnow@usc.edu Dan Simon , University of Southern California Law School dsimon@law.usc.edu We show that constraint satisfaction processes (coherence based reasoning) play an important role in social reasoning, and that social reasoning violates key assumptions of classic models of judgment and decision-making. Constraint satisfaction models predict a bi-directional flow of influence between evaluations of evidence for a judgment and the judgment itself, such that an evolving judgment can influence the evaluation of evidence. In contrast, models of judgment and decision-making, such as Bayesian models, SEU and Information Integration Theory assume that the direction of influence is only from the evaluations to the judgment. We examined two very different social reasoning tasks, a judgment about whether a young dating couple would get married, and a legal case about the guilt of a defendant charged with murder. In both studies, subjects exhibited strong coherence effects, such that the evaluation of evidence shifted over time to become more coherent with the final judgment. We also note the similarity of constraint satisfaction models to classic cognitive consistency theories in social psychology, such as Cognitive Dissonance Theory.|../mac/prof186.html",
    "Phonological and Distributional Cues in Syntax Acquisition: Scaling-Up the ...||Paper Track Phonological and Distributional Cues in Syntax Acquisition: Scaling-Up the Connectionist Approach to Multiple-Cue Integration Florencia Reali , Department of Psychology; Cornell University; fr34@cornell.edu Morten H. Christiansen , Department of Psychology; Cornell University mhc27@cornell.edu Padraic Monaghan , Department of Psychology, University of Warwick Padraic.Monaghan@warwick.ac.uk Recent work in developmental psycholinguistics suggests that children may bootstrap grammatical categories and basic syntactic structure by exploiting distributional, phonological, and prosodic cues. Previous connectionist work has indicated that multiple-cue integration is computationally feasible for small artificial languages. In this paper, we present a series of simulations exploring the integration of distributional and phonological cues in a connectionist model trained on a fullblown corpus of child-directed speech. In the first simulation, we demonstrate that the connectionist model performs very well when trained on purely distributional information represented in terms of lexical categories. In the second simulation we demonstrate that networks trained on distributed vectors incorporating phonetic information about words also achieve a high level of performance. Finally, we employ discriminant analyses of hidden unit activations to show that the networks are able to integrate phonological and distributional cues in the service of developing highly reliable internal representations of lexical categories.|../mac/prof187.html",
    "Eyetracking and Selective Attention in Category Learning...||Paper Track Eyetracking and Selective Attention in Category Learning Bob Rehder , Department of Psychology, New York University bob.rehder@nyu.edu Aaron B. Hoffman , Department of Psychology, New York University abh226@nyu.edu An eyetracking version of the classic Shepard, Hovland and Jenkins (1961) experiment was conducted. Forty years of research has assumed that category learning includes learning how to selectively attend to only those stimulus dimensions useful for classification. We confirmed that participants learned to allocate their attention optimally. However, we also found that neither associationist accounts of gradual learning nor hypothesis-testing accounts accurately predicted the pattern of eye movements leading up to successful learning. The implication of these results, and the use of eyetracking technology more generally, for categorization theory are discussed.|../mac/prof188.html",
    "Information Aggregation in Groups: The Approach of Simple Group-Heuristics ...||Paper Track Information Aggregation in Groups: The Approach of Simple Group-Heuristics (SIGH) Torsten Reimer , Department of Psychology, University of Basel, Missionsstrasse 60/62, 4055 Basel, Switzerland t.reimer@mpib-berlin.mpg.de Ulrich Hoffrage , Center for Adaptive Behavior and Cognition (ABC) at the Max Planck Institute for Human Development hoffrage@mpib-berlin.mpg.de In this paper, we examine interaction organizations in communication through drawing. We report that both sequential and parallel interaction organizations take place in actual human graphical communication, despite the fact that the media use involved in drawing has resource characteristics that better match parallel organizations. We argue that activity-dependent constraints imposed by the task performed in the interaction play a significant role in determining the interaction organization. We propose two types of sequentiality constraints and demonstrate that they can override the resource characteristics of the drawing media, thereby enforcing a sequential turntaking organization similar to those observed in verbal interactions.|../mac/prof189.html",
    "The Nature and Timing of Interruptions in a Complex, Cognitive Task: Empiri...||Publication-Based Talk Track The Nature and Timing of Interruptions in a Complex, Cognitive Task: Empirical Data and Computational Cognitive Wayne D. Gray , Cognitive Science Department, Rensselaer Polytechnic Institute grayw@rpi.edu Michael J. Schoelles , Cognitive Science Department, Rensselaer Polytechnic Institute schoem@rpi.edu Much can be revealed regarding the cognitive control of interactive behavior by understanding the way in which multiple tasks are interleaved. Across three studies we held one task constant and varied the nature of the secondary task. The main task was the Argus Prime (Schoelles & Gray, 2001) simulated task environment (Gray, 2002). Argus is a complex task designed to mimic aspects of a radar operator’s job. The basic paradigm entails keeping track of 20 moving targets on a radar-screen interface, calculating the threat value of each, and classifying the target by its threat value. In a 12-min scenario, 70-85 accurate calculations and classifications are required for perfect performance. Across all three studies, part 1 entailed an hour of training and one hour of practice on four 12-min scenarios. The three studies differed in the last hour. In study 1, subjects continued the basic Argus task for 4 additional, 12-min scenarios. Study 2 added a perceptual-motor secondary task. As subjects performed the primary task, they were required to use the mouse to track a plane that was randomly moving around the right-side of the screen. With accurate tracking the cursor turned blue. As inaccuracies increased, it turned yellow, and then red. Avoiding “red” required that subjects monitor cursor color and switch from Argus to the secondary task whenever the cursor turned yellow. Study 3 traded the perceptual-motor task for a cognitive “alpha” task. For this task subjects heard one randomly sampled letter of the alphabet every 4-s. Their task was to press “x” if the current letter (n) came before the prior letter (n-1) in the alphabet (e.g., n = c, n-1 = m) or “v” if it came after (e.g., n = s, n-1 = m). Schoelles (2002) created 24 simulated-human users (SHUs) in ACT-R 5.0. Each SHU incorporated a different combination of strategies observed by our subjects as they performed study 1. The fit of model performance to various measures of human performance was quite good and accurately mimicked overall performance as well as performance on each of 4 within-subject interface conditions. For study 2, the same 24 SHUs were used. Predictions as to when subjects would interrupt the classification task to switch to the tracking task were made based on a task analysis of Argus. The only change to the 24 SHUs was to add productions that would consider switching to the tracking at the completion of certain subtasks. For study 2, the models predicted overall performance on Argus and on each of the four within-subject interface conditions as well as they did for study 1. As the models were based on study 1 data, we argue that their use in study 2 is essentially a zero-parameter fit. Unfortunately, more detailed examination of the data showed that human subjects switched from Argus to the tracking task with a much higher frequency than did the SHUs. Screen shot of Argus Prime from Study 2. Radar display is on left-side. Operator position is represented by the small + at the bottom left. The flying numbers represent planes. Right-side shows information window at top with information on the currently selected target. Bottom right shows the randomly moving target and the cursor with which the operator is supposed to track the target. For the original model of task switching, control of behavior was top-down (i.e., goal directed). We have created models that incorporate top-down and/or bottom-up control of task switching. During the talk we will compare and contrast the performance of these models in fitting the study 2 and study 3 data.|../mac/prof19.html",
    "Choice set options affect the valuation of risky prospects...||Paper Track Choice set options affect the valuation of risky prospects Stian Reimers , Department of Psychology, University of Warwick stian.reimers@warwick.ac.uk Neil Stewart , Department of Psychology, University of Warwick neil.stewart@warwick.ac.uk Nick Chater , Department of Psychology, University of Warwick nick.chater@warwick.ac.uk A series of experiments is used to investigate the extent to which people’s valuation of a risky prospect is affected by the values from which a participant selects a response. Three variables were considered: a smaller risk-free amount, a larger risky amount, and the probability of winning the larger amount. There were three conditions: in each, two of the three variables were held constant, and participants chose the value for the third variable that made the risky and risk-free options worth the same to them. This was done first by a freechoice valuation, and then, with different participants, by choosing one of four options that were either all below or all above the population free-choice median. The options presented had a strong effect on valuation of the missing variable. This effect remained even when the free-choice and multiple-choice conditions were presented within subjects. This demonstrates that people showing rational and consistent risk evaluation strategies could have their risk aversion manipulated by context. Overall, the experiments suggest that people’s propensity for risk aversion is manipulable by context. This is problematic finding for traditional, context-independent, theories of decision under risk.|../mac/prof190.html",
    "Counterfactual Reasoning: How to Organize a Possible World...||Paper Track Counterfactual Reasoning: How to Organize a Possible World Russell Revlin , Department of Psychology, University of California revlin@psych.ucsb.edu Dustin P. Calvillo , Department of Psychology, University of California calvillo@psych.ucsb.edu mautone@psych.ucsb.edu , Department of Psychology, University of California Counterfactual reasoning appears to be a universal phenomenon of human inference from childhood to adulthood, yet, the prevailing explanations seem able to capture only a limited aspect of the process and are in need of an overarching framework. We propose that David Lewis’ possible worlds analysis offers a first approximation to such a framework and gives a psychologically plausible account of counterfactuals. It identifies the unique properties of our ability to reason from false assumptions—whether talking about pretense or revising our beliefs. Three experiments are offered to suggest the plausibility of this account.|../mac/prof191.html",
    "The Role of Strategies and Instructions in Relational Deductive Reasoning...||Paper Track The Role of Strategies and Instructions in Relational Deductive Reasoning A. Emanuel Robinson , School of Psychology, Georgia Tech gte581m@mail.gatech.edu Christopher Hertzog , School of Psychology, Georgia Tech Christopher.hertzog@psych.gatech.edu Deductive reasoning is often seen as being composed of an immutable mechanism, universal to all reasoning situations and consisting of either mental models (e.g., Johnson-Laird, 1983) or formal-rules (e.g., Rips, 1994). Many researchers have questioned whether these positions are truly mutually exclusive (e.g., Roberts, 1993, 2000). Most deductive reasoning research has largely ignored the influence of instructions and strategies on the reasoning process. The present experiment was conducted to investigate reasoning strategies along with metacognitive measures of those strategies. Instructions were given to use a particular strategy (e.g., spatial, verbal, or naïve). Items were separated into two levels: simple and complex, defined by the amount of premises in each. Premise times, accuracy, and metacognitive strategy reports were collected. Instructions had an effect on performance, as seen in premise times and accuracy. Also, strategy reports indicated a distribution of strategies utilized by participants. Strategy reports proved vital in corroborating differential patterns of performance indicative of varied approaches to solving this task.|../mac/prof192.html",
    "Attending to auditory and visual input with flexibility: Evidence from 4-ye...||Paper Track Attending to auditory and visual input with flexibility: Evidence from 4-year-olds Christopher W. Robinson , Center for Cognitive Science, The Ohio State University robinson.777@osu.edu Vladimir M. Sloutsky sloutsky.1@osu.edu Center for Cognitive Science, The Ohio State University Previous research established that while infants and young children show a preference for auditory input, adults have a preference for visual input. In this research, we hypothesize that children and adults are flexible attenders, and they may switch their modality preference under different stimuli conditions. The results from the current experiments demonstrate that very simple changes in visual stimuli yield a change in attentional weights in 4-year-olds, but not adults. Understanding how changes in auditory and visual information influence shifts in attention at various points in development may provide an important tool for understanding the effects of labels and other forms of auditory information on semantic development.|../mac/prof193.html",
    "Computational offloading: Supporting distributed team working through visua...||Paper Track Computational offloading: Supporting distributed team working through visually augmenting verbal communication Yvonne Rogers , Interact Lab, School of Cognitive and Computing Sciences, yvonner@cogs.susx.ac.uk Harry Brignull , Interact Lab, School of Cognitive and Computing Sciences harrybr@cogs.susx.ac.uk Distributed team working often involves close-knit groups collaborating over a large geographical space performing time-critical tasks. We present a field study of the way a dispersed team of technicians coordinate their work, highlighting the phenomenon of extraneous ‘detective work’ – where much communication, via walkie-talkies, needs to take place to resolve uncertainty arising in their work. We suggest one way of improving the way team members maintain their awareness of what is going on in different places and times is to offload some of the computation involved, by augmenting the verbal channel with visual information. Using the external cognition framework, we describe how we designed a dynamic visualization that allowed salient verbal information to be re-represented as an external cognitive trace. To test our assumption about externalization and computational offloading, we carried out an experiment, with three different conditions: visualization, pen and paper and no cognitive aid. Our findings showed that allowing users to create and view a dynamic visualization improves awareness of what is going on and the way distributed work is coordinated.|../mac/prof194.html",
    "Category Structure and Recognition Memory...||Paper Track Category Structure and Recognition Memory Yasuaki Sakamoto , Department of Psychology, University of Texas yasu@psy.utexas.edu Bradley C. Love , Department of Psychology, University of Texas love@psy.utexas.edu Current models of human category learning and subsequent recognition are either exemplar-based, rule-based, or some combination of both approaches. We present learning and recognition data that cannot be accounted for by current approaches. The data suggest that the degree to which an item is remembered is determined by the strength of the expectation it violates. In our study, expectations take the form of simple, imperfect rules where the strength of a rule is determined by the number of items that follow the rule in training. Exemplarbased models cannot account for the results because they do not posit organizing knowledge structures that can be violated. The frequency insensitivity of rule-based accounts leads to their failure. We propose a cluster-based approach that is consistent with our findings, as well as schema, stereotype, and basic memory research.|../mac/prof195.html",
    "Toward a Unified Framework for Tracking Cognitive Processes...||Paper Track Toward a Unified Framework for Tracking Cognitive Processes Dario D. Salvucci , Department of Computer Science, Drexel University salvucci@cs.drexel.edu Timothy Siedlecki , Department of Computer Science, Drexel University ts62@drexel.edu In this paper we present an initial specification of a general, robust, and efficient computational framework for tracking cognitive processes — that is, inferring a persons’ thoughts from their actions. Our framework, which we call the mind-tracking architecture, centers on two core processes: generating predicted cognitive and action sequences using computational cognitive models, and tracking observed actions through robust matching with predicted actions. In essence, the mind-tracking architecture “thinks along” with the person in predicting a set of possible thoughts and actions, and then matches these to the person’s observed actions to infer their most likely thoughts. In the paper we provide a background of related work (e.g., for intelligent tutoring systems), outline the basic components of the architecture, and demonstrate its usefulness for a sample real-world application — real-time inference of driver intentions.|../mac/prof196.html",
    "Meta-Cognitive Architecture for Team Agents...||Paper Track Meta-Cognitive Architecture for Team Agents Alexei V. Samsonovich , Krasnow Institute for Advanced Study, George Mason University asamsono@gmu.edu Kenneth A. De Jong , Department of Computer Science and Krasnow Institute for Advanced Study, George Mason University kdejong@gmu.edu A key element of our approach is the interpretation of “self” in a meta-cognitive sense: that is, “self” is understood as a virtual character representing an agent as the subject of experience, as the target of attribution of experiences and deliberate actions performed by this agent. Thus understood, “self” can be represented as an element in an agent’s cognitive system and can be used for meta-cognitive processing: i.e., reasoning about one’s own self and other selves. This general idea reflects a simulationist theory-ofmind viewpoint (Nichols & Stich, 2000), which is taken as the basis for our approach. Our model of an agent’s mind includes multiple instances of “self” representing notions of INow, I-Yesterday, I-Imagine, I-Goal, etc. Each instance of “self” is represented by a “chart” with a set of properties and mental states attributed to it. Thus, mental states in this framework are representations of experiences attributed to a particular instance of “self”. This attribution further implies certain rules and constraints imposed on the contents and the dynamics of representations. The result is a general architecture that will enable in intelligent agents a metacognitive “common sense”, which proves to be vital in a variety of paradigms and scenarios requiring cooperation within a team.|../mac/prof197.html",
    "Sequence Effects in Solving Knowledge-Rich Problems: The Ambiguous Role of ...||Paper Track Sequence Effects in Solving Knowledge-Rich Problems: The Ambiguous Role of Surface Similarities Katharina Scheiter , Department of Applied Cognitive Psychology and Media Psychology, University of Tuebingen k.scheiter@iwm-kmrc.de Peter Gerjets , Multimedia and Hypermedia Research Unit, Knowledge Media Research Center p.gerjets@iwm-kmrc.de Sequence effects are said to occur whenever the problemsolving performance varies as a function of the order in which problems are solved. We present a framework that explain sequence effects as a result of (a) learning during solving a problem and of (b) transferring the learned content on succeeding problems. In two experiments we studied the ambiguous influence of surface similarities on sequence effects among succeeding, structurally dissimilar knowledgerich problems. These experiments demonstrate that surface similarities may either foster performance or lead to negative transfer depending on whether problem solvers are already aware of the structural features of the problems prior to solving them.|../mac/prof198.html",
    "Interactive Processing of Morphosyntactic Features in the Bilingual Lexicon...||Paper Track Interactive Processing of Morphosyntactic Features in the Bilingual Lexicon Matthias Scheutz , Department of Computer Science and Engineering, University of Notre Dame mscheutz@nd.edu Kathleen Eberhard , Department of Psychology, University of Notre Dame keberhar@nd.edu Kathleen Targowski , Department of Psychology, University of Notre Dame ktargows@nd.edu We investigated whether morphosyntactic representations in bilinguals’ native language become activated by form-related representations in their second language. Specifically, we examined whether, relative to English monolinguals, German- English bilinguals would exhibit a bias to initially interpret English nouns ending in –er (e.g., partner) as referring to a male due to the association of masculine gender with this ending in German. This question was tested by recording bilinguals' and monolinguals' eye movements as they read English sentences containing a stereotypically male, female, or neutral –er noun that was an antecedent of a gendermatching or –mismatching reflexive. The bias to initially interpret the –er nouns' referent as male was assessed by the size of the mismatch effect or the difference between the participants' total fixation duration on the gender-matching vs. gender-mismatching reflexives. The results showed a mismatch effect for both the bilinguals and monolinguals in the stereotypical male condition, no mismatch effect for either group in the neutral condition, and a mismatch effect only for the monolinguals in the stereotypical female condition. The results were simulated using a connectionist interactive activation model, and the implications for future study are|../mac/prof199.html",
    "Latent Semantic Analysis: Theory, Use and Applications...||Tutorials Track Latent Semantic Analysis: Theory, Use and Applications Simon Dennis , Institute of Cognitive Science, University of Colorado Simon.Dennis@colorado.edu Tom Landauer , Institute of Cognitive Science, University of Colorado Tom.Landauer@colorado.edu Walter Kintsch , Institute of Cognitive Science, University of Colorado Walter.Kintsch@colorado.edu Jose Quesada , Institute of Cognitive Science, University of Colorado Jose.Quesada@colorado.edu Latent Semantic Analysis (LSA) is a theory and method for extracting and representing the contextual-usage meaning of words by statistical computations applied to a large corpus of text (Landauer and Dumais, 1997). The underlying idea is that the aggregate of all the word contexts in which a given word does and does not appear provides a set of mutual constraints that largely determines the similarity of meaning of words and sets of words to each other. These constraints can be solved using linear algebra methods, in particular, the Singular Value Decomposition. The adequacy of LSA’s reflection of human knowledge has been established in a variety of ways. For example, its scores overlap those of humans on standard vocabulary and subject matter tests; it mimics human word sorting and category judgments; it simulates word–word and passage– word lexical priming data; and it accurately estimates passage coherence, learnability of passages by individual students, and the quality and quantity of knowledge contained in an essay. In addition, LSA has found application in a number of areas including selecting educational materials for individual students, guiding online discussion groups, diagnosing mental disorders from prose, matching jobs with candidates and facilitating automated tutors.|../mac/prof2.html",
    "The Social Emergence of Communication in Spatialized Arrays of Neural Nets...||Publication-Based Talk Track The Social Emergence of Communication in Spatialized Arrays of Neural Nets Patrick Grim , Group for Logic & Formal Semantics, Dept. of Philosophy, SUNY Stony Brook pgrim@notes.cc.sunysb.edu Paul St. Denis , Group for Logic & Formal Semantics, Dept. of Philosophy, SUNY Stony Brook pstdenis@notes.cc.sunysb.edu Trina Kokalis , Group for Logic & Formal Semantics, Dept. of Philosophy, SUNY Stony Brook tkokalis@notes.cc.sunysb.edu We build on a background of game-theoretic work regarding cooperation (Axelrod 1984, Nowak and Sigmund 1992), particularly the emergence of higher levels of cooperation in the Spatialized Prisoner’s Dilemma (Grim 1995, Grim 1996, Grim, Mar, & St. Denis 1998). Here we extend that work to the larger topic of communication, exploring spatialized models with large arrays of agents in an environment of wandering food sources and predators. Our individuals are simple neural nets; we use two species of neural nets in two series of runs. In each case we start with a spatialized cellular automata array of over 4,000 individuals with randomized weights and biases, and have them do a partial training on the behavior of more successful neighbors. In this environment each individual is capable of making arbitrary sounds and of responding to sounds from immediate neighbors by opening its mouth, hiding, or coasting in neutral. An individual whose mouth is open in the presence of a wandering food source is ‘fed’ and gains points; an individual who fails to hide in the presence of a predator is ‘hurt’ by losing points. But opening mouths, hiding, and making sounds each exact an energy cost. Our models differ from most of their predecessors in that (1) all action and neural net training is purely local, and (2) all gains are individual; there is no symmetrical reward for communication per se.|../mac/prof20.html",
    "Examining the Role of Prediction in Infants' Physical Knowledge...||Paper Track Examining the Role of Prediction in Infants' Physical Knowledge Matthew Schlesinger , Department of Psychology, Southern Illinois University MATTHEWS@Siu.Edu Michael E. Young , Department of Psychology, Southern Illinois University MEYOUNG@Siu.Edu The violation-of-expectation paradigm investigates infants' physical knowledge by exploiting their tendency to look longer at events that are surprising, unexpected, or physically impossible. The current simulation study examines the role of prediction as a fundamental component of infants' expectations in physical-knowledge studies. A simple recurrent network is presented with a computer-animated version of Baillargeon’s “car study” (1986; Baillargeon & DeVos, 1991), in which a car rolls down a ramp and behind a screen. After learning to predict the outcome of a training event, the model is then tested on possible and impossible events from the same study. During testing, the model successfully predicts only superficial features of the test events. These results are used to argue for the necessity of prior physical knowledge, and perhaps also a built-in capacity for mental representation, in order for a prediction system to work.|../mac/prof200.html",
    "Analyzing Effects of Goal Competition and Task Difficulty in Multiple-Task ...||Paper Track Analyzing Effects of Goal Competition and Task Difficulty in Multiple-Task Performance: Volitional Action Control within ACT-R Tina Schorr , Virtual Ph.D. Program: Knowledge Acquisition and Knowledge Exchange with New Media t.schorr@iwm-kmrc.de Peter Gerjets , Multimedia and Hypermedia Research Unit, Knowledge Media Research Center p.gerjets@iwm-kmrc.de Katharina Scheiter , Department of Applied Cognitive Psychology and Media Psychology, University of Tuebingen k.scheiter@imw-kmrc.de In this paper we present a cognitive-modeling analysis of processes of volitional action control in multiple-task performance. We simulated experimentally obtained effects of goal competition and task difficulty on processing strategies and performance in a high-level learning and problem-solving task within the ACT-R architecture. Beyond the model's capacity for explaining the empirical pattern of results, the fundamental processing mechanisms used for modeling, i.e., activation mechanisms and executive control productions are in line with current explanations of elementary executive control processes. Thus, we consider our modeling approach to be a solution to fill the gap between volitional control demands in real-world multiple-task performance and experimental findings on elementary executive functions.|../mac/prof201.html",
    "Some New Evidence for Concept Stability...||Paper Track Some New Evidence for Concept Stability Sam Scott , Department of Cognitive Science, Carleton University sscott@carleton.ca It is generally assumed that concepts are stored as relatively invariant chunks in long term memory. This is the default assumption in most linguistic, psychological, and philosophical work on concepts. But contrary to this assumption is the recent suggestion that both the structure and content of concepts is highly flexible and unstable. I critically review the evidence behind this assertion, then present some new experimental evidence that supports the traditional view.|../mac/prof202.html",
    "Incubation in Problem Solving as a Context Effect...||Paper Track Incubation in Problem Solving as a Context Effect Rachel Seabrook , Department of Psychology, Oxford Brookes University rseabrook@brookes.ac.uk Zoltán Dienes , Department of Psychology, Sussex University Z.Dienes@sussex.ac.uk Anagrams were used to test the hypothesis that incubation is the result of a change in context between two attempts at a problem. The context was manipulated between two sessions of work on the anagrams by presenting word searches containing words from a single category (either animals or fruit and vegetables) prior to each session of problem solving. Some of the anagrams had solutions from one of these two categories; these were compared with distracters, the solutions of which belonged to neither category. The anagrams subject to the context manipulation showed an incubation effect (superior performance for items attempted in two sessions relative to controls attempted in only one session) whereas distracters did not, thus supporting the hypothesis.|../mac/prof203.html",
    "Your Task is my Task: Shared Task Representations in Dyadic Interactions...||Paper Track Your Task is my Task: Shared Task Representations in Dyadic Interactions Natalie Sebanz , Max Planck Institute for Psychological Research, sebanz@psy.mpg.de Günther Knoblich , Max Planck Institute for Psychological Research, knoblich@psy.mpg.de Wolfgang Prinz , Max Planck Institute for Psychological Research, prinz@psy.mpg.de While some cognitive scientists regard social interactions as just another form of environmental interaction, others have proposed that social interactions place special demands on the cognitive systems involved, and may have shaped individual minds in particular ways. One consequence of the demands of action coordination between several individuals could be that in social interactions, others’ tasks are represented and integrated in one’s own action plans even when coordination is not required. To test this assumption, we investigated the performance of pairs of participants carrying out complementary and different tasks alongside each other, and compared this to performance in exactly the same tasks carried out alone. The task of another agent influenced individual performance in the group setting. Performance was also modulated by the nature of the other’s task. This suggests that individuals shared task representations and integrated each other´s task in their own action planning. The results are consistent with the view that the demands of joint action have shaped mental processes, and support the claim that cognition must be studied in relation to social context.|../mac/prof204.html",
    "Do We Really Reason about a Picture as the Referent?...||Paper Track Do We Really Reason about a Picture as the Referent? Atsushi Shimojima , Japan Advanced Institute of Science and Technology ATR Media Information Science Research Labs ashimoji@jaist.ac.jp Takugo Fukaya tfukaya@atr.co.jp A signi?cant portion of the previous accounts of inferential utilities of graphical representations (e.g.,Sloman, 1971; Larkin&Simon,1987) implicitly relies on the existence of what may be called inferences through hypothetical drawing. However,conclusi ve detections of them by means of standard performance measures have turned out to be dif?cult (Schwartz,1995). This paper attempts to ?ll the gap and provide positive evidence to their existence on the basis of eye-tracking data of subjects who worked with external diagrams in transitive inferential tasks.|../mac/prof205.html",
    "Accuracy of Tutors’ Assessments of their Students by Tutoring Context...||Paper Track Accuracy of Tutors’ Assessments of their Students by Tutoring Context Stephanie A. Siler , University of Pittsburgh; Learning Research and Development Center siler@pitt.edu Kurt VanLehn , University of Pittsburgh; Learning Research and Development Center vanlehn@pitt.edu In this study we compared the accuracy of tutors’ assessments of their students’ general competence, conceptual knowledge and affective state in two different tutoring contexts: face-to-face (FTF) and computer-mediated (CM). We found that the accuracy of tutors’ assessments of their students was dependent on both the type of student information that was assessed, and, to a lesser extent, the tutoring context . Only tutors’ assessments of their students’ general competence, as opposed to their assessments of their students’ individual conceptual knowledge or their students’ motivation, was affected by the manipulations.|../mac/prof206.html",
    "Causal Models can be Used to Predict Base-Rate Neglect...||Paper Track Causal Models can be Used to Predict Base-Rate Neglect Steven A. Sloman , Department of Cognitive & Linguistic Sciences, Brown University Steven_Sloman@brown.edu Causal models are used to predict individuals' probability judgments on the taxicab problem of Tversky and Kahneman (1982). Predictions are based on the hypothesis that judgments take into account only those variables that are judged causally relevant. Two versions of the problem were tested, one with and one without causally-relevant base rates. The results showed that causal models were able to predict judgments reasonably well. However, the data failed to replicate Tversky and Kahneman's finding of a difference between the two conditions.|../mac/prof207.html",
    "Cross-Cultural Differences in the Input to Early Word Learning...||Paper Track Cross-Cultural Differences in the Input to Early Word Learning Jesse Snedeker , Department of Psychology, Harvard University snedeker@wjh.harvard.edu Peggy Li P. Li, Department of Psychology, Harvard University pegs@wjh.harvard.edu Sylvia Yuan , Department of Psychology, Harvard University yuansy@wjh.harvard.edu Young English learners typically have vocabularies dominated by nouns, while their Mandarin learning counterparts have relatively more verbs. Novice language learners must initially learn words from the situational contexts in which they appear, since they have limited access to syntactic or semantic context. These studies explore whether differences in these situational contexts could account for the difference in early vocabulary composition. We assessed the potency of this word-to-world mapping procedure in each language by asking adults to identify nouns and verbs from their extralinguistic contexts maternal speech. The full pattern is examined in English native speakers given either English or Mandarin input (Exp. 1) and Mandarin native speakers given English or Mandarin input (Exp. 2). Results indicate that adults identify more nouns than verbs in English input, but identify as many verbs as nouns in Mandarin input. Thus cross-linguistic differences in the information that is available in extralingistic contexts may account for some of the differences in early vocabulary composition.|../mac/prof208.html",
    "Interventions do not solely benefit causal structure learning: Being told w...||Paper Track Interventions do not solely benefit causal structure learning: Being told what to do results in worse learning than doing it yourself David M. Sobel , Department of Cognitive and Linguistic Sciences, Brown University David_Sobel_1@brown.edu Tamar Kushnir , Department of Psychology, Tolman Hall, University of California Tkushnir@socrates.berkeley.edu Previous research has demonstrated that causal structure learning is facilitated by observing interventions on a causal system (e.g., Lagnado & Sloman, 2002). However, the question remains as to whether the origins of those interventions influence learning. Sobel (2003) demonstrated that causal structure learning was facilitated when learners observed the results of their own interventions as opposed to when they observed the results of another’s interventions, even though the data they observed was identical. He proposed that this difference could be accounted for by “active learning” Bayesian algorithms, following Steyvers et al. (in press). The present experiment compared learning when learners were able to intervene on a causal system themselves vs. when learners were forced to make particular interventions. Although learners made the same interventions and observed identical data, learning was better when participants observed the results of their own interventions, consistent with the predictions of these Bayesian algorithms.|../mac/prof209.html",
    "Inductive Simplicity and the Matrix...||Publication-Based Talk Track Inductive Simplicity and the Matrix Gilbert Harman , Department of Philosophy, Princeton University harman@princeton.edu Sanjeev Kulkarni , Department of Electrical, Princeton University kulkarni@princeton.edu In certain statistical learning problems, a policy of choosing simpler rules that account fairly well for data is likely to have less error on new cases than a policy of choosing complex rules that have less error on the data. The relevant kind of simplicity is not to be measured in terms of the number of parameters needed to specify a given member of a class of rules but might be measured in terms of the VC dimension of such a class. The rationale for using simplicity so measured can be extended to allow simplicity to decide among empirically equivalent hypotheses. The extended rationale provides reasons of simplicity to reject certain sorts of philosophical skepticism. curve that fits the data, or, more precisely, that inductive generalization must balance considerations of data-coverage against considerations of simplicity. If so, we need an account of the sort of simplicity that plays or could legitimately play such a role in inductive reasoning. Simplicity Distinguish two goals for an account of simplicity in this sense. One is to account for the sorts of simplicity people actually use in inductive generalization. Another is to say what sorts of simplicity it is reasonable|../mac/prof21.html",
    "Unsupervised efficient learning and representation of language structure...||Paper Track Unsupervised efficient learning and representation of language structure Zach Solan , Faculty of Exact Sciences Tel Aviv University rsolan@post.tau.ac.il David Horn , Faculty of Exact Sciences Tel Aviv University horn@post.tau.ac.il Eytan Ruppin , Faculty of Exact Sciences Tel Aviv University ruppin@post.tau.ac.il Shimon Edelman , Department of Psychology Cornell University se37@cornell.edu We describe a linguistic pattern acquisition algorithm that learns, in an unsupervised fashion, a streamlined representation of corpus data. This is achieved by compactly coding recursively structured constituent patterns, and by placing strings that have an identical backbone and similar context structure into the same equivalence class. The resulting representations constitute an efficient encoding of linguistic knowledge and support systematic generalization to unseen sentences.|../mac/prof210.html",
    "Linguistic Cues Influence Acquisition of Number Words...||Paper Track Linguistic Cues Influence Acquisition of Number Words Eugenia Steingold , Laboratory for Developmenta Studies Harvard University Genya@wjh.harvard.edu Elizabeth Spelke , Laboratory for Developmenta Studies Harvard University Spelke@wjh.harvard.edu Audrey Kittredge , Brown University Audrey_Kittredge@brown.edu A cross-linguistic study investigated how linguistics cues influence young children s number representations. We compared how English monolinguals and Russian-English bilinguals perform on number discrimination tasks. Subjects performances strongly suggest that linguistic cues, such as plural markings, guide children s initial acquisition and representation of number.|../mac/prof211.html",
    "No Unified Scales for Perceptual Magnitudes: Evidence from Loudness...||Paper Track No Unified Scales for Perceptual Magnitudes: Evidence from Loudness Neil Stewart , Department of Psychology, University of Warwick neil.stewart@warwick.ac.uk Nick Chater , Department of Psychology, University of Warwick nick.chater@warwick.ac.uk In this article we consider whether perceptual magnitudes are represents on unified underlying scales. We exploite the ubiquitous sequential effects seen in judgments concerning the attributes of simple perceptual stimuli. Participants made judgments about the intensity of sinusoidal tones and white noise hisses. On each trial in Experiment 1, participants heard a tone and a hiss and judged which was the louder. The loudness of a stimulus was assimilated only to the loudness of the same type of stimulus on the previous trial. In Experiment 2, the effect of the stimulus on the previous trial in an absolute identification of loudness task was larger when previous and current stimuli were of the same type. The attenuation of sequential effects by a switch of stimulus types suggests that the loudness of tones is not represented in the same way as the loudness of hisses. We argue that these sequential effects are indicative of the relativity of perceptual judgment.|../mac/prof212.html",
    "Schema-driven Memory and Structural Alignment...||Paper Track Schema-driven Memory and Structural Alignment C. Hunt Stilwell , Univeristy of Texas at Austin, Department of Psychology stilwell@psy.utexas.edu Arthur B. Markman , Univeristy of Texas at Austin, Department of Psychology markman@psy.utexas.edu In the 1960s and 70s, researchers paid a great deal of attention to schematic memory, or memory for thematic information. In the 1980s and 90s, this research ground to a halt, in large part because researchers lacked a strong theoretical framework for the study of schematic memory. We suggest that structure-mapping theory, a theory of analogy and similarity comparisons, may provide such a framework. We conduct two experiments designed to replicate classic findings from the schematic memory literature - that memory for schema-relevant information is better than that for schema-irrelevant information, and that information in a schema can intrude on memory for instances - using a paradigm from work in analogical reasoning. Because we can replicate schema-driven findings in an analogy paradigm, we should be able to use what we know about analogy to understand schematic memory.|../mac/prof213.html",
    "Accounting For Discovery in a Conginitive Architecture...||Paper Track Accounting For Discovery in a Conginitive Architecture Ron Sun , Department of Cognitive Science, Rensselaer Polytechnic Institute, Troy, NY 12180, USA rsun@rpi.edu Xi Zhang , Department of CECS, University of Missouri xzf73@mizzou.edu|../mac/prof214.html",
    "Beyond the Bounds of Cognition...||Paper Track Beyond the Bounds of Cognition Tarja Susi , Department of Computer Science, University of Skövde tarja@ida.his.se Jessica Lindblom , Department of Computer Science, University of Skövde jessica@ida.his.se Tom Ziemke , Department of Computer Science, University of Skövde tom@ida.his.se One of the questions that frequently come up in discussions of situated, embodied and distributed cognition is where to draw the boundary between cognisers and their environment. Adams and Aizawa (2001) have recently formulated a critique of what they consider a “radical view of tool use”, i.e., the view of tools as part of the cognitive system. We analyse their critique and show that much of what they consider ‘radical’ turns out to be compatible with what they consider ‘common sense’. Hence, we argue that much of the debate boils down to a disagreement over different uses of the term ‘cognitive’, whereas there is growing agreement about the central role that agent-environment interaction in general, and tool use in particular, play in cognitive processes. We therefore suggest to drop the ‘bounds of cognition’ debate, and conclude by raising what we consider more important questions in the study of cognitive tool use.|../mac/prof215.html",
    "Constructive Perception: A Skill for Coordinating Perception and Conception...||Paper Track Constructive Perception: A Skill for Coordinating Perception and Conception Masaki Suwa , PRESTO, JST & School of Computer and Cognitive Sciences, Chukyo University suwa@sccs.chukyo-u.ac.jp Barbara Tversky , Department of Psychology, Stanford University bt@psych.stanford.edu One aspect of expertise, in chess, in diagrammatic reasoning, in design, and in other domains, is inferring conceptual ideas from perception. We propose that underlying a skill for this is a process we call constructive perception, the deliberate adoption of perceptual strategies in the service of cognition, from comprehension to creativity. In the case of enabling new ideas in design, this seems to be a coordination of two processes: reorganizing perception and associating ideas. The present research presents evidence for the two components underlying constructive perception. Generating new interpretations of ambiguous sketches was correlated independently with a perceptual ability, reorganizing parts of figures, and with a conceptual ability, associative fluency. We speculate on implications for expertise and for its’ nurturing.|../mac/prof216.html",
    "Constraints on Generalization: Why are Past-Tense Irregularization Errors s...||Paper Track Constraints on Generalization: Why are Past-Tense Irregularization Errors so Rare? Niels Taatgen , Department of Artificial Intelligence niels@ai.rug.nl Mariëtte Dijkstra , Departments of Humanities Computing and English m.e.dijkstra@student.rug.nl We present an extension of the ACT-R model by Taatgen and Anderson (2002) of learning the English past tense that can take into account the production of irregularization errors. These errors are produced by using examples of irregular verbs as the basis for analogy. The relative rareness of irregularization errors puts constraints on the rule generalization process The model explains this by the fact that the probability for a particular irregularization is too low to establish a rule.|../mac/prof217.html",
    "Learning Causal Laws...||Paper Track Learning Causal Laws Joshua B. Tenenbaum , Department of Brain and Cognitive Sciences MIT jbt@mit.edu Sourabh Niyogi , Department of Electrical Engineering and Computer Science MIT niyogi@mit.edu|../mac/prof218.html",
    "The Concept of Voluntary Motor Control in the Recent Neuroscientific Liter...||Paper Track The Concept of Voluntary Motor Control in the Recent Neuroscientific Literature Paul Tibbetts, Ph.D. , Philosophy and Cognitive Science, University of Dayton paul.tibbetts@notes.udayton.edu The concept of voluntary motor control (VMC) frequently appears in the neuroscientific literature, specifically in the context of cortically-mediated, intentional motor actions. For cognitive scientists, this concept of VMC raises a number of interesting questions: (i) Are there dedicated, modular-like structures within the motor system associated with VMC? Or (ii) is it the case that VMC is distributed over multiple cortical as well as subcortical structures? (iii) Is there any one place within the so-called hierarchy of motor control where voluntary movements could be said to originate? And (iv) in the current neurological literature how is the adjective voluntary in VMC being used? These questions are here considered in the context of how higher- and lower-levels of motor control plan, initiate, coordinate, sequence, and modulate goal-directed motor outputs in response to changing internal and external inputs. Particularly relevant are the conceptual implications of current neurological modeling of VMC concerning intentional agency. Key Words: causal agency, intentional behavior, voluntary, voluntary motor control, will.|../mac/prof219.html",
    "Epistemic Mediators and Manipulative Abduction in Model-Based Reasoning...||Publication-Based Talk Track Epistemic Mediators and Manipulative Abduction in Model-Based Reasoning Lorenzo Magnani , Department of Philosophy and Computational Philosophy Laboratory lmagnani@unipv.it Science expresses one of the most creative forms of human reasoning. It has been stressed that scientific discovery is a mysterious process, involving irrationality and any kind of “unexplainable” feelings and emotions. By rejecting that point of view, the recent epistemological and cognitive studies concentrate on the concept of abduction, as a rational means to originate and refine new ideas. I maintain that abduction is the process of “inferring” certain facts and/or laws and hypotheses that render some sentences plausible, that “explain” or “discover” some (eventually new) phenomenon or observation; it is the process of reasoning in which explanatory hypotheses are formed and evaluated. In (Magnani, 2001a) I have introduced the concept of theoretical abduction, as a form of internal processing. There are two kinds of theoretical abduction, “sentential”, related to logic and to verbal/symbolic inferences, and “model-based”, related to the exploitation of internalized models of diagrams, pictures, etc. Moreover, I have described two main epistemological meanings: 1) abduction that only generates “plausible” hypotheses (“selective” or “creative”), and 2) abduction considered as inference “to the best explanation”, which also evaluates hypotheses. Traditional cognitive science and computational accounts concerning abduction aim at illustrating discovery and creativity processes in terms of “theoretical” and “internal” aspects. A neglected issue, worth of a deepest investigation inside artificial intelligence and cognitive science, is that “discovery” and hypotheses generation are often related to a complex cognitive task involving the use and the manipulation of the external world. Theoretical abduction certainly illustrates and cognitively integrates much of what is important in creative reasoning in science, in humans and in computational programs, but fails to account for many cases of explanations occurring in science when the exploitation of environment is crucial. The concept of manipulative abduction (1) aims at capturing a large part of scientist’s thinking where the role of action is central, and where the features of this action are implicit and hard to be elicited: action can provide otherwise unavailable information that enables the agent to solve problems by starting and by performing a suitable abductive process of generation or selection of hypotheses. Concrete manipulations of the external world and of external representations constitute a fundamental passage in scientific discovery: by a process of manipulative abduction it is possible to build prostheses (epistemic mediators) for human minds, by interacting with external artifacts and representations in a constructive way. In this manner it is possible to create “implicit” knowledge through doing and to produce various opportunity to find, for example, anomalies and fruitful new risky perspectives. As a kind of model-based reasoning (Magnani & Nersessian, 2002, Magnani, Nersessian, & Pizzi, 2002) and of embodied and unexpressed knowledge manipulative abduction holds a key role in the processes of scientific comprehension and discovery. I have called “templates” some prototypical manipulative behaviors which account for the most common cognitive and epistemic acting related to scientific discovery.|../mac/prof22.html",
    "Gender Difference in Effects of Conflict on Cognitive Change...||Paper Track Gender Difference in Effects of Conflict on Cognitive Change Eiji Tomida , Department of Behavioral and Health Sciences, Graduate School of Human-Environment Studies, Kyushu University e.t.edu@mbox.nc.kyushu-u.ac.jp Shunichi Maruno , Department of Human Sciences, Faculty of Human-Environment Studies, Kyushu University syunedu@mbox.nc.kyushu-u.ac.jp The aim of this study was to examine whether gender difference mediates the effects of conflict on cognitive change through social interaction. Forty three undergraduate participants were divided into 10 groups and asked to jointly construct a naïve path model explaining the cause of Japanese teenager’s aggression through discussion. They were also asked to personally construct a similar naïve model before and after the discussion sessions. These personal models were compared within individuals and individually assessed the degree of change in the explanation. Coded the transcriptions of all discussion sessions, we counted the frequencies of utterance, for example, counter-arguing, interpreting, and agreeing etc. Examined the relationships between frequencies of being uttered by others and degree of change in explanation, we found that degree of change in explanation was correlated with frequency of being interpreted, but not with of being counter-argued. However, only in case of counterargument, male and female samples showed different types of relationship with degree of change in explanation. Specifically, whereas females had a negative relationship, males had a positive one. This result indicates that we need to examine the effect of personal backgrounds affecting the interpretation of specific utterances to fully explicate the mechanisms of cognitive change through social interaction.|../mac/prof220.html",
    "Interaction Organization in Graphical Communication...||Paper Track Interaction Organization in Graphical Communication Ichiro Umata , ATR Media Information Science Laboratories umata@atr.co.jp Atsushi Shimojima , Japan Advanced Institute of Science and Technology ashimoji@jaist.ac.jp Yasuhiro Katagiri , ATR Media Information Science Laboratories katagiri@atr.co.jp Nik Swoboda , ATR Media Information Science Laboratories nswoboda@atr.co.jp In this paper, we examine interaction organizations in communication through drawing. We report that both sequential and parallel interaction organizations take place in actual human graphical communication, despite the fact that the media use involved in drawing has resource characteristics that better match parallel organizations. We argue that activity-dependent constraints imposed by the task performed in the interaction play a significant role in determining the interaction organization. We propose two types of sequentiality constraints and demonstrate that they can override the resource characteristics of the drawing media, thereby enforcing a sequential turntaking organization similar to those observed in verbal interactions.|../mac/prof221.html",
    "Two Reasoning Mechanisms for Solving the Conditional ‘Fallacies’....||Paper Track Two Reasoning Mechanisms for Solving the Conditional ‘Fallacies’. Niki Verschueren , Laboratory of Experimental Psychology University of Leuven Niki.Verschueren@psy.kuleuven.ac.be Walter Schaeken , Laboratory of Experimental Psychology University of Leuven Walter.Schaeken@psy.kuleuven.ac.be Géry d’Ydewalle , Laboratory of Experimental Psychology University of Leuven Géry.Dydewalle@psy.kuleuven.ac.be There are two different reasoning mechanisms for solving ‘ifthen’- problems: one is based on likelihood-estimates and is rather heuristic; the other one takes counterexamples into account and is analytic in nature. Based on the difference in input of the two reasoning mechanisms we found that the AC problem is mainly solved by using likelihood-information, while the DA problem is rather solved using counterexampleinformation. Mental models adepts have proposed some explanations to account for the differences in processing difficulty and speed between AC and DA. Considering the reasoning mechanism for AC and DA from a dual process perspective provides an extra explanation for the observed effects. This study indicates that framing observations in a dual process account can provide additional explanations for well-known phenomena.|../mac/prof222.html",
    "Assessment of Resource Coordination Effectiveness Through Analysis of Distr...||Paper Track Assessment of Resource Coordination Effectiveness Through Analysis of Distributed Cognitive Traces in Team Decision Making Rita M. Vick , CIS Interdisciplinary Doctoral Program, University of Hawaii vick@hawaii.edu Martha E. Crosby , Information and Computer Sciences Department, University of Hawaii crosby@hawaii.edu Brent Auernheimer , Computer Science Department, California State University brent@CSUFresno.edu The results of the research undertaken for this study will augment current knowledge of how complex computer supported collaborative work (CSCW) and computer supported collaborative learning (CSCL) systems are used in the contexts of work and learning. The work has significant implications for effective design, deployment, and use of collaborative systems. The group task chosen for this study is decision making, a task that is inherent in almost every kind of work and learning activity. The purpose of the study is to capture evidence of artifact usage, emergence of knowledge and process structures, and resource coordination patterns during group interaction process in order to determine how resource coordination during group interaction facilitates the decision process. It is proposed that interactive use of emergent as well as stored cognitive information traces in the work environment results in the formation of an emergent group mental model through a discourse-driven distillation process. The group mental model is the distributed shared understanding of the team’s problem solution instantiated in the final decision model as the physical representation of the problem solution.|../mac/prof223.html",
    "Two Apparent “Counterexamples” To Marcus: A Closer Look...||Paper Track Two Apparent “Counterexamples” To Marcus: A Closer Look Marius Vilcu , School of Computing Science, Simon Fraser University mvilcu@cs.sfu.ca Robert F. Hadley , School of Computing Science, Simon Fraser University hadley@cs.sfu.ca Marcus, Vijayan, Bandi Rao, Vishton’s experiment (1999) concerning infant ability to discriminate between simple syntactic structures has prompted many connectionists to strive to demonstrate that certain types of neural networks can replicate those results. In this paper we take a closer look at two such attempts: Shultz & Bale (2001) and Altmann & Dienes (1999). We were not only interested in how well these two models matched the infants’ reported results, but also whether or not they were able to learn the grammars involved in this process. After performing an extensive set of experiments, we found that, at first blush, Shultz & Bale’s model replicated the infant’s known data, but the model largely failed to learn the grammars. We also discovered serious problems with Altmann & Dienes’ model, which failed to match most of the infant’s results and to learn the syntactic structure of the input patterns.|../mac/prof224.html",
    "Effects of Sequential Context on Judgments and Decisions in Prisoner’s Dile...||Paper Track Effects of Sequential Context on Judgments and Decisions in Prisoner’s Dilemma Game Ivaylo Vlaev , Department of Experimental Psychology, University of Oxford ivaylo.vlaev@psy.ox.ac.uk Nick Chater , Department of Psychology, University of Warwick nick.chater@warwick.ac.uk Existing models of interactive game-theoretic decision making typically assume that only the attributes of the game need be considered when reaching a decision, i.e., these theories assume that the utility of a strategy is determined by the utility of the outcomes of the game, and transforms of the probabilities of each outcome. The strategic decisions are assumed to be based on these utilities. The two experiments presented here provide strong evidence against these assumptions. We investigated choice and predictions about the choices of other players in Prisoners Dilemma game. The cooperativeness of the games in each condition was varied and the results demonstrate that the average cooperation rate and the predicted cooperation of the other player in each game strongly depended on the cooperativeness of the preceding games, which suggests that games are not considered independently. It is proposed that people have poor notions of absolute cooperativeness, risk, and utility, and instead make their judgments and decisions in relative terms. The proposed accounts for these results are based on existing psychophysical and cognitive theories of perception and judgment of magnitude information.|../mac/prof225.html",
    "I’ll Never Grow Up: Adult and Child Understanding of Aspect...||Paper Track I’ll Never Grow Up: Adult and Child Understanding of Aspect Laura Wagner , Department of Psychology, Harvard lwagner@wjh.harvard.edu This paper revisits the aspectual under-extension found in children’s production data, in which children preferentially link telic predicates with perfective/past morphology and atelic predicates with imperfective/present morphology. I argue that these aspectual groupings reflect a deep property of linguistic/conceptual organization and are manifested in various ways throughout the lifespan. The results of a new sentence comparison task show that adults judge sentences which conform to the children’s under-extended groupings as better than those which do not.|../mac/prof226.html",
    "How People Represent and Reason from Graphs...||Paper Track How People Represent and Reason from Graphs Lara Webber , Department of Psychology, University of Durham l.j.webber@durham.ac.uk Aidan Feeney , Department of Psychology, University of Durham aidan.feeney@durham.ac.uk In this paper we examine how people represent graphical information. We present a constrained graphical reasoning task isomorphic in logical structure to a three-term series reasoning problem. Participants were shown pairs of simple line graphs (premise graphs) and were then required to verify a third line graph (conclusion graph). We found that participants reordered the premise graphs in order to construct integrated representations. The order of the terms in the premises (their figure) modulated the accuracy and speed with which participants subsequently verified conclusions against these representations. These findings suggest a role for analogical representation in graph comprehension and call into question the common assumption that graph comprehension processes may accurately be modelled using propositional representations only.|../mac/prof227.html",
    "Reasoning from Causal and Noncausal Conditionals: Reasoning from Causal an...||Paper Track Reasoning from Causal and Noncausal Conditionals: Reasoning from Causal and Noncausal Conditionals:Testing an Integrated Framework Andrea Weidenfeld , Allgemeine Psychologie I, University of Potsdam aweiden@rz.uni-potsdam.de Klaus Oberauer , Allgemeine Psychologie I, University of Potsdam ko@rz.uni-potsdam.de We suggest and test an integrated framework explaining how the interpretation of and the reasoning from causal conditionals (e.g., If you fertilize a flower it will bloom) depends on exceptions. In the model availability of exceptional situations (e.g., the flower was not watered enough) reduces the subjective conditional probability of the consequent given the antecedent, P(q p). The conditional probability corresponds to the subjective degree of belief in the conditional, P(p®q). The degree of belief in the conditional affects the willingness to accept the valid inferences modus ponens and modus tollens. Additionally to this probabilistic path the framework contains a mental model path: n direct influence of exceptional situations on the willingness to accept modus ponens and modus tollens. Three internetbased experiments supported the framework for causal but not for arbitrary conditional statements in which no meaningful relation between antecedent and consequent was present.|../mac/prof228.html",
    "Speeded categorization: the effects of perceptual processing and decision-m...||Paper Track Speeded categorization: the effects of perceptual processing and decision-making time. A.K. Welham , School of Psychology, University of Exeter a.k.welham@ex.ac.uk M.J. Schnadt , School of Psychology, University of Cambridge mschnadt@onetel.net.uk A.J. Wills , School of Psychology, University of Exeter a.j.wills@ex.ac.uk The effects of limited processing time were investigated for the binary categorization of artificial multidimensional objects. Following Lamberts and Freeman (1999), in the first stage of the experiment participants learnt to categorize 9 stimuli into two categories. In the second stage, the same stimuli were presented for categorization, and both the display time, and the time available in which to make a decision, were varied independently. It was found that each of these variables had a significant effect on accuracy of categorisation, as well as response latency. Lamberts and Freeman (1999) demonstrated that restricting presentation time of a certain stimulus in their category structure caused a reversal in category assignment. We found evidence of the same reversal, but it was dependent on the time available to make a decision rather than the duration of stimulus display. Importantly, changes in accuracy due to response deadline were not explicable in terms of truncation of processing by the limited time. The study provides an empirical investigation of the intuitive notion that both perceptual processing and decision making components are time dependent.|../mac/prof229.html",
    "iSTART: An Automated Reading Strategy Tutor...||Publication-Based Talk Track iSTART: An Automated Reading Strategy Tutor Danielle S. McNamara , University of Memphis Irwin Levinstein , Old Dominion University Keith Millis , Northern Illinois University Joe Magliano , Northern Illinois University Katja Wiemer-Hastings , Northern Illinois University This talk will describe iSTART (Interactive Strategy Trainer for Active Reading and Thinking), a reading strategy tutor that we have developed in which animated agents teach students to self-explain text using a variety of active reading strategies. iSTART was developed to help students learn how to more actively process less cohesive, challenging text, particularly science text. The trainer is based on previous experiments which demonstrated that self-explanation coupled with reading strategy training increased comprehension scores and course grades. This training has been particularly effective for low-knowledge readers. This talk will (a) describe the iSTART system and the theoretical motivation for the various components, (b) briefly describe two experiments which have been conducted with college and middle-school students to test the system, and (c) describe the success of several approaches that we have used to analyze the verbal protocols and provide feedback to the user. The system: iSTART delivers reading strategy training using an interactive and adaptive format. Pedagogical agents interact with each other and with the user to increase active processing and participation by the student. The student first learns about self-explanation and reading strategies (comprehension monitoring, paraphrasing, predictions, elaborations, and bridging inferences). The student then practices self-explanation by typing in explanations to sentences from a science text. The system analyzes the self explanations and provides feedback to the user.|../mac/prof23.html",
    "Acquisition of concepts and causal rules in SHRUTI...||Paper Track Acquisition of concepts and causal rules in SHRUTI Carter Wendelken , International Computer Science Institute carterw@icsi.berkeley.edu Lokendra Shastri , International Computer Science Institute shastri@icsi.berkeley.edu The SHRUTI model demonstrates how complex cognitive functions can be realized by neural circuitry. In this paper, we address how key elements of this circuitry can be learned in a neurally plausible manner. Two basic learningmechanisms, causal Hebbian learning and recruitment learning, are employed in the learning of relational concepts and causal rules.|../mac/prof230.html",
    "Progression of one Student Group’s Work...||Paper Track Progression of one Student Group’s Work Muffie Wiebe , School of Education, Cognition and Learning Lab, Stanford wiebe@stanford.edu This paper argues that the construction of classroom practice is the learning which takes place, and that it is jointly accomplished by the teacher and students through the ongoing organization of student work. A case of classroom interaction is examined in which the student learning is shown to be an accomplishment at the level of the group. This learning is evidenced through a 4-step progression, in which the student group shifts to more expert and more autonomous activity and the teacher is shown shifting from less explicit whole-class instruction to more tailored group interventions. A three-level analytic framework is presented and two examples are offered to demonstrate its utility in explaining the development of this progression.|../mac/prof231.html",
    "Thematic Integration in the Similarity of Abstract and Concrete Items...||Paper Track Thematic Integration in the Similarity of Abstract and Concrete Items Katja Wiemer-Hastings , Department of Psychology, Northern Illinois University KATJA@NIU.EDU Xu Xu , Department of Psychology, Northern Illinois University XXU@NIU.EDU Thematic integration plays a role in similarity judgments of pairs of items that are unrelated taxonomically, like soup and spoon. We hypothesized that integration serves as a more central process in the similarity evaluation of abstract items because of their temporality, their large variability, and relational nature. Participants rated the similarity of concrete or abstract word pairs, followed by a written explanation of the ratings. Explanations were coded as thematic integration and taxonomic comparison. Consistent with our hypothesis, the rate of integration responses was higher overall for abstract items, and integration of abstract items occurred frequently even for item pairs for which taxonomic comparison was possible, unlike for concrete nouns. and Ba|../mac/prof232.html",
    "When two heads are better than one expert...||Paper Track When two heads are better than one expert Jennifer Wiley , Department of Psychology, University of Illinois at Chicago jwiley@uic.edu Cara Jolly , Department of Psychology, University of Illinois at Chicago cjolly1@uic.edu In a line of research on expertise and creative problem solving (Wiley, 1998; Wiley, 1999), I have found an instance in which two heads do seem better than one. In a study using a Remote Associates Task, when high knowledge participants were paired with a novice, solution rates actually increased beyond what would be expected from both individuals working independently. The discovery of a “process gain” condition is quite rare within the social psychology and group problem solving literature. Further, from a cognitive standpoint, it is interesting that experts may sometimes need the assistance of novices in order to be most effective, flexible or innovative in their problem solving. That is, when others state their ideas it may cause an interruption in the idea generation process and cause an individual to lose their chain of thought. Working with others adds information, but also adds new channels that need to be attended to. Further, more time may be spent on off-task topics (Dugosh, Paulus, Roland & Yang, 2000). In the end, individuals may be more burdened and enjoy less intact cognitive processing working in groups than when working alone. Studies have directly tested this notion of interrupted cognitive processing among groups, which has|../mac/prof233.html",
    "Effects of language on color discriminability ...||Paper Track Effects of language on color discriminability Nathan Witthoft , MIT witthoft@mit.edu Jonathan Winawer , MIT winawer@mit.edu Lisa Wu , MIT skuld@mit.edu Michael Frank , Department of Psychology, Stanford mcfrank@stanford.edu Alex Wade , Department of Psychology, Stanford wade@white.stanford.edu Lera Boroditsky , MIT lera@mit.edu Different languages divide the color spectrum in different ways. Can such linguistic codes affect color discrimination? Results of three experiments suggest that color language can influence people?s color judgments even in conditions when all color stimuli are present at the same time and need not be stored in memory. Two experiments showed that color- discrimination performance within a language group is affected by verbal interference (and not spatial interference). The third experiment showed that color discrimination performance across a boundary that exists in one language but not another can be altered by linguistic interference only for the language group that codes that linguistic distinction.|../mac/prof234.html",
    "Structural Differences of Physical and Mental Events and Processes...||Paper Track Structural Differences of Physical and Mental Events and Processes Xu Xu , Department of Psychology, Northern Illinois University XXU@NIU.EDU Katja Wiemer-Hastings , Department of Psychology, Northern Illinois University KATJA@NIU.EDU Event structure has been intensively studied through different approaches. A recent study by Rips and Estin (1998) applied a property listing approach and documented structural differences between physical and mental events. Physical events were shown to have more distinct parts than mental events, which were described as more homogeneous. The present study replicated this study with one modification. We added a factor that may have been confounded with physicality in their study. This factor further distinguishes events and processes. We examined whether the observed structural differences between physical and mental events was partly due to this factor. The results show that distinctiveness of parts was a function not only of tangibility (physical or mental event), but also of temporal characteristics (event vs. process). Furthermore, the distinction of process and event was the only significant factor when distinctive properties were weighted against common properties.|../mac/prof235.html",
    "Dual Processes in the Acquisition of Categorical Concepts...||Paper Track Dual Processes in the Acquisition of Categorical Concepts Takashi Yamauchi , Department of Psychology, Texas A&M University tya@psyc.tamu.edu Two experiments and computational simulations investigated the way people make classifications and inferences when the information about category membership was available to participants. On a classification question, participants were asked to predict the category to which a stimulus belongs, and on an inference question, participants were asked to predict the feature value of a stimulus given the category membership of the stimulus. The results from this study indicate that classification and inference rely on markedly different cognitive processes. While classification judgments are more likely to employ a similarity-based associative process, inference judgments appear to be carried out primarily by a rule-based abstract process. Given classification questions, participants were highly liable to modifications introduced to the concrete appearance of individual stimuli, but the same manipulations barely influenced inference judgments.|../mac/prof236.html",
    "A theory of rerepresentation in analogical matching...||Paper Track A theory of rerepresentation in analogical matching Jin Yan , Qualitative Reasoning Group, Northwestern University j-yan@northwestern.edu Kenneth D. Forbus , Qualitative Reasoning Group, Northwestern University forbus@northwestern.edu Dedre Gentner , Psychology Department, Northwestern University gentner@northwestern.edu Psychologically, rerepresentation appears to be an important technique for achieving flexibility in analogical matching. This paper presents a concise theory of rerepresentation in analogical matching. It divides the problem into detecting opportunities for rerepresentation, generating rerepresentation suggestions based on libraries of general methods, and strategies for controlling the rerepresentation process. We show that the kinds of opportunities can be exhaustively derived from the principles of structuremapping, and the methods for detecting them derived from consideration of how the SME algorithm works. Four families of rerepresentation methods are proposed, as well as task-independent and task-dependent constraints on strategies. Implemented simulation examples are used for illustration.|../mac/prof237.html",
    "Conjunctions are easier than disjunctions:A study of logical reasoning prob...||Paper Track Conjunctions are easier than disjunctions:A study of logical reasoning problems in the GRE Yingrui Yang , Department of Cognitive Sciences, Department of Decision Sciences and Engineering Systems, Rensselaer Polytechnic Institute yangyri@rpi.edu P.N. Johnson-Laird , Department of Psychology, Princeton University phil@princeton.edu We report three experiments corroborating a prediction of the theory of mental models about reasoning. In realistic problems deriving from those used in the Graduate Record Examination, logically untrained individuals are able to cope better with conditional assertions, which have only a single explicit mental model, than with disjunctive assertions, which have multiple explicit mental models. The experiments showed that a manipulation of a sentence in the text of the problems have only a marginal effect, whereas a manipulation of the response options – whether they were both conditionals or both disjunctions – had robust effects both on the accuracy of performance and the latency of response.|../mac/prof238.html",
    "Mental metalogic and its empirical justifications: The case of reasoning wi...||Paper Track Mental metalogic and its empirical justifications: The case of reasoning with quantifiers and predicates Yingrui Yang , Department of Cognitive Science; Department of Decision Sciences and Engineering Systems, Rensselaer Polytechnic Institute, yangyri@rpi.edu Selmer Bringsjord , Department of Cognitive Science; Department of Computer Science; Department of Decision Sciences and Engineering Systems, Rensselaer Polytechnic Institute, selmer@rpi.edu We report on a series of experiments designed to test a new, unifying theory in psychology of deductive reasoning: mental metalogic (MML, for short; Yang & Bringsjord, 2001), which marks the unification of mental logic (ML) and mental model (MM) theory, and stands in part on the strength of empirical investigation of strategic interactions between logical syntactic and logical semantic processes in human reasoning. MML promises to resolve the long-standing controversy between ML and MM.|../mac/prof239.html",
    "Explanatory Content and Multi-Turn Dialogues in Tutoring...||Publication-Based Talk Track Explanatory Content and Multi-Turn Dialogues in Tutoring Stellan Ohlsson , Department of Psychology, University of Illinois at Chicago stellan@uic.edu Andrew Corrigan-Halpern , Department of Psychology, University of Illinois at Chicago ahalpe1@uic.edu Barbara Di Eugenio , Department of Computer Science, University of Illinois at Chicago bdieugen@cs.uic.edu Xin Lu X. Lu, Department of Computer Science, University of Illinois at Chicago xlu@cs.uic.edu Michael Glass , Department of Mathematics and Computer Science, Valparaiso University michael.glass@valpo.edu We aim to systematically investigate the effects of explanatory content and multi-turn dialogues on learning during tutoring by implementing alternative tutoring regimens in an intelligent tutoring system for a letter sequence extrapolation task. It is plausible that the power of tutoring resides, in part, in the fact that a tutor can extend right/wrong discourse moves like OK and are you sure with explanatory content (this is right/wrong, because ...). In past work, we have proposed a computational theory of learning from information about wrong answers (Ohlsson, 1996). However, some empirical tutoring studies have found an effect of explanatory content, others have not. The effects of explanatory content might depend on the linguistic devices used to communicate it. For example, backward references in multi-turn dialogues (remember what we said before about ....) are part of normal discourse, and evidence suggest that they add pedagogical power. However, resolving such references requires working memory capacity, so they might interfere with learning. A deeper understanding of the interaction of explanatory content and linguistic form is important for determining the optimal design of natural language interfaces for intelligent tutoring systems (DiEugenio, 2001). We investigate this issue by systematically varying the content and the linguistic form of tutoring explanations, and assessing the effects in controlled experiments. The relevant dimensions of variation are identified via the analysis of tutoring dialogues generated by novice and expert tutors.|../mac/prof24.html",
    "Causal Induction and the Revision of Belief...||Paper Track Causal Induction and the Revision of Belief Daniel G. Yarlett , Department of Psychology, Stanford University yarlett@psych.stanford.edu Michael J.A. Ramscar , Department of Psychology, Stanford University michael@psych.stanford.edu We propose that the process of causal induction can be regarded as a form of belief-revision, and formalize this idea using a discrepancy-based learning algorithm similar to that employed in the Rescorla-Wagner model of associative learning (Rescorla and Wagner, 1972) and the Belief-Adjustment Model (Hogarth and Einhorn, 1992). We then demonstrate that this model can account for con- flicting patterns in human induction judgments reported by Wasserman et al. (1993) and Buehner and Cheng (1997), two data-sets which it is difficult for other models to satisfactorily explain.|../mac/prof240.html",
    "Sound Symbolism and Early Word Learning in Two Languages...||Paper Track Sound Symbolism and Early Word Learning in Two Languages HanakoYoshida , Psychology Department, Indiana University hayoshid@indiana.edu Linda B. Smith , Psychology Department, Indiana University smith4@indiana.edu Verb learning seems especially difficult for children learning many languages, at least relative to nouns. Many have speculated because this is because verbs refer to relational events that are components of much more complex events. Consider a buying-selling event. If the child hears the verb buy , does it mean sell, get, handover, give money, or perhaps smile? By this view, a key problem in learning verb is parsing complex events into relevant components and then attending to the right components. This study presents evidence that parents use sound symbolism to direct children s attention to the proper component and that children readily use that information when learning new verbs. The study specifically compares two languages that differ in their sound-symbolic words Japanese language with its class of mimetics that are heavily used in speech to children and English language which is commonly considered to be a less sound symbolic language. The results indicate that both Japanese- and English-speaking parents use sound symbolism to teach verbs and that both children learning Japanese- and English benefit from this symbolism. This work provides new insight into verb learning, the nature of the input, and the universal aspects of sound symbolism in language use.|../mac/prof241.html",
    "The Role of Embodied Intention in Early Lexical Acquisition...||Marr Prize The Role of Embodied Intention in Early Lexical Acquisition Chen Yu , Department of Computer Science; University of Rochester yu@cs.rochester.edu Dana H. Ballard , Department of Computer Science; University of Rochester dana@cs.rochester.edu Richard N. Aslin , Department of Brain and Cognitive Sciences; University of Rochester aslin@cvs.rochester.edu We examine the influence of inferring interlocutors’ referential intentions from their body movements at the early stage of lexical acquisition. By testing human subjects and comparing their performances in different learning conditions, we find that those embodied intentions facilitate both word discovery and word-meaning association. In light of empirical findings, the main part of this paper presents a computational model that can identify the sound patterns of individual words from continuous speech using non-linguistic contextual information and employ body movements as deictic references to discover word-meaning associations. To our knowledge, this work is the first model of word learning which not only learns lexical items from raw multisensory signals to closely resemble natural environments of infant development, but also explores the computational role of social cognitive skills in lexical acquisition.|../mac/prof242.html",
    "Blending of Non-Similar Episodes as a Result of Analogical Mapping with a T...||Paper Track Blending of Non-Similar Episodes as a Result of Analogical Mapping with a Third One Neda Zareva-Toncheva , Central and East European Center for Cognitive Science, Department of Cognitive Science and Psychology, New Bulgarian University nzareva@cogs.nbu.bg Boicho Kokinov , Central and East European Center for Cognitive Science, Department of Cognitive Science and Psychology, New Bulgarian University bkokinov@nbu.bg This paper is making an attempt to establish a previously unknown phenomenon: blending of dissimilar episodes in long-term memory. In contrast with previous work which demonstrates that similar episodes can be blended, the current experiment shows that even highly dissimilar episodes can be blended if these episodes have simultaneously been mapped onto a third one. In this way analogy-making can produce memory distortions even between episodes that are neither superficially, nor structurally similar. This result confirms a prediction made by the AMBR model of analogy-making.|../mac/prof243.html",
    "What’s that Thing Called Embodiment?...||Paper Track What’s that Thing Called Embodiment? Tom Ziemke , Department of Computer Science, University of Skövde tom@ida.his.se Embodiment has become an important concept in many areas of cognitive science. There are, however, very different notions of exactly what embodiment is and what kind of body is required for what kind of embodied cognition. Hence, while many nowadays would agree that humans are embodied cognizers, there is much less agreement on what kind of artefact could be considered as embodied. This paper identifies and contrasts five different notions of embodiment which can roughly be characterized as (1) structural coupling between agent and environment, (2) historical embodiment as the result of a history of structural coupling, (3) physical embodiment, (4) ‘organismoid’ embodiment, i.e. organism-like bodily form (e.g., humanoid robots), and (5) organismic embodiment of autopoietic, living systems.|../mac/prof244.html",
    "Microlevel analysis constrains models of serial learning...||Member Abstract Track Microlevel analysis constrains models of serial learning Kelly M. Addis , Volen Center for Complex Systems, Brandeis University MS013, Waltham, MA 02454-9110 USA addis@brandeis.edu Michael J. Kahana , Volen Center for Complex Systems, Brandeis University MS013, Waltham, MA 02454-9110 USA kahana@brandeis.edu|../mac/prof245.html",
    "The N400 in Exact and Approximate Mental Arithmetic...||Member Abstract Track The N400 in Exact and Approximate Mental Arithmetic Eric C. Anderson , School of Cognitive Science, Hampshire College ecanderson@hampshire.edu Joanna Morris Florack , School of Cognitive Science, Hampshire College jmorris@hampshire.edu Recent evidence suggests that people may rely on different mental representations and brain regions when completing certain number processing/tasks (Dehaene Spelke, Pinel, Stanescu, and Tsivkin 1999). For example, solving arithmetic problems that require exact calculation is thought to rely on verbal processing while approximate calculation is thought to rely on visuo-spatial processing. Event-related potential research has shown that when participants are presented with multiplication problems followed by incongruent possible answers, an N400 component is generated similar to those elicited in language paradigms when subjects are presented with sentences ending with semantically inappropriate words (Niedeggen, Rosler, and Jost, 1999). This study was designed to combine these two areas of research by trying to elicit an N400 in both exact and approximate addition tasks. If exact arithmetic relies on verbal pathways, it makes sense that a linguistic-like N400 could be generated. However if approximate arithmetic relies on different visio-spatial pathways, it seems likely that approximate addition would not generate an N400. negativity after both foils (near and far) in exact addition but not to foils in approximate addition.|../mac/prof246.html",
    "A Criterion-Specific Advantage for Small Samples in the Detection of Correl...||Member Abstract Track A Criterion-Specific Advantage for Small Samples in the Detection of Correlation Richard B. Anderson , Department of Psychology, Bowling Green State University randers@bgnet.bgsu.edu Michael E. Doherty , Department of Psychology, Bowling Green State University mdoher2@bgnet.bgsu.edu The present work examines the counterintuitive hypothesis that small samples provide better grounds for inferring the existence or non-existence of a population correlation than do larger samples. Researchers have long cited capacity limitation as an explanation for sub-optimal performance (e.g., Miller, 1956; Broadbent, 1958). Yet, recent work (e.g., Kareev, 2000) has challenged the notion that more information is always better—and this challenge takes place in the domain of correlation detection which is, without question, fundamental to learning and cognition. Kareev (e.g., Kareev, 2000) noted that the sampling distribution of the Pearson correlation coefficient is skewed, and that the amount of skew increases as n (the number of elements in each sample) decreases. The top half of Figure 1 illustrates two such distributions (n = 5 and n = 10) sampled from a population with a correlation (?) of .56. Consistent with Kareev’s analyses (e.g., Kareev, 2000), the median and modal correlation (r) in the top half of Figure 1 exceed the value of ?, and the proportion of sample rs exceeding an arbitrary criterion, c (the rightmost dashed line in the top half of the figure) is greater when n = 5 than when n = 10. Thus, there appears to be a small-sample advantage for inferring whether ? = 0 or ? > 0. One feature of Kareev’s work, as well as later work by Juslin and Olsson (2000), is that the decision criterion is used not to decide whether ? = 0 or ? > 0, but to distinguish “useful” correlations from correlations that are too small to be predictively useful (see Kareev, 2000). In contrast to previous research, we built a simulation that used a straightforward means of defining various types of correct and incorrect inferences about ?. Using a signal detection paradigm, we included samples drawn from populations in which ? = 0, as well as from populations in which ? > 0. A false alarm occurred when ? = 0 and when the sample correlation (r) was either greater than an arbitrary decision criterion, c, or less than -c. Likewise, a hit occurred when ? > 0 and when r was either greater than c or less than -c. The criterion was manipulated across five levels (± .5, .6, .7, .8, and .9). Figure 1 shows the actual sampling distributions generated by the simulation, with the dashed lines showing c = ± .8. Performance was measured as the hit rate minus the false alarm rate (D), the components of which are illustrated in Figure 1 (Note that D was computed separately for n = 5 and n = 10). H1, H2, F1, F2, L, and Q denote regions of the sampling distributions, where H1 and H2 are hits (i.e., signal samples falling outside the range of the criteria), and F1 and F2 are false alarms (i.e., noise samples falling outside the range of the criteria). Thus, D = [(H1+H2)/(H1+H2+L)] - [(F1+F2)/(F1+F2+Q)]. Note: An alternate version of the simulation counted Area H2 as misses, and produced results virtually identical to the ones reported here. The results showed that the existence of a small-sample advantage depended on the placement of c: When c was ± .8 or ± .9, there was indeed a small-sample advantage (i.e., D was greater for n = 5 than for n = 10). But when c was ± .5, .6, or .7, there was a large-sample advantage.|../mac/prof247.html",
    "Gender differences in interacting with AnimalWatch...||Member Abstract Track Gender differences in interacting with AnimalWatch Ivon Arroyo , Department of Computer Science, University of Massachusetts ivon@cs.umass.edu Carole R. Beal , Department of Computer Science, University of Massachusetts cbeal@psych.umass.edu Little research has been carried out that examines how the two genders interact with software. This is of special importance in the case of educational software for mathematics, given the difficulties that girls and women have encountered while learning the subject (Eccles et al., 1993). We have observed the interactions of 350 elementary school students with a mathematics intelligent tutoring system, AnimalWatch (Beal et al., 2002). We found gender differences at two levels. At the coarse-grained level, there was a distinct impact of the amount of help on students’ attitudes towards mathematics: girls decreased the value they attributed to math after working with a version of AnimalWatch providing the right answer when students made mistakes. At the fine-grained level, we observed the amount of time students paid attention to help: boys were less willing to be helped. This gender difference was larger for the best students in the class.|../mac/prof248.html",
    "Thinking About Music: Novice and Expert Inductive Reasoning...||Member Abstract Track Thinking About Music: Novice and Expert Inductive Reasoning Liz Baraff , MIT Brain and Cognitive Science Department liz_b@mit.edu John D. Coley , Northeastern University Department of Psychology j.coley@neu.edu Recent research (e.g. López, Atran, Coley, Medin & Smith, 1997; Proffitt, Coley & Medin, 2000; Shafto & Coley, in press) has revealed striking expert -novice differences in category-based induction in the domain of folk biology. In this paper we examine the generality of those findings by investigating expert-novice differences in category-based induction in the domain of music. Experiment 1 revealed that experts and novices showed extremely high agreement in terms of how they sorted the names of 24 musical composers into groups. Experiment 2 employed a standard strength-of-argument rating task to assess the degree to which measures of taxonomic distance derived from Experiment 1 predicted categorybased inferences. Results were precisely as previously reported for folk biology; novices demonstrated effects of both premise-conclusion similarity and premise diversity, where experts showed similarity but not diversity. Experiment 3 replicated Experiment 2 except that expert and novices both rated argument strength under speeded conditions. Under cognitive load, premise-conclusion similarity persisted for both experts and novices. In contrast, under cognitive load novice premise diversity disappeared, whereas for experts diversity was evident only under cognitive load. These results suggest that patterns of reasoning previously reported for folk biological induction may be more generally applicable. They also suggest important processing differences between experts and novices.|../mac/prof249.html",
    "Effects of Dialogue Structure on the Activation of Syntactic Information...||Publication-Based Talk Track Effects of Dialogue Structure on the Activation of Syntactic Information Martin J. Pickering , Department of Psychology, University of Edinburgh martin.pickering@ed.ac.uk Holly P. Branigan , Department of Psychology, University of Edinburgh holly.branigan@ed.ac.uk Janet F. McLean , Department of Psychology, University of Edinburgh janet.mclean@ed.ac.uk Many theories assume that language production involves the activation of linguistic information (e.g., Dell, 1986; Levelt, Roelofs & Meyer, 1999). In such models, prior context can affect processing by altering the relative activation of different elements. Production and self-monitoring of their own utterances, or comprehension of others’ utterances, may all activate speakers’ linguistic representations to some degree. We report three experiments that investigate how prior context can affect syntactic activation in dialogue. We identify three accounts of syntactic activation in production. Each makes different predictions regarding syntactic priming (re-use of particular structures) (Bock, 1986). Under one account, distinct syntactic information is activated in production and comprehension (Bock & Loebell, 1990). This predicts production-to-production priming but not comprehension-to-production priming. Evidence from sentence recall (Potter & Lombardi, 1998) and picture description in dialogue (Branigan, Pickering & Cleland, 2000) argues against this account, but is consistent with a model where production and comprehension activate shared syntactic information in the same way, and to the same extent. This model predicts equivalent comprehensionto- production and production-to-production priming. In a third model, syntactic information is shared, but the degree of activation is not equivalent: production involves both production processes and the comprehension processes implicated in self-monitoring (Postma, 2000), giving rise to stronger activation, whereas comprehension involves the activation of information by comprehension processes only. This model predicts stronger production-to-production than comprehension-to-production priming. All three experiments used a picture description task. In Experiment 1, participants produced picture descriptions after either producing or comprehending another description. Participants produced more Prepositional Object (PO) descriptions like The chef handing the jug to the swimmer after producing or comprehending a PO description, and more Double Object (DO) descriptions like The chef handing the swimmer the jug after a DO description. However, the self-priming effect was stronger than the other-priming effect. This is compatible with a model in which production involves syntactic activation from both production and self-monitoring processes. However, linguistic behavior in dialogues is influenced by interactivity (Fay, Garrod & Carletta, 2000): Could the low effects of comprehension-to-production priming actually reflect low interactivity? (Each participant produced sequences of utterances). To exclude this explanation, Experiments 2 and 3 manipulated degree of interactivity. Speakers produced either alternating utterances or sequences of utterances. Priming was unaffected by this manipulation. Overall, our results support a production-andmonitoring interpretation of Experiment 1. They suggest that prior interactivity plays a relatively reduced role in syntactic processing, or may only exert an influence in dialogues involving no external constraints on turn-taking.|../mac/prof25.html",
    "Strategies in Spatial Navigation...||Member Abstract Track Strategies in Spatial Navigation Olessia Blajenkova , Rutgers University Maria Kozhevnikov , Rutgers University The goal of this research was to investigate the relation between individual differences in strategies used in navigation and mental representations of the environment reflected in sketch-map drawings.|../mac/prof250.html",
    "Cross-linguistic differences in the representations of events: Verb aspect...||Member Abstract Track Cross-linguistic differences in the representations of events: Verb aspect and completion in English and Russian Lera Boroditsky , MIT Eugenia Trusova , MIT|../mac/prof251.html",
    "Interdependence and Past Experience in Menu Choice Assessment...||Member Abstract Track Interdependence and Past Experience in Menu Choice Assessment Duncan P. Brumby , School of Psychology, Cardiff University BrumbyDP@cardiff.ac.uk Andrew Howes , School of Psychology, Cardiff University HowesA@cardiff.ac.uk During goal-directed menu search the user assesses the relevance of an item to their goal and then has a choice between selecting the current item or continuing to assess the remaining items in the choice set. An obvious influence on selection is the relevance of the goal label to the task description (Franzke, 1995). Interestingly, Young (1998) has proposed that there is interdependency between the assessments of each item. Previous models, in contrast, have tended to assume that assessments were independent. A novel prediction to emerge as a consequence of Young’s normalization assumption is that the relevance of the distracter items to the search goal will affect the decision to terminate a search and select an item. More specifically, the assumption predicts that the presence of lower relevance distracters will result in fewer items being assessed. Another potential influence on the decision to select an item or assess further items is the history of previous experience. In problem solving tasks Lovett and Anderson (1996) have shown that an operator is more likely to be selected if it was previously successful. In menu selection it is an open question as to whether previous history of evaluation leads to more or less evaluation. The aim of the present study was to demonstrate that the decision to select an item is not simply determined by the relevance of the goal item to the search goal, but is strongly influenced by the relevance of the remaining distracter items and by past experience.|../mac/prof252.html",
    "Counterfactual ‘only if ’ conditionals...||Member Abstract Track Counterfactual ‘only if ’ conditionals Ruth M.J. Byrne , Psychology Department, University of Dublin rmbyrne@tcd.ie Suzanne M. Egan , Psychology Department, University of Dublin smegan@tcd.ie Juan A. García-Madruga , Developmental and Educational Psychology Department, UNED jmadruga@psi.uned.es People understand a conditional, 'if A then B', such as 'if Peg went swimming then she felt well' by keeping in mind only true possibilities, e.g., A and B, not-A and not-B, not-A and B (Johnson-Laird & Byrne, 2002). Initially they think about only a few of them, e.g., A and B, because of working memory limitations. As a result, they make some inferences readily e.g., modus ponens (MP), given A, therefore B. But they find other inferences difficult, e.g., modus tollens (MT), given not-B, therefore not-A. The information does not correspond to their initial possibility and they must think about other true ones, e.g., not-A and not-B. Counterfactual conditionals, such as 'if Peg had gone swimming then she would have felt well' are different. 'If A had been then B would have been' conveys a conjecture, A and B, but it also conveys a presupposition that the facts are the opposite, not-A and not-B. Reasoners keep in mind two possibilities from the outset. As a result, they make more of the inferences that depend on access to the negative possibility, e.g., MT, and the denial of the antecedent (DA), given not-A, therefore not-B (Byrne & Tasso, 1999). An 'only if' conditional, e.g., 'Peg went swimming only if she felt well' is logically equivalent to an 'if' conditional. ‘If A then B’ and ‘A only if B’ are both false in the same situation, A and not-B. However, their everyday interpretations have long been debated. Reasoners make more MP than MT from ‘if’ but the difference disappears with ‘only if’ (Evans, 1993). We suggest that reasoners keep more possibilities in mind to understand 'only if' then ‘if’ (Johnson-Laird & Byrne, 2002), and they prefer to process them in the direction B to A (Evans, 1993). so for DA (84% versus 63%) although not for MT (71% versus 79%), and there were no differences for the affirmative inferences, as Table 1 shows. As we expected, there were no differences for the negative inferences from counterfactual 'only if', for DA (68% versus 79%), and MT (92% versus 96%), and also no differences for the affirmative inferences. Table 1: The percentages of inference endorsement Inference MP AC MT DA If Factual Counterfactual Only if Factual Counterfactual 100 67 79 63 100 67 71 84 95 84 96 79 98 73 92 68 The results support the suggestion that reasoners keep in mind two possibilities to understand factual and counterfactual ‘only if’. In contrast, they keep in mind a single possibility to understand factual ‘if’ and two possibilities to understand counterfactual ‘if’ (Thompson & Byrne, 2002). Acknowledgments This research was supported by a grant from Enterprise Ireland.|../mac/prof253.html",
    "Reasoning and Belief Revision with Deontic, Causal, and Arbitrary Condition...||Member Abstract Track Reasoning and Belief Revision with Deontic, Causal, and Arbitrary Conditionals Dustin P. Calvillo , Department of Psychology, University of California, Santa Barbara calvillo@psych.ucsb.edu Russell Revlin , Department of Psychology, University of California, Santa Barbara revlin@psych.ucsb.edu Deductive reasoning is the process of drawing or evaluating conclusions from a set of premises. A key finding in the deductive reasoning literature is that the content of the premises affects the conclusions of reasoners. Using the Wason (1966) selection task, researchers have found that participants are more likely to respond in accord with normative logic when reasoning with deontic rules than other types of rules (for review, see Evans, Newstead, & Bryne, 1993). A deontic rule specifies what is permissible or obligatory given some set of circumstances. For example, If someone is drinking beer, then that person must be at least 18 years old (Griggs & Cox, 1982). Some researchers have claimed there is something special about reasoning with deontic rules (Cheng & Holyoak, 1985; Cosmides, 1989), while others have questioned the supporting evidence, claiming the selection task is not a reasoning task at all (Sperber, Girotto, Cara, 1995). Nevertheless, conditional reasoning theories, such as mental models and mental logic theories, have included pragmatic considerations to deal with these content effects (Braine & O’Brien, 1991; Johnson-Laird & Byrne, 2002). The current experiment examines content effects in reasoning and belief revision without using the selection task. We presented 66 undergraduates at the University of California, Santa Barbara, with 16 problems, each containing three tasks in a fixed order. First, in the inference identification task, we presented participants with a conditional premise (if p then q) and a categorical assertion (e.g., p) and asked participants to identify, among multiple alternatives, what, if anything, follows from the premises. The problems’ categorical assertions either affirmed the antecedent (p), denied the antecedent (not p), affirmed the consequent (q), or denied the consequent (not q). Second, in the contradiction recognition task, we presented a third premise that contradicted the inference that could have been drawn in the inference identification task. For example, if given if p then q and p, reasoners should infer q. We then told them that they know for sure that not q is true. The participants’ task was to determine whether all three premises (if p then q; p; not q) were consistent. Finally, in the belief revision task, we asked participants, when they believed there was an inconsistency in the previous task, to resolve the inconsistency by rejecting either the conditional or categorical assertion. We employed a 3x4 mixed-model design. The type of conditional used was a between-subjects variable with three levels: deontic, causal, and arbitrary. The logical structure of the problem was a within-subjects variable with four levels corresponding to which categorical assertion accompanied the conditional. We counterbalanced premise order within the problem set to control for order effects. For the inference identification task, accuracy for problems with deontic rules was greater than problems for causal and arbitrary rules, particularly when the categorical assertion affirmed the consequent or denied the antecedent. Likewise, reasoners were more accurate when reasoning with deontic conditionals in the contradiction recognition task only when the categorical assertion affirmed the consequent or denied the antecedent. For the belief revision task, we predicted that participants would revise their belief in deontic conditionals less often than arbitrary or causal conditionals because a contradiction of a deontic conditional represents a violation of the rule, but it does not disprove it. A contradiction of a causal or arbitrary rule, on the other hand, disproves it since these are empirical generalizations. This hypothesis was not supported; neither of the independent variables had an effect on revision preferences and we found no interaction. Overall, our results provide evidence for the uniqueness of deontic reasoning without using the selection task. Participants responded in accord with normative logic more often with deontic rules than with causal or arbitrary rules, suggesting that logical reasoning is facilitated by deontic rules in a deductive reasoning task, not simply an artifact of the selection task, and theories of conditional reasoning are justified in their inclusion of pragmatic considerations. Acknowledgments We thank Dana Goodgame for collecting data. References Braine, M. D. S., & O’Brien, D. P. (1991). A theory of if: A lexical entry, reasoning program, and pragmatic principles. Psychological Review, 98, 182-203. Cheng, P. W., & Holyoak, K. J. (1985). Pragmatic reasoning schemas. Cognitive Psychology, 17, 391-416. Cosmides, L. (1989). The logic of social exchange: Has natural selection shaped how humans reasoning? Studies with the Wason selection task. Cognition, 31, 187-276. Evans, J. St. B. T., Newstead, S. E., & Byrne, R. M. J. (1993). Human reasoning: The psychology of deduction. Hillsdale, NJ: Lawrence Erlbaum Associates. Griggs, R. A., & Cox, J. R. (1982). The elusive thematicmaterials effect in the Wason selection task. British Journal of Experimental Psychology, 73, 407-420. Johnson-Laird, P. N., & Byrne, R. M. J. (2002). Conditionals: A theory of meaning, pragmatics, and inference. Psychological Review, 109, 646-678. Sperber, D., Cara, F., & Girotto, V. (1995). Relevance theory explains the selection task. Cognition, 57, 31-95. Wason, P. (1966). Reasoning. In B. M. Foss (Ed.), New horizons in psychology. Harmondsworth: Penguin.|../mac/prof254.html",
    "Do we think about music in terms of space? Metaphoric representation of mus...||Member Abstract Track Do we think about music in terms of space? Metaphoric representation of musical pitch. Daniel Casasanto , Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology djc@mit.edu Lera Boroditsky , Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology lera@mit.edu We often talk about musical pitch using spatial language. In English, pitches can be high or low, melody lines can rise or fall, and we can sing at the top or the bottom of our range. Are spatial metaphors for pitch merely linguistic conventions, or is it possible that patterns in language reveal something fundamental about the way we mentally represent pitch? There are several reasons to suspect that pitch and space are importantly related in the brain and mind. Auditory nuclei are organized tonotopically: changes in pitch correspond to analogous changes in the localization of activity on a neural map (Schreiner & Langer, 1997). PET data show that the same regions in right prefrontal cortex are engaged during attention to sounds that vary in musical pitch or spatial location (Zatorre, Mondor & Evans, 1999). Right hemisphere damage that compromises visuospatial memory also impairs the learning of new melodies (Samson & Zatorre, 1991). Linguistic metaphors suggest that the relationship between pitch and space may be highly specific. We tend to borrow unidimensional, vertical spatial terms to describe pitch (e.g., up, down). Do we think about pitch in terms of vertical space? A nonlinguistic psychophysical paradigm, used previously to explore the spatialization of temporal representations (Casasanto & Boroditsky, 2002), was adapted to investigate the spatialization of pitch.|../mac/prof255.html",
    "An ACT-R Model of the Wickens Tracking Task...||Member Abstract Track An ACT-R Model of the Wickens Tracking Task Alex K. Chavez , Department of Computer Science, Drexel University achavez@drexel.edu Dario D. Salvucci , Department of Computer Science, Drexel University salvucci@cs.drexel.edu Our primary goal is to develop a comprehensive model of human multitasking in a cognitive architecture. We are taking a bottom-up approach to this problem, starting by modeling small-scale, albeit complex, dual-task experiments. Using the ACT-R cognitive architecture (Anderson & Lebiere, 1998), we modeled the Martin- Emerson and Wickens (1992) tracking and choice discrimination task (described below). By analyzing our results, we hope to gain insight and understanding into human multitasking at a low, perceptual-motor level. This will serve as a building block in the development of more general models of multitasking currently being worked on.|../mac/prof256.html",
    "Returning to Unsolved Creative Problems...||Member Abstract Track Returning to Unsolved Creative Problems Bo T. Christensen , Learning Research and Development Center, University of Pittsburgh bo@virksom.dk Christian D. Schunn , Learning Research and Development Center, University of Pittsburgh schunn@pitt.edu Incubation refers to the effect in problem solving that spending time away from a creative problem may be beneficial to performance compared to continuous problem solving. The Opportunistic Assimilation theory (Seifert et al., 1995) explains this phenomenon with benefits from cuing from the environment during time away from the problem. The theory states that reaching an impasse on problems will encode ‘failure indices’ in memory that will trigger memory of the problem if the subject later opportunistically encounters information that is relevant to solving the problem. The theory has so far received mixed empirical support. The present research tested hypotheses derived from the Opportunistic Assimilation theory of incubation using insight tasks and an innovative design that allowed for testing analogical retrieval and analogical mapping separately in the same design. In particular, the theory predicts that presenting analogous cues during time away from a problem will 1) trigger immediate analogical retrieval of the relevant problem, and 2) enhance performance (through analogical mapping) on the relevant problem.|../mac/prof257.html",
    "Inhibitory Effects of Aging on Category Learning...||Member Abstract Track Inhibitory Effects of Aging on Category Learning Kristin N. Christy , Department of Psychology, Miami University christkn@muohio.edu Leslie M. Blaha , Department of Psychology, Miami University blahalm@muohio.edu Robin D. Thomas, PhD , Department of Psychology, Miami University thomasrd@muohio.edu While there exists an abundant amount of literature on the effects of aging on memory, priming, and classical conditioning, very little has been said about the effects of aging on category learning. Age-related declines in performance have been consistently found in tasks involving explicit memory, but normal aging has little to no effect on implicit memory operations (e.g. Light & Albertson, 1989). It is easy to assume, then, that category tasks that require the use of explicit memory will take longer for the elderly to learn than implicit memory categorization tasks. A general decrease in neural plasticity as a result of normal aging has been well documented (Woodruff-Pak, 1997). Most neural communication is inhibitory (Kandel et al., 2000), and it has been proposed that this prevalence of inhibition allows necessary functions to be performed without interference. If this were true, it would be likely that the decline of plasticity may have the greatest effect on inhibitory communication, which would lead to difficulties in learning and memory tasks as a result of unnecessary brain activity. Maddox et al. (1998) demonstrated that deficits in inhibitory processes affect attentional capacities of older adults during category learning tasks. In particular, older adults showed poorer performance in selectively attending to single dimensions of integral-dimension stimuli. These results can be attributed to a lack of inhibition, which results in high interference during the learning of the category structures. Instead of focusing on the idea that the elderly are unable to learn as quickly as younger participants, the current study looked at how well elderly participants were able to inhibit their learning. We propose that not only do the attentional processes decline with age, but the inhibition of inappropriate learning declines as well. This lack of negative feedback in the categorization process results in slower learning of an explicit category structure. It is also possible that age-related declines in implicit category learning will be demonstrated due to decreased ability to ignore irrelevant dimensions Using a category learning paradigm, the current study compared elderly participants’ ability to learn explicit category structures to their ability to learn implicit category structures, both of which require attention to more than one dimension. We compared the results of the older adults to those of younger adults who performed the same categorization tasks. A categorization model was created incorporating the decline of attentional processes and the reduction of the influence of negative feedback due to aging. This model was then used to simulate the participants’ responses. Post hoc analysis was performed based on Ashby et al.’s (1998) COVIS model that postulates two modules responsible for category learning, explicit and implicit, where the module predicting the most accurate responses becomes responsible for a person’s output. Modeling techniques were employed to examine the decision bounds used by each participant during categorization. Due to the lesser effect of aging on the implicit module, we expect older adults to use an implicit decision bound on both implicit and explicit tasks more often than younger adults.|../mac/prof258.html",
    "The Use of Pictorial Examples in Problem Solving:Fixation in a Design-Relat...||Member Abstract Track The Use of Pictorial Examples in Problem Solving:Fixation in a Design-Related Task Evangelia G. Chrysikou , Department of Psychology, Temple University lila@temple.edu Robert W. Weisberg , Department of Psychology, Temple University weisberg@temple.edu The notion of fixation refers to an unhelpful reliance on the past during problem solving, when a new perspective is needed (e.g., Scheerer, 1963; Duncker, 1935/1945). Although the concept was introduced in psychology, fixation may occur in many other domains. Recent studies in the field of engineering design, for example, have suggested that presentation of pictorial examples with a tobe- solved problem may lead to fixation in design, even in cases where the example is specifically described as problematic (Purcell & Gero, 1996; Purcell et al., 1993; Jansson & Smith, 1991). The inherently cognitive nature of the design process makes these findings of potential significance to cognitive psychology. The design-fixation studies suggest that fixation occurs when the physical characteristics of the example design overlap with the expertise of the designer. However, none of those studies examined fixation in a group of nonspecialists. Fixation may affect equally designers and naive participants alike. This study examined whether fixation occurs when non-expert individuals are exposed to a laboratory design-problem-solving situation. Our aim was to investigate whether the inclusion of examples would negatively influence performance in a design problemsolving task. According to the example-expertise overlap hypothesis, introduced by Purcell et al. (1993), one would expect negligible fixation with naïve participants.|../mac/prof259.html",
    "An Attentional Constraint on Spatial Meaning...||Publication-Based Talk Track An Attentional Constraint on Spatial Meaning Terry Regier , Department of Psychology, University of Chicago regier@uchicago.edu Mingyu Zheng , Department of Psychology, University of Chicago mzheng@midway.uchicago.edu|../mac/prof26.html",
    "An Action-based Hierarchical Task Analysis for Evaluating Medical Device Sa...||Member Abstract Track An Action-based Hierarchical Task Analysis for Evaluating Medical Device Safety Phillip H. Chung , Department of Psychology, Rice University pchung@rice.edu Jiajie Zhang , School of Health Information Sciences, University of Texas jiajie.zhang@uth.tmc.edu Todd R. Johnson , School of Health Information Sciences, University of Texas todd.r.johnson@uth.tmc.edu Since the release of the Institute of Medicine medical error report in 1999, human errors in medicine and patient safety have become of great concern in the medical field. Device level incidents stemming from poorly designed interfaces are becoming particularly problematic, with advances in device technology. This is not, however, a new problem. The literature has long suggested that the number of injuries resulting from these types of problems far exceed that of injuries due to actual device failures (Cooper, Newbower, Long, & McPeek, 1978). Hence, the motivation for this effort was to apply existing cognitive theory to evaluate the safety of a simple yet pervasive medical device, the volumetric infusion pump. The initial step in our evaluation methodology was to construct Hierarchical Task Analyses (HTA; Kirwan & Ainsworth, 1992) of three typical infusion tasks for six different pump models from three different manufacturers. In order to bring the relevant cognitive mechanisms into consideration, we conducted our analyses following Norman’s Action cycle (Norman, 1986), which describes interaction between human and computer in seven basic stages of user activity (Figure 1). Our most basic prediction from this theory was that interfaces requiring a greater number of internal operations (less information on the interface) and more steps would generate higher frequencies of human error. Compiling our HTAs in a tabular format facilitated counting of the number of tasks, subtasks, external representations, internal representations, error affordances, and basic operators. From these basic measures we were then able to generate general predictions of human error propensity given a pump interface. For all six single and triple channel pumps, the overall trend for the number of affordances, task steps, and internal operations (representations) followed that of the number of actual user problems reported on the FDA’s MAUDE database (FDA, 2003). Initial results from pump tests, currently in progress with human subjects, show a similar trend, although differences not captured by the proposed evaluation methodology were additionally reported (i.e. preference). These findings generally support our proposal and prediction that cognitive theories, such as Norman’s Action theory, hold great value for deductive evaluations of medical device safety.|../mac/prof260.html",
    "An Action-based Task Analysis for Error Prediction in Medical Devices...||Member Abstract Track An Action-based Task Analysis for Error Prediction in Medical Devices Phillip H. Chung , Department of Psychology, Rice University Pchung@Rice.Edu Human Error in Medicine Since the release of the Institute of Medicine medical error report in 1999, human errors in medicine have become of great interest and concern in the medical field. Like other landmark events in the history of human factors, events like this have increased overall awareness of human factors within this domain. Of particular concern are device level incidents stemming from poorly designed interfaces. For some time, literature has suggested that the number of injuries resulting from these types of problems far exceeds that of injuries due to device failures (e.g., Cooper, Newbower, Long, & McPeek, 1978). Volumetric Infusion Pumps One pervasive device in medicine that has led to numerous medical error incidents is the innocuous volumetric infusion pump. From the standpoint of pure research, the volumetric infusion pump provides a benchmark medical device on which the development of human-device error evaluation and intervention techniques may be developed and tested.|../mac/prof261.html",
    "A Language-Specific Form of Attention that Underlies L2-Proficiency...||Member Abstract Track A Language-Specific Form of Attention that Underlies L2-Proficiency Wai Men Noel Chung , Department of Psychology & Centre for the Study of Learning and Performance, Concordia University wmn_chun@alcor.concordia.ca Norman Segalowitz , Department of Psychology & Centre for the Study of Learning and Performance, Concordia University segalow@vax2.concordia.ca This research investigated the role of a specific form of attention underlying second language (L2) proficiency. Some cognitive linguists have proposed that a principal function of language, beyond referring to events, objects and their properties, is to direct attention to relationships between elements in a message (Talmy, 2000). Grammaticized elements (grammatical morphemes, inflections, and word order patterns) refer to categories that are not experienced directly in our perceptual, sensorimotor, and practical dealings with the world (Slobin, 1996, p. 91). Instead, their meanings derive from how they relate various message elements to each other. In The boy was holding a red ball, the grammaticized elements (the/a, was, ing) refer to definiteness, time, and how actions unfold. These meanings are not directly available to perception as are those of boy, hold, red, and ball. Languages vary in the use of grammaticized elements -- a potential challenge for L2 learners. For example, not all languages mark definiteness (the/a) in the same way, if at all. Slobin suggested that in L2-proficiency, grammaticized elements obligatorily (automatically) direct attention to such relationships. We hypothesized that L2 proficiency would correlate positively with efficiency of attention control with L2 grammaticized elements, after taking into account attention with non-grammaticized elements. The grammaticized stimuli were pronouns, prepositions, conjunctions and verb forms. The non-grammaticized stimuli were concrete nouns (e.g, apple) with clear perceptual referents but otherwise unrelated to language structure, and abstract nouns (e.g., hour) naming concepts lacking clear perceptual referents but that otherwise resemble concrete nouns.|../mac/prof262.html",
    "SCoT: A Model of Conversational and Tutorial Intelligence...||Member Abstract Track SCoT: A Model of Conversational and Tutorial Intelligence Brady Clark , Center for the Study of Language and Information, Stanford bzack@csli.stanford.edu Elizabeth Owen Bratt , Center for the Study of Language and Information, Stanford ebratt@csli.stanford.edu Karl Schultz , Center for the Study of Language and Information, Stanford schultzk@csli.stanford.edu Martha W. Evens , Department of Computer Science, Illinois Institute of Technology evens@iit.edu Human tutoring is known to be more e ective than classroom instruction (Bloom 1984). Sophisticated intelligent tutoring systems (ITS) are only about half as e ective as human tutors, however (Graesser et al. 2001). To approach the e ectiveness of human tutors, ITS may need to not only use sophisticated tutoring strategies, but also use natural language di- alogue (Graesser et al. 2001). How should dialogue be modeled as part of an ITS? One approach has been to catalogue the dia- logue moves found in human tutorial dialogues, and realize them as states in finite state automata that govern the dialog move option space for any input. This approach intertwines the mechanisms of dia- logue (e.g., discourse markers, turn management) with the mechanisms of tutoring (e.g., hints, ex- planations). The joint activity theory (Clark 1996) separates conversational intelligence, i.e. how to use dialogue mechanisms in conversation, from the ac- tivity that a dialogue accomplishes. This separation provides for a clearer representation of how and why the structure of a task changes the structure of a dialogue. Joint activities are activities in which participants have to coordinate their individual actions to suc- ceed (e.g., a tutor identifying and addressing a stu- dent’s misconceptions), using language and other signals. The functions of many of these signals are shared across domains (e.g., discourse markers), whereas dialogue structure varies as a consequence of the activity the dialogue serves. This suggests that linguistic knowledge should be kept separate from domain knowledge in an ITS.|../mac/prof263.html",
    "Learning the Role of Intervocalic Constraints in Vowel-to-Vowel Dynamics...||Member Abstract Track Learning the Role of Intervocalic Constraints in Vowel-to-Vowel Dynamics Orlando Bisacchi Coelho , UMC / FEEC & IEL – UNICAMP orlandoc@terra.com.br Edson Françozo , LAFAPE – IEL – UNICAMP edson@iel.unicamp.br Eleonora Albano , LAFAPE – IEL – UNICAMP albano@iel.unicamp.br Renato Basso , LAFAPE – IEL – UNICAMP renatomb@iel.unicamp.br LAFAPE – IEL – UNICAMP|../mac/prof264.html",
    "Spatial Ability in the Representation of Cross Sections...||Member Abstract Track Spatial Ability in the Representation of Cross Sections Cheryl Ann Cohen , Department of Psychology, University of California c_cohen@psych.ucsb.edu Mary Hegarty , Department of Psychology, University of California hegarty@psych.ucsb.edu Madeleine Keehner , Department of Psychology, University of California keehner@psych.ucsb.edu Daniel R. Montello , Department of Geography, University of California montello@geog.ucsb.edu|../mac/prof265.html",
    "Age and Gender: Similarly Cued from Silhouetted Face Profiles...||Member Abstract Track Age and Gender: Similarly Cued from Silhouetted Face Profiles Nicolas Davidenko , Psychology Department, Stanford University ndaviden@psych.stanford.edu Michael Ramscar , Psychology Department, Stanford University michael@psych.stanford.edu Our ability to recognize one another relies heavily on our capacity to extract identifying information from faces. Since most faces share the same basic features and configurations, we must attend to more subtle differences to successfully distinguish among them. Prior to identifying individual faces, we must be able to extract socially relevant information such as gender and age. Wild et al. (2000) showed that people can identify gender from faces in the absence of sex-stereotyped cues. Here we further this claim and show that both gender and age can be determined from information-scarce images of faces. A linear model reveals that the cues we use to estimate age overlap with those we use for gender identification.|../mac/prof266.html",
    "Do typical birds usually fly normally?...||Member Abstract Track Do typical birds usually fly normally? Kristien Dieussaert , Department of Psychology, University of Leuven kristien.dieussaert@psy.kuleuven.ac.be ‘Birds fly, Tweety is a bird, does Tweety fly?’ has become the canonical example to introduce default reasoning. Since its introduction many variations on this example, stretching from ‘Birds normally fly...’ (Lifschitz, 1988) to ‘If something is a bird, then it flies...’, (Schurz, 2002) appeared in the literature. All these sentences are believed to express the same default rule. But do they? In order to do proper experimental research on default reasoning, this is an important question to be answered.|../mac/prof267.html",
    "MP Suppression and Belief Revision, two sides of the same coin?...||Member Abstract Track MP Suppression and Belief Revision, two sides of the same coin? Kristien Dieussaert , Department of Psychology, University of Leuven Kristien.Dieussaert@psy.kuleuven.ac.be Wim De Neys , Department of Psychology, University of Leuven Wim.DeNeys@psy.kuleuven.ac.be MP Suppression. Byrne (1989) showed that the Modus Ponens (MP) inference was suppressed by adding a logically irrelevant, but semantically relevant, premise to the classical MP-problems (if A, then B, A). She presented participants with statements as follows: If it rains, then M. will get wet If she walks outside, then M. will get wet It rains The semantic relevance of the second premise lies in the fact that it triggers a possible exception to the first premise (namely, staying inside). Given A, participants (pps) become uncertain whether the more specific rule applies here or not, resulting in fewer persons accepting the MP conclusion B. Thus, although no direct conflict is apparent and one could easily deny the second conditional premise, the additional information is taken into account as a disabler and creates an inconsistent state that pps wish to resolve. Cummins et al. (1991) showed that the more disablers one can think of, the more suppression of MP takes place. Belief Revision. Elio (1997) showed that when MP was explicitly denied, pps lowered their belief in the conditional premise. She presented pps with statements as follows: Initially, you believe If it rains, then M. will get wet It rains From this, you believe M. gets wet You do further investigation and discover: M. does not get wet In accordance with Cummins et al. (1991), she also showed that the more disablers one could imagine, the lower the degree-of-belief in the conditional.|../mac/prof268.html",
    "Collaboration in Design: Evolving Conceptual Diagrams...||Member Abstract Track Collaboration in Design: Evolving Conceptual Diagrams Fehmi Dogan , College of Architecture, Georgia Institute of Technology fehmi.dogan@arch.gatech.edu Nancy J. Nersessian , College of Computing, Program in Cognitive Science, Georgia Institute of Technology nancyn@cc.gatech.edu This study presents the results of cognitive-historical analysis (Nersessian, 1995) of two case studies from architectural practice: the First Unitarian Church of Rochester by Louis I. Kahn and the New Staatsgalerie by Stirling and Wilford Associates. The investigations trace the evolution of the cognitive system in each case through studying sketches and focusing on one particular kind of external representation: the conceptual diagrams. What we mean by a ‘conceptual diagram’ is an abstract representation that embeds a conceptualization of a design situation (Dogan & Nersessian, 2002). These diagrams are concise, yet powerful in that they provide high-level commitments constraining potential solutions. They embed generic characteristics and convey the form of possible specific design schemes. That they are not detailed prevents early commitment to a specific scheme and, thus, they facilitate exploratory reasoning. At the same time they are not ambiguous in the way other kinds of sketches are in that they fix meaning and constrain a set of related solutions. In the design process of the Unitarian Church, Khan’s conceptual diagram became a socially agreed upon conceptualization facilitating and regulating communication between client and architect. Kahn’s commitment to his form concept specified constraints that simplified the search for a satisfactory design. His individual commitment to the form concept, however, did not block design flexibility. His client’s contributions were made effective by the joint use of Kahn's conceptual diagram. The client appreciated the conceptualization represented in diagram, which provided criteria for assessment of later schemes that could meet their additional constraints. In the case of Kahn, the conceptual diagram was instrumental in collaboration with the client, whereas in the case of Staatsgalerie it became instrumental facilitating the collaboration among peer designers. For the Staatsgalerie project, Stirling designed a satisfying scheme only after re-sketching earlier drawings of junior designers in his office. Stirling's re-sketching helped him to understand the design ideas involved in the earlier schemes and to advance those ideas. The junior designers in the office worked together side by side and they presented their ideas through sketches to Stirling. This was a collaboration conducted largely through sketches. According to the historical records, this significant collaborative effort evolved mainly through sketches with few discussions and conversations. The collaboration evolved in terms of successive stages. First, a wide ranging diagrammatic exercise carried out to establish all possible ways of configuring the building. The alternatives produced at the end of this stage would then be presented to Stirling who would work on these sketches to select, edit, alter, add. He would do this by taking the A4 photocopied clip and putting an A4 tracing paper on top and doodling. Finally, the concept thereby results from a myriad of ideas (Wilford, 1996, p. 14). This was a process of simplification of the complexity through re-sketching to leave only what is deemed necessary and sufficient for achieving a design solution. The process of collaboration in both projects is an informative example of distributed cognition within either a group of professional designers or between a designer and his client. The design collaboration is a problem-solving effort that is sustained in time and space through external representations that are in the form of abstract, simplified drawings, sketches or diagrams. In both cases conceptual diagrams helped the designers represent complex relationships through perceptual characteristics of diagrams. Studies in diagrammatic reasoning have shown that diagrams facilitate reasoning and problem solving by elucidating complex relationships (See, e.g., (Bauer & Johnson-Laird, 1993)). Conceptual diagrams in design simplify the complexity of conceptualizations by embodying the most salient, constraining features of design situations and their relationship to one another. Through these kinds of diagrams, conflicting constraints are resolved through establishing dependency relationships between them or by modifying them. This serves to enhance collaboration.|../mac/prof269.html",
    "Neural Voting Machines...||Publication-Based Talk Track Neural Voting Machines Whitman Richards , Artificial Intelligence Lab MIT wrichards@mit.edu Sebastian Seung , Artificial Intelligence Lab MIT seung@mit.edu In theories of cognition that view the mind as a system of interacting agents, there must be mechanisms for aggregate decision-making, such as voting. Here we show that certain voting procedures studied by social scientists can be implemented as recurrent neural networks. For example, a standard winner-take-all network can determine which of a number of competing alternatives garners a plurality of votes. Similarly, in the special case where voters share a model governing the different rankings of alternatives, the Borda procedure can easily be computed. In the face of voter un-certainties, this Borda network returns the maximum likelihood choice.|../mac/prof27.html",
    "The Nature of the Phonological Code Accessed Early in Visual Word Recogniti...||Member Abstract Track The Nature of the Phonological Code Accessed Early in Visual Word Recognition during Sentence Reading Brianna M. Eiter , State University of New York at Binghamton, Department of Psychology bj92542@binghamton.edu Albrecht W. Inhoff , State University of New York at Binghamton, Department of Psychology inhoff@binghamton.edu There is a great deal of evidence which suggests that readers determine the phonological representation of a word early during visual word recognition (Pollatsek, Lesch, Morris, & Rayner, 1992), and that a word’s phonological code is maintained in working memory during sentence reading (Folk & Morris, 1995). However it is unclear whether the phonological code used for a word’s identification is also the code used in working memory storage. Recently Inhoff, Connine, Eiter, Radach, & Heller (in press) proposed that the phonological code maintained in working memory includes the representation of speech-like properties. If the early-activated phonological code is similar to the code maintained in working memory, and therefore possesses qualities comparable to the phonological code of the spoken word, then providing readers with a spoken word, which is essentially the phonological code of the visual word, should benefit readers by providing a head start for visual word identification.|../mac/prof270.html",
    "“The Artmedicine Education Model (AM-EM) and Observation Method” (AM-OM): t...||Member Abstract Track “The Artmedicine Education Model (AM-EM) and Observation Method” (AM-OM): their Use in Medical Education. Carlos H. Espinel , The Blood Pressure Center, Washington, and Georgetown University Medical Center, Dept. of Medicine espinelmd@aol.com To study the use of “The Artmedicine Education Model” (AM-EM) and “The Artmedicine Observation Method” (AM-OM) in the learning of clinical information by medical students.|../mac/prof271.html",
    "The “Autonoetic Hypothesis” on Creativity:Memory and Cognition in Picasso’s...||Member Abstract Track The “Autonoetic Hypothesis” on Creativity:Memory and Cognition in Picasso’s Guernica. Carlos H. Espinel , The Blood Pressure Center, Washington, and Georgetown University Medical Center, Dept. of Medicine espinelmd@aol.com To study Picasso’s memories in relation to the creation of his masterpiece, Guernica.|../mac/prof272.html",
    "The Influence of Conceptual Relations on Lexical Retrieval...||Member Abstract Track The Influence of Conceptual Relations on Lexical Retrieval Zachary Estes , Department of Psychology, University of Georgia estes@uga.edu Lara L. Jones , Department of Psychology, University of Georgia laraj@uga.edu The purpose of the present investigation is to argue for another source of lexical priming that is independent of associations and similarity, namely, conceptual relations. In a priming study, Estes (2003) found that target combinations (e.g, MOUNTAIN SNAKE) were comprehended faster when the prime used the same relation (e.g., JUNGLE BIRD) than when it used a different relation (e.g., STEREO HEADPHONE). This study shows that conceptual relations play an important role in semantic decisions. We claim that conceptual relations will also facilitate lexical retrieval of target words. Retrieval of a lexical item will activate its associated relations. This activation of relations will then spread to other lexical concepts that complement the prior lexical concept in the activated relation. Moreover, if these conceptual relations are activated automatically, then they should facilitate retrieval even in a lexical task for which the retrieval of conceptual relations is not (logically) necessary. Prior evidence supports this prediction. Gagné and Shoben (1997, Experiment 2) found that lexical decisions were faster for sensical word pairs (e.g., MOUNTAIN BIRD) than for nonsensical word pairs (e.g., PICTURE SOUP ). However, these results may be confounded since their word pairs were not matched for word length, frequency of occurrence, or frequency of co-occurrence (as this was not their empirical interest).|../mac/prof273.html",
    "Decomposition of compound words: an MEG measure of early access to constitu...||Member Abstract Track Decomposition of compound words: an MEG measure of early access to constituents Robert Fiorentino , Department of Linguistics, University of Maryland cnlrob@wam.umd.edu David Poeppel , Department of Linguistics, University of Maryland dpoeppel@deans.umd.edu The psychological reality of the morphological complexity difference between compounds (teacup) and single words (crescent) is highly controversial, and the conditions and time course of morphological decomposition remain contested in the psycholinguistic literature (McQueen & Cutler, 1998). Decomposition is argued to occur early or late (Andrews, 1986), in novel but not lexicalized compounds (Van Jaarsveld & Rattink, 1988), and in long but not short compounds (Bertram & Hyönä, 2003). This study investigates decomposition in compounds using magnetoencephalography (MEG) in a visual lexical decision paradigm comparing compounds (CW) and single words (SW). The results suggest early decomposition, regardless of lexicalization or constituent length. Response time (RT) is sensitive to both early and late processes, whereas the latency of the MEG component at 300-400ms after word onset (M350) indexes early lexical activation but not post-lexical processing (Embick et al. 2001, Pylkkänen et al. 2002). When the first and second CW constituents have higher log frequency (1.96/1.96 vs. 0.455) and shorter letter-length (3.82/4.0 vs. 7.8) than the SW, early decomposition predicts faster RT and earlier M350 than SW, reflecting constituent over whole-word properties. Late decomposition predicts faster RT but not earlier M350. Since the CW are lexicalized and have short constituents, lexicalization and length constraints predict no RT or M350 differences. Whole-word-only accounts likewise predict no RT or M350 differences.|../mac/prof274.html",
    "The Use of Verbal Protocols in the Study of Insight...||Member Abstract Track The Use of Verbal Protocols in the Study of Insight Jessica I. Fleck , Department of Psychology, Temple University jfleck@temple.edu Robert W. Weisberg , Department of Psychology, Temple University weisberg@temple.edu Verbal protocols have been used as data in the study of the cognitive processes underlying many problem-solving tasks (see Ericsson & Simon, 1993, for a review). However, the utility of verbal protocols in the analysis of insight and related processes has been questioned because “verbal overshadowing” was reported in participants solving insight problems (Schooler, Ohlsson, & Brooks, 1993). If this effect applies broadly to insight problems, then the use of verbal protocols in the study of insight would be called into question. However, Fleck and Weisberg (2003) demonstrated that verbal overshadowing did not extend to the Candle Problem. The present research examined the effect of verbalization on three additional insight problems, to examine the potential usefulness of verbal protocols in insight problem solving research.|../mac/prof275.html",
    "Analysis of strategies in expert tutoring dialog for use in Intelligent Tut...||Member Abstract Track Analysis of strategies in expert tutoring dialog for use in Intelligent Tutoring System development Donald R. Franceschetti , Department of Physics, The University of Memphis, Memphis, TN 38152 dfrncsch@memphis.edu Amy B. Adcock , College of Education, The University of Memphis, Memphis, TN 38152 aadcock@memphis.edu Arthur C. Graesser , Department of Psychology, The University of Memphis, Memphis, TN 38152 a-graesser@memphis.edu One-on-one tutoring that encourages students to explain their answers has long been known to be an effective means of increasing student performance, even when the tutors are far from experts in the field concerned (e.g., Chi, de Leeuw, Chiu, and LaVancher, 1994; Bloom, 1984). The design of effective Intelligent Tutoring Systems (ITS) is an area of active research that attempts to take advantage of the benefits of this type of tutoring with the added convenience of automated, just-in-time teaching interventions. Validating ITS dialogues requires comparison with human tutors constrained to conditions similar to those found in ITS interfaces. A basic assumption in the design of ITS is that student productions (questions, statements, and side comments) can be categorized in a way that permits selection of an appropriate tutor response. Advanced ITSs attempt to use Natural Language Processing (NLP) components to give the student an intervention tailored to their specific needs. For these systems to work, a detailed modeling of the conversations that occur during a domain specific tutoring session is desirable. This study addresses two questions posed by the comparison of ITSs to human tutors. The first is the degree of variance that can be expected between expert tutors in a given discipline, in this instance physics. The second is the extent to which the productions of expert tutors vary from one tutor to another and if experience has any impact on the set of dialog moves employed by domain expert tutors. Answers to these questions could be key to the development of a robust ITS. AutoTutor is an ITS that teaches physics by using NLP components to conduct a dialog with the student (Graesser et al., 2000). Students are asked questions in conceptual physics and AutoTutor responds based on the quality of the student response. The overall selection of tutor responses is based on an extensive analysis of the moves employed by nonexpert human tutors across a broad range of subjects (Graesser & Person, 1994). In the process of developing and validating a version of Auto Tutor for conceptual physics, a set of 17 verbatim transcripts of tutoring sessions between students and expert physics tutors were collected. These transcripts represent well over 100 hours of human physics tutoring in a chat room environment. A turn by turn analysis of the transcripts was conducted by an experienced physics professor and a graduate student in educational technology using a modified form of the classification scheme introduced by Graesser and Person (1994). This analysis sheds light on how expert tutors use dialog to elicit deep processing of conceptual physics problems for use in improving intelligent tutoring of physics.|../mac/prof276.html",
    "How do Presentation and Context Influence Representation for Functional Fix...||Member Abstract Track How do Presentation and Context Influence Representation for Functional Fixedness Tasks? Michael C. Frank , Department of Philosophy, Stanford mcfrank@stanford.edu Michael Ramscar , Department of Psychology, Stanford ramscar@psych.stanford.edu A truism in many kinds of problem solving is that if the problem is set up right, the solution is obvious. The trick lies in finding the best way to represent the task. Functional fixedness tasks are tasks modeled around this “trick”—in order to solve them optimally, participants need to use objects in a novel way. In the candle task, first described in Maier (1931), participants must use a box of tacks and a book of matches to mount a candle on a wall so that it can burn normally and without dripping. The optimal solution is to take the tacks out of their box, tack the box to the wall, and place the candle on the box, but participants only discover this solution approximately one quarter of the time. On the other hand, if the tacks are presented next to the box instead of inside it, the task is trivial. The candle task is a functional fixedness task because in order to solve it, participants must overcome their representation of the box as having a fixed function as a container in order to represent it as a possible support for the candle. In a classic study, Glucksberg & Weisberg (1966) revealed that participants who failed to produce the correct solution were subsequently less likely to produce the word “box” in a free association task, leading Glucksberg and Weisberg to conclude that participants failed to overcome their functional fixedness because did not have the “box” concept available to them. In a study, conducted on 68 undergraduate students at Stanford University, we tested the veracity of this view by presenting participants with written descriptions of the candle task and asking them to describe any solution they could find. In all of the conditions, the participants were presented with the word “box” so that the lexical concept was explicitly available in the stimuli. In the control condition, the text of the problem was printed unmodified. In the first experimental condition, the phrases “candle,” “book of matches,” and “box of tacks” were underlined. In the second experimental condition, the words “candle,” “book,” “matches,” “box,” and “tacks” were all underlined. The percentage of solution in the first condition was 23%, whereas the percentages in the other two were 55% and 47%, respectively. T-tests showed there was a significant difference both between the control condition and the second condition (p|../mac/prof277.html",
    "Overjustification of Effort by Pigeons...||Member Abstract Track Overjustification of Effort by Pigeons Andrea M. Friedrich , Department of Psychology, University of Kentucky amfrie0@uky.edu Thomas R. Zentall , Department of Psychology, University of Kentucky zentall@uky.edu When humans prefer outcomes that require greater effort to obtain, such preferences are attributed to cognitive dissonance or overjustification of effort. We demonstrate a similar effect in pigeons and attribute it to a within-trial contrast effect.|../mac/prof278.html",
    "Testing Additional Category Knowledge in a One-Dimension Sorting Task...||Member Abstract Track Testing Additional Category Knowledge in a One-Dimension Sorting Task Gyslain Giguère , Department of Computer Science, Université du Québec à Montréal gyslain.giguere@videotron.ca Guy L. Lacroix , Centre for the Study of Learning and Performance, Concordia University glacroix@education.concordia.ca Serge Larochelle , Department of Psychology, Université de Montréal serge.larochelle@umontreal.ca|../mac/prof279.html",
    "Improving statistical reasoning: representational format and valid intuitio...||Publication-Based Talk Track Improving statistical reasoning: representational format and valid intuitions Peter Sedlmeier , Department of Psychology, Chemnitz University of Technology peter.sedlmeier@phil.tu-chemnitz.de In numerous studies, statistical reasoning has been found to be a notoriously difficult topic for lay people and experts alike, and resulting biases in judgments under uncertainty have been regarded as being very resistant to training attempts. Although this still seems to be a widespread view, there have been moderately successful attempts to make people reason according to probability theory. For instance, Nisbett, Krantz, Jepson, and Kunda (1983) hypothesized that it is mainly the salience of three chance factors – clarity of the sample space and the sampling process, recognition of the role of chance in producing an event, and cultural prescriptions to think statistically in a particular domain – that determine whether people take sample size into account properly. However, even if these chance factors hold, several kinds of sample-size problems still seem to be very difficult to solve (Sedlmeier, 1998). Why? After a thorough review of the literature on people's use of sample size, Sedlmeier and Gigerenzer (1997) advanced an explanation: People have a valid intuition conforming to the empirical law of large numbers. This intuition is, however, only applicable to tasks that involve judgments about single means or proportions and is usually not invoked in judgments about sampling distributions. Valid intuitions are not restricted to judgments about the impact of sample size but cover a large variety of probabilistic judgments (Sedlmeier, 1999; in press a). However, the use of these intuitions depends on the way statistical information is processed: When applicable, intuitions work best if the information is given in a format that is close to the naturally occurring format (e.g., frequencies instead of probabilities) or if, after training, participants are able to translate numerical information into that format (Sedlmeier, 1999). Valid statistical intuitions, (such as the size-confidence intuition, that is, the intuition that means or proportions from larger samples should be given more confidence than means or proportions from smaller samples) can be explained as a by-product of associative learning (Sedlmeier, 1999; 2002; in press b). The hypothesis that the representational format makes a decisive difference in learning to reason statistically was explored in several studies that compared the effectiveness of different kinds of computerized tutoring systems. These systems were built either in a way that conformed to our view or in a way that was held to be nearly equivalent but conformed to more traditional ways of teaching probability theory. For tasks involving conditional probabilities, the impact of sample size, and probability revision, the training versions optimized to elicit valid intuitions led to remarkably better results than traditionally built training programs, especially in respect to long-term learning effects (Sedlmeier, 1997; 1999; 2000; Sedlmeier & Gigerenzer, 2001). As a practical side effect of our research, a textbook on probability theory (including a program on CD) resulted, which is currently being used in several German high schools (Sedlmeier & Köhlers, 2001; Wassner, Martignon & Sedlmeier, 2002).|../mac/prof28.html",
    "Collaborative Model-Building and Peer Critique Online...||Member Abstract Track Collaborative Model-Building and Peer Critique Online Janice Gobert , The Concord Consortium jgobert@concord.org The paper describes a large scale design study involving a total of 1100 middle and high school students from California and Massachusetts who collaborated on-line about plate tectonic activity in their respective location. The students, drawn from demographically diverse schools, collaborated on-line using WISE (Web-based Inquiry Science Environment; Linn, 1999). WISE is a web-based integrated set of software resources to engage students in many types of scientific inquiry, including prompted reflection, electronic discussions, evidence sorting and argument mapping, collaborative search for evidence, collaborative design, and analysis (Linn & Hsi, 2000). Following the WISE design framework, the two main pedagogical principles embodied in the present study were: Make thinking visible and help students learn from one another. In terms of making thinking visible, we engaged students in two visual modes of representation. First, using the drawing tool in WISE, students drew their models and used these models as artifacts for reiterative cycles of critique and model-revision. Secondly, students viewed a set of dynamic, runnable models of plate tectonic phenomena in order to better visualize the dynamic, causal, and temporal processes. In terms of helping students learn from one another, we engaged students in tasks in which they critiqued their learning partners’ models from the opposite coast. We did this to provide students with an opportunity to both think deeply about the domain in order to do the critiques, as well as to promote their understanding of the nature of models in science. Data from 15 classrooms is described and examples of student work are presented. Data analysis was focussed on primarily three areas: 1) measuring overall content gains of the domain, 2) measuring epistemological gains about the nature of models as both representations of causal systems and as communication tools, and 3) characterizing the nature of students’ model critiques and subsequent revisions on the basis of these critiques. Results suggest that the unit and the east-west coast collaboration was successful in promoting deep content learning as measured by the content gains (Gobert et al, 2002a). Additionally, the task of evaluating and critiquing their peers’ models fostered students’ epistemologies of models (Gobert et al, 2002b). Follow up analyses are currently focussed on the ways in which students’ epistemologies of models served as bootstrapping devices for driving model building, deep content understanding, and transfer.|../mac/prof280.html",
    "Modeling Across the Curriculum: Scaling up Modeling using Technology...||Member Abstract Track Modeling Across the Curriculum: Scaling up Modeling using Technology Janice Gobert , The Concord Consortium jgobert@concord.org Paul Horwitz , The Concord Consortium paul@concord.org Bob Tinker , The Concord Consortium bob@concord.org Barbara Buckley , The Concord Consortium bbuckley@concord.org Uri Wilensky , Center for Connected Learning and Computer-Based Modeling, Northwestern University Sharona Levy , Center for Connected Learning and Computer-Based Modeling, Northwestern University Chris Dede , Graduate School of Education, Harvard University Modeling environments, we believe, can make the most significant contribution of information technologies to improved science learning. A partnership of the Concord Consortium, Harvard and Northwestern Universities, with the help of the Josiah Quincy Upper School and the Fitchburg and Lowell Public Schools, is studying the impact of modeling on secondary science learning. The project uses tools we call CIE models: environments that incorporate core science content that students learn through inquiry and exploration. Increased learning is expected from a curriculum focused on a few core science concepts that students learn primarily through guided exploration of CIE models. The project goal is to test this expectation in real schools as students encounter a representative set of CIE models over three years in sequential physical science, chemistry, and biology courses. The tools that are used in these courses are Dynamica (Horwitz et al, 2001), Connected Chemistry (Stieff, M & Wilensky, in press) and BioLogica™ (Horwitz & Christie, 1999), respectively. The project uses a powerful tool, Pedagogica, developed at The Concord Consortium that provides online lessons with embedded guidance and assessment. Pedagogica controls all aspects of the learners’ interactions with the software tools by changing the available options, the nature of the scaffolding, the instructions, and the assessments. Pedagogica can automatically report student progress through these lessons via the Internet, providing real-time, fine-grained, educationally significant data on student learning. The project consists of a series of studies using as outcome measures, gains in content knowledge, modeling skills, epistemological understanding, and motivation. The first group of studies focuses on improving the instructional design through individual interviews of students and teachers. The classroom-based studies evaluate the impact of variation in at least two important pedagogical dimensions: degree of constructivism and the type and amount of scaffolding. The design involves randomly assigning the same science content, but different scaffolding, to different students within a class, thereby limiting the problems of bias and self-selection that often confound in-school research. At the same time, Pedagogica will automatically document how students interact with the material and the pedagogy, and track their learning during treatment. We are also conducting a longitudinal study of three-year implementations of project materials in the Partner Schools and, during the latter years, in ten Member Schools. Cumulative gains in the content areas, transfer to other topics, modeling skills, epistemological understanding, and attitudes about science are being measured. All data are collected from Pedagogica, annual administration of a project-generated assessment, extensive interviews, and standardized tests. With this, we are addressing how this can be adapted in order to scale up to serve large numbers of students across the country.|../mac/prof281.html",
    "Conventionality, Similarity, and the Metaphor / Simile Distinction...||Member Abstract Track Conventionality, Similarity, and the Metaphor / Simile Distinction Selin Gokcesu , Department of Psychology, Indiana University bgokcesu@indiana.edu Brian F. Bowdle , Department of Psychology, Indiana University bbowdle@indiana.edu Nominal metaphors (figurative statements of the form X is Y) can often be restated as similes (figurative statements of the form X is like Y). Thus, one can say both Time is a river and Time is like a river. This grammatical alternation is often assumed to be arbitrary, with metaphors and their simile counterparts being synonymous. (Indeed, metaphors are traditionally defined as elliptical similes.) However, a handful of studies have suggested that the two forms may in fact not be equivalent. For example, Gibb and Wales (1990) found that similes are preferred over metaphors for figurative statements with concrete predicates, and Reynolds and Ortony (1980) found that simile comprehension developmentally precedes metaphor comprehension. What accounts for these differences between the two forms? Two recent proposals have been made. According to Gentner and Bowdle’s (2001) career of metaphor hypothesis, conventionality plays a key role. On this view, novel figurative statements are processed as comparisons between the target concept (a-term) and the base concept (bterm). Therefore, similes will be strongly preferred over metaphors for novel statements, as the grammatical form of similes suggests comparison, whereas the grammatical form of metaphors suggests categorization. As metaphoric base terms become conventionalized, however, they take on additional, more abstract meanings that can act as metaphoric categories. Thus, conventional figurative statements can be processed either as comparisons involving the original sense of the base term, or as categorizations involving the derived sense of the base term, which means that they can be phrased either as similes or as metaphors. Aisenman (1999) has proposed an alternative account. According to Aisenman, the distinction between metaphors and similes is closely tied to the type of similarity expressed by the figurative statement. If the similarity between the target and base is primarily attributional – that is, if both items are perceptually alike – then the simile form will be preferred. If, on the other hand, the similarity between the target and base is primarily relational – that is, if both items are functionally alike – then the metaphor form will be preferred. The basic assumption here is that function tends to be more conceptually central than physical appearance. Thus, relational similarity should call for the “stronger” grammatical form: namely, the metaphor We tested these two alternative accounts of the metaphor / simile distinction by presenting subjects with both novel and conventional figurative statements whose target and base terms were either attributionally similar or relationally similar. In one experiment, subjects received all four types of statements (novel + attributional, novel + relational, conventional + attributional, conventional + relational) as both metaphors and similes, and were asked to indicate which form they preferred. A main effect of conventionality was obtained, with the simile form being preferred more strongly for novel statements than for conventional statements. However, there was only a marginal effect of similarity type in the direction predicted by Aisenman’s account. In a second experiment, subjects were presented with the same statements, each statement being phrased either as a metaphor or as a simile. Subjects were asked to rate the aptness of the statements. An interaction between conventionality and grammatical form was obtained, such that similes were rated as more apt than metaphors for novel statements, but not for conventional statements. There was no interaction between similarity type and grammatical form. Relational statements were not rated as more apt when phrased as metaphors, and attributional statements were not rated as more apt when phrased as similes. Overall, the results of these experiments support the career of metaphor hypothesis and its approach to the metaphor / simile distinction. Nevertheless, we believe that type of similarity could play a key role during the process of conventionalization. Relational figurative comparisons may be more likely than attributional figurative comparisons to lead to the abstraction of conventional metaphoric categories. Indeed, our research suggests that the majority of conventional metaphor bases are relationally defined.|../mac/prof282.html",
    "The Effect of Speaker-Specific Information in On-Line Pragmatic Inferencing...||Member Abstract Track The Effect of Speaker-Specific Information in On-Line Pragmatic Inferencing Daniel J. Grodner , Department of Cognitive and Linguistic Sciences, Brown University dgrodner@brown.edu Julie C. Sedivy , Department of Cognitive and Linguistic Sciences, Brown University julie_sedivy@brown.edu Sentences frequently convey more information than they explicitly contain. For instance, suppose a speaker uses a noun phrase (NP) modified by an adjective, as in “the tall cup.” In this situation, perceivers infer the existence of two sets of entities in the discourse: (1) a target set corresponding to the literal denotation of the expression (e.g., a tall cup), and (2) a contrast set containing an object matching the noun, but differing by virtue of the property expressed by the adjective (e.g., a short cup). Evidence for the latter inference comes from monitoring perceiver eye-movements as they listen to spoken instructions. In particular, in the presence of a contrasting object, individuals identify target objects faster and make fewer spurious looks to competitors that share the adjectival property (e.g, a tall pitcher) (Sedivy, et al, 1999). These effects can be observed within two hundred ms of the onset of the head noun. Existing evidence suggests that contrastive inferences arise from an understanding between conversational participants that speakers are only as informative as they need to be (Grice, 1975). A simple NP (e.g., “the cup”) would suffice to pick out the intended referent in a context with only a single entity. When a speaker uses a more elaborate form, the perceiver infers that a different state of affairs prevails. The adjective is most easily made informative by attributing to it a distinguishing function. In support of this view, contrastive inferences only arise for adjective types that are not used to label objects in isolation. Scalar and material adjectives are rarely used to label objects in isolation, whereas color adjectives are frequently included in such descriptions. Correspondingly, scalar and material adjectives are interpreted contrastively, but color adjectives are not (Sedivy, 2001). This effect is driven by discourse-level expectations and not the semantic class of the adjective. When color is a highly predictable property of an object (e.g., a yellow banana), color modifiers are rarely used in default labels. In this case, the use of a modifier is seen as overly informative and the modifier is interpreted contrastively.. (Sedivy, in press). Because default descriptions are statically linked with individual referents, the inferences explored in the above studies do not require that perceivers consider the circumstances of the immediate discourse. Perceivers might reflexively infer a contrast upon deviation from a stored default form. In theory, many different factors might be weighed in deciding whether a modified form should be interpreted contrastively. These include the intrinsic properties of a referent, the linguistic context, the reliability of the speaker, the intentions of the speaker, and so on. It is frame that contextual contrast effects have been observed.|../mac/prof283.html",
    "Psychophysiological Studies of Emotional Arousal to Bilingual Speakers' Fir...||Member Abstract Track Psychophysiological Studies of Emotional Arousal to Bilingual Speakers' First and Second Languages Catherine L. Harris , Department of Psychology, Boston charris@bu.edu Second language users have frequently reported that concepts expressed in their first language have greater emotional resonance than concepts expressed in their second language. Prior work used electrodermal recording to verify this psychophysiologically. Late learners of English (Turkish-English bilinguals) rated words and phrases for unpleasantness while electrodermal responsiveness was recorded via finger-tip electrodes (Harris, Aycicegi, & Gleason, 2003). Taboo words and childhood reprimands (e.g., Shame on you!) were compared to neutral words, positive words (bride, joy) and negative words (cancer, death). Skin conductance amplitudes were greater in the second language (L2) than in the first language (L1). While skin conductance responses were overall largest for taboo words, the difference between L1 and L2 was greatest for the reprimands, indicating that hearing Go to your room caused little arousal when such phrases were read or heard in L2. Little research exists on the emotional accompaniments of language. These findings on Turkish late learners of English are a reminder that the knowledge of the cognitive aspects of language is only one aspect of speaking a language. It is well-known that ultimate proficiency in a second language is strongly related to the age of immersion in that language (Birdsong & Molis, 2001; Newport & Johnson, 1989). Does the capacity to have native-levels of emotional arousal depend similarly on age of acquisition? Or do levels of proficiency influence emotional response independently of age of acquisition?|../mac/prof284.html",
    "Effect of Visual-Spatial Ability on Total Performance Score in Endoscopic S...||Member Abstract Track Effect of Visual-Spatial Ability on Total Performance Score in Endoscopic Simulation Training Leif Hedman, (leif.hedman@psy.umu.se; leif.hedman@cfss.ki.se) Pehr Andersson (pehr.andersson@psy.umu.se) Pär Ström (per.strom@cfss.ki.se) Ann Kjellin (ann.kjellin@cfss.ki.se) Lars Särnö (lars.sarna@mta.hs.sll.se) Torsten Wredmark (torsten.wredmark@cfss.ki , Department of Psychology, Umeå University Karolinska Institutet at Huddinge University Hospital Pehr Andersson , Department of Psychology, Umeå University pehr.andersson@psy.umu.se Pär Ström , Karolinska Institutet at Huddinge University Hospital per.strom@cfss.ki.se Ann Kjellin , Karolinska Institutet at Huddinge University Hospital ann.kjellin@cfss.ki.se Lars Särnö , Karolinska Institutet at Huddinge University Hospital lars.sarna@mta.hs.sll.se Torsten Wredmark , Karolinska Institutet at Huddinge University Hospital torsten.wredmark@cfss.ki.se Li Felländer-Tsai , Karolinska Institutet at Huddinge University Hospital li.tsai@cfss.ki.se We report on a study that investigates the relationship between spatial ability tests scores and total score of instrument navigation in Procedicus KSA (simulator for training endoscopic surgery). Wanzel and colleagues (2002) tested whether junior surgical residents with high visual-spatial scores in six tests of visual-spatial ability, would perform better in completing and learning a spatial-complex surgical procedure than those with low scores in the tests. They found that visual-spatial ability, assessed by Mental Rotation test (MRT, Vanderberg and Kuse, 1978), correlates significantly with initial competency and quality of results in two- and four-flap Z-plasty procedures in pig thighs. We hypothesize that students with higher scores in visual-spatial tests would perform better in a KSA instrument navigation procedure than those with low scores in the tests.|../mac/prof285.html",
    "Inferences from Weather Maps: Do you need a Weatherman to know which way th...||Member Abstract Track Inferences from Weather Maps: Do you need a Weatherman to know which way the Wind Blows? Mary Hegarty , Department of Psychology, University of California hegarty@psych.ucsb.edu Matt Canham , Department of Psychology, University of California canham@psych.ucsb.edu Naomi Shimozawa , Department of Psychology, University of California shimozaw@psych.ucsb.edu Although there is a growing body of literature on how people interpret graphical displays, few studies to date have examined how people apply domain knowledge to make inferences from complex graphical displays. Meteorology is an excellent domain in which to investigate these questions, because the displays in this domain can be quite complex and because they are typically used to make predictions about future weather conditions. Weather forecasting depends both on the ability to understand the current weather conditions from a map, and the ability to apply knowledge of meteorological principles to make predictions about how these weather conditions will change in the future.|../mac/prof286.html",
    "Testing the Equality of Learning Rates Using a Linear Hypothesis...||Member Abstract Track Testing the Equality of Learning Rates Using a Linear Hypothesis Sébastien Hélie , Department of Computer Science, Université du Québec À Montréal helie.sebastien@courrier.uqam.ca Denis Cousineau , Department of Psychology, Université de Montréal denis.cousineau@umontreal.ca For over a hundred years, psychologists have tried to describe and understand human learning. While many models disagree on the nature of the process, most agrees that the performance is well described by a power function (Newell and Rosenbloom, 1981). This equation has a curvature parameter as well as two scaling parameters. Recent models are making quantitative predictions on the learning rate, which is represented by the curvature parameter of the function. For example, Logan’s race model (Logan, 1988) predicts that standard deviations and means of response times will decrease at the same rate. The most intuitive way to test this hypothesis is to estimate the best-fitting parameters and to apply a statistical test on those estimations. The problem with this approach is that those estimations are highly biased (Cousineau, Hélie and Lefebvre, in press). In the late fifties, Rao (1959) proposed a test of linear hypothesis. By assuming that the learning rates are equal, the power functions (or any other learning model postulated) become a linear combination of each other, irrespective of the scaling parameters. However, since learning data tends to be noisy, the power of Rao’s test was limited. In order to reduce the effects of noise, Cousineau, Hélie and Lefebvre (in press) proposed to apply Rao’s test on block-average data.|../mac/prof287.html",
    "A Declarative Approach to Modeling Interactive Cognition...||Member Abstract Track A Declarative Approach to Modeling Interactive Cognition Andrew Howes , School of Psychology, Cardiff University HowesA@cardiff.ac.uk Alonso Vera , Human Computer Interaction Institute, Carnegie Mellon University avera@mail.arc.nasa.gov Roger Remington , NASA Ames Research Center rremington@arc.nasa.gov We have constructed a prototype software tool, called CPMX, for predicting the time course of interactive cognitive skill. The tool calculates a prediction of an operator schedule, given a declarative description of the task and psychological constraints. In this respect, the tool shares similarities with the syndetic modeling technique developed by Duke and Duce (1999) and can be contrasted to simulation-based approaches to modeling cognition such as ACT-R (Anderson and Lebiere, 1998) and EPIC (Kieras and Meyer, 1997). Predictions are derived by reasoning about constraints on cognitive processing rather than by simulating transitions through a state-space. The task constraints are described in terms of a hierarchal analysis of the interaction between a person and the task environment. The psychological constraints are described in terms of a distributed set of processors each with its own processing capabilities. In this respect, CPM-X is influenced by the Model Human Processor (MHP; Card, Moran, Newell, 1983). Each processor is defined in terms of a set of properties. For example, a processor might typically execute an operator once every 50ms. Following Gray, John and Atwood (1993) operators are defined in terms of their processor requirements and dependencies. (In contrast to a simulation architecture, the state changes caused by operators are not represented.) Dependencies specify the preconditions for operator scheduling. For example, a dependency may specify that a vision operator which is for perceiving words must be preceded by a cognitive operator that attends to the appropriate modality. Similarly, given the task to type ‘ab’, the operator which is for initiating ‘type b’ is dependent on the completion of the operator for initiating ‘type a’. In CPM-X no commitment is made to a particular set of processors or operators. Instead, within the declarative description language, it is possible to define the set of processors and operators that correspond to the required theory of the human cognitive architecture. The modeling language therefore provides a means of expressing the class of theories in which human cognition is conceived of as an interacting set of simple information processors. In the models that we have built in CPM-X we have used a set of processors that were derived from the CPM-GOMS approach to modeling cognition (Gray, John, Atwood, 1993). These include a cognitive processor, one or more motor processors, and one or more perceptual processors. The operators that are executed by the motor and perceptual the cognitive processor, reflecting a central-bottleneck theory of mental processing. However, it should be possible to express alternative theoretical viewpoints. In addition to a description of the psychological processors and their operators, CPM-X also requires the description of strategies by which the processors work together to produce desired behaviors. For a particular description of the task and psychological constraints there are usually many possible ways in which the cognitive, perceptual, and motor operators can be scheduled. CPM-X uses a heuristic method to calculate an operator schedule that is consistent with the specified constraints and which, given the heuristics, takes as little overall time as possible. Following Vera, Remington, Matessa, John, Freed (2003) the schedule is presented as a PERT chart. The CPM-X prototype is implemented in Prolog and has been applied to a handful of toy Human-Computer Interaction problems. Our initial findings indicate that the technique can be used to make millisecond accurate predictions of skilled interactive behavior.|../mac/prof288.html",
    "Connectionist modelling of Chinese character pronunciation...||Member Abstract Track Connectionist modelling of Chinese character pronunciation Janet Hui-wen Hsiao , School of Informatics, University of Edinburgh h.hsiao@sms.ed.ac.uk Richard Shillcock , School of Informatics, University of Edinburgh rcs@inf.ed.ac.uk Psychologists and cognitive scientists aim to understand the universals involved when the brain deals with written language. Seidenberg and McClelland’s “triangle model” of the reading of monosyllabic English words has been substantially developed (e.g. Harm & Seidenberg, 1999), but with little application to languages other than English. Here we report our initial application of this approach to the reading of Chinese, a radically different orthography from English, illuminating processing universals in reading. The granularity of Chinese orthography is a fundamental issue in modelling its pronunciation. The character is the unit of pronunciation. Several formats exist, but a central paradigm exists in which a phonetic radical, on the right, specifies pronunciation information, and a semantic radical, on the left, specifies information about meaning (Figure 1). Chen et al. (1996) argue that these radicals (plus any remaining sub-character material) are the functional processing units of Chinese reading. Therefore, the typical granularity is substantially coarser than that found in English orthography. Is there thus a principled architecture for modelling Chinese pronunciation that uses distributed representations in a non-trivial way? We resolve the issue by referring to the anatomy of the visual pathways.|../mac/prof289.html",
    "How Studies of Hypertext-Assisted Learning Inform Educational System Design...||Publication-Based Talk Track How Studies of Hypertext-Assisted Learning Inform Educational System Design Amy M. Shapiro , University of Massachusetts Dale S. Niederhauser , Iowa State University The question of how we learn from hypertext is more complicated than how we learn from traditional text. While all the same elements of character decoding, word recognition, sentence comprehension, and so forth remain, a number of features unique to hypertext produce added complexity. Primary among these is the nonlinear structure of hypertext. One question that many researchers have addressed is how various nonlinear structures affect hypertext-assisted learning. The literature can appear downright contradictory but some common themes have emerged in our own work and is supported by the broader literature. In a nutshell, the effectiveness of good structures like hierarchies tends to hinge on interactions among learners’ prior knowledge, goals, and the level of activity and/or metacognition in the learners’ approach. The proposed talk will explain two general conclusions drawn from the literature and explain in detail the ways in which these variables interact to influence learning. The discussion will be grounded in theories of text-based learning (e.g., the construction-integration model). The first conclusion we make from our own work and that of others is that research on organizing tools and system structure indicates that well-defined structures (such as hierarchies) are helpful if the learning goal is to achieve simple, factual knowledge. Such structures are also helpful for domain novices. As the constructionintegration model (Kintsch, 1988) predicts, learning that does not require situation model construction (i.e., integration with prior knowledge) is facilitated by transparent information structures. In keeping with research in text-based learning, however, promoting active learning is also an important consideration. By providing a structure that is highly organized or simple to follow, even beginning learners may become passive. The challenge for designers is to structure the hypertext in ways that sufficiently challenge learners while not overburdening them to the point where learning is compromised. Our second conclusion is that ill-structured systems are often beneficial for advanced learners. Providing less obvious organizational structures has the effect of challenging the learner to seek coherence within the system. The overall effect is to promote active strategies, integration with prior knowledge, and improve learning. As the construction-integration model predicts, facilitating users to employ prior knowledge enhances learning outcomes. We do not claim, however, that ill-structured systems are always best for advanced learners, as they do not always apply their prior knowledge. A passive learner will garner little beyond explicitly stated facts from any hypertext system. It has become clear that hypertext (e.g., the internet) is not just another technological fad. It is here for a good long while, both in school and less formal educational settings. As such, it is important for educators and researchers to understand how students learn from such technology. The purpose of this talk is to present data-based conclusions|../mac/prof29.html",
    "Selective attention to Contextual Information in Japan...||Member Abstract Track Selective attention to Contextual Information in Japan Keiko Ishii , Department of Behavioral Science, Hokkaido University ishii@let.hokudai.ac.jp Shinobu Kitayama , Faculty of Integrated Human Studies, Kyoto University kitayama@hi.h.kyoto-u.ac.jp Recent studies have demonstrated that people in Western cultures tend to attend primarily to an object, whereas people in Eastern cultures tend to attend to the entire field, which consists of both the object and its context. Two of the tasks that have been used in this literature are of interest because the cognitive operations involved in these tasks appear to be largely automatic and relatively detached from semantic world knowledge (which is demonstrably different across cultures). To begin with, the relative attentional attunement to context for Easterners has been shown with a vocal Stroop task. Specifically, Ishii, Reyes, and Kitayama (2003) showed that in comprehension of emotional utterance Americans more spontaneously attended to verbal content than to paralinguistic cues (i.e., vocal tone), whereas Japanese and Filipinos more spontaneously attended to paralinguistic cues than to verbal content. Moreover, a conceptually consistent pattern of findings has been obtained in a completely non-social domain. Kitayama, Duffy, Kawamura, and Larsen (2003) found that Americans were more accurate than Japanese when they had to judge the absolute extent of a line stimulus while ignoring its surrounding frame. In contrast, Japanese were more accurate than Americans when they had to judge the relative extent of the same line by comparing it with the height of the surrounding frame. The present study examined whether the attentional attunement to context, demonstrated for Japanese in these two tasks, could be replicated with new groups of Japanese in a much larger age range. Moreover, we wanted to see if there is any correlation in performance between the two tasks.|../mac/prof290.html",
    "Toward a Kantian Defense of Jackendoff’s Psychologism...||Member Abstract Track Toward a Kantian Defense of Jackendoff’s Psychologism Luke Jerzykiewicz , Institute of Cognitive Science, Carleton University ljerzyki@ccs.carleton.ca Terry Dartnall (2000) suggests that much cognitive modeling unwittingly commits a variant of psychologism – the fallacy of assimilating mathematical truth to psychological fact. The worry appears particularly acute for Ray Jackendoff’s (2002) conceptual semantics (CS) – a naturalist, internalist, generative account of meaning and of the language faculty. Appearances are however deceptive. I argue that CS’s psychologism, while real, is benign. Jackendoff’s theory of abstract objects can answer antipsychologistic arguments if CS explicitly recognizes its Kantian leanings (cf. Falkenstein 1995). Furthermore, if this is correct, Jackendoff’s metaphysics gives grounds to reconsider whether sophisticated psychologism truly is a fallacy.|../mac/prof291.html",
    "What Representation Results from Feature Inference Category Learning?...||Member Abstract Track What Representation Results from Feature Inference Category Learning? Mark K. Johansen , Department of Psychology, University College London mark.johansen@ucl.ac.uk John K. Kruschke , Department of Psychology, Indiana University kruschke@indiana.edu Recently, a variety of research has been conducted in the feature inference category learning paradigm, the main results of which are summarized by Markman and Ross (in press). Participants learn about several perceptual categories by feature inference on the instances composing them-- given the category label and several features predict a missing feature--rather than by a more typical classification procedure--given several features predict the category label. trials that they responded to without corrective feedback.|../mac/prof292.html",
    "An ACT-R Model of Human Object-Location Memory...||Member Abstract Track An ACT-R Model of Human Object-Location Memory Todd R. Johnson , University of Texas Health Science Center Houston todd.r.johnson@uth.tmc.edu Hongbin Wang , University of Texas Health Science Center Houston hongbin.wang@uth.tmc.edu Jiajie Zhang , University of Texas Health Science Center Houston jiajie.zhang@uth.tmc.edu One important aspect of human spatial memory is the ability to remember the location of objects. Such locations are represented in multiple frames of reference, including eyecentered, body-centered, and object-based (world-centered) frames. The viewer-centered representations appear to require attention to encode; however, encoding object-toobject relations presents a computational problem. It seems unlikely that all object-to-object relations in the field of attention could be encoded, raising the question of which relations, if any, are available automatically, through early computation, and which may need to be generated later. To address this issue, we adopted an experimental paradigm developed by Milner and colleagues in the 1990s, (e.g., Owen, Milner, Petrides, & Evans, 1996). There are two phases in the Milner paradigm. In the encoding phase, eight black-and-white drawings (objects) are presented, one at a time, on a computer screen, with each accompanied by two solid black squares as landmarks. Participants are asked to remember the locations of these drawings, relative to the landmarks. In the retrieval phase, participants are presented a pair of identical study drawings (say, two chairs), together with some retrieval cues. However, the two identical drawings are presented at different locations, with one (the target), in its original location relative to the retrieval cues, and the other one (the distracter) in a different location. Participants must select the target out of the target-distracter pair. Milner varied the retrieval cues and whether the array of objects were presented in the same screen position as during encoding (fixed array) or shifted to one of the four corners of the display (shifted array). Retrieval cues were either the two landmarks or two of the drawings. This resulted in four conditions: fixed-landmark, fixed-object, shifted-landmark, and shifted-object. In fixed conditions, the target could be identified using absolute screen location. In shifted conditions, object to object relationships were required. Furthermore, shifted-object required object-toobject relationships among drawings that were never simultaneously displayed during study. To test the effect of retrieval cues, we modified the Milner paradigm by adding a fixed-nocue condition, in which only the target and distractor were present. The results of the experiment are shown as red lines in Figure 1. The main effects of cue-type and array-type, as well as their interaction were significant. There was no difference between fixed-no-cue and fixed-landmark. To model this data we extended ACT-R to automatically encode object-to-object relations between the previously and currently attended objects. Attending to an object produces an encoding of that object that includes its visual features and its screen-based location. Shifting attention to a second object builds a relation between the two objects. These relations are stored in declarative memory where they may later be retrieved. In all fixed conditions, the model attends to a drawing (either target or distractor), retrieves the screen-based study location of the object, and compares it to the current screen location. If the locations match, the model selects the drawing, otherwise the model chooses the alternative drawing. In shifted-landmark, the model attends to one of the drawings, then attends to one of the landmarks, resulting in a relation between the test drawing and landmark. It then tries to retrieve a study relation that matches the test relation. If one can be retrieved, it selects the current drawing, otherwise, it selects the other. All object-cue conditions first require a visual search for the pair of drawings that are the target and distractor. In shiftedobject, the model locates the target-distractor pair, then retrieves a study relation between one of the drawings and a landmark. It then uses this relation to shift attention to where the landmark would be if the test drawing were in its correct location. It then determines whether this hypothesized landmark location is in the same relative location to one of the cue drawings as it was during study. Using all of ACT-R’s default parameters, the model produces a good fit (R2 = 0.92) to the data.|../mac/prof293.html",
    "Using an Externalized recall procedure to unmask age-related deficits in in...||Member Abstract Track Using an Externalized recall procedure to unmask age-related deficits in inhibition Michael J. Kahana , Volen Center for Complex Systems kahana@brandeis.edu Emily D. Dolan , Volen Center for Complex Systems edolan@brandeis.edu Colin L. Sauder , Volen Center for Complex Systems csauder@brandeis.edu Arthur Wingfield , Volen Center for Complex Systems wingfiel@brandeis.edu|../mac/prof294.html",
    "Decision Making for Patient Safety in Critical Care Equipment Selection...||Member Abstract Track Decision Making for Patient Safety in Critical Care Equipment Selection Alla Keselman , Laboratory of Decision Making and Cognition, Department of Biomedical Informatics, Columbia University ak454@columbia.edu Vimla L. Patel , Laboratory of Decision Making and Cognition, Department of Biomedical Informatics, Columbia University patel@dmi.columbia.edu Mark J. Graham , Laboratory of Decision Making and Cognition, Department of Biomedical Informatics, Columbia University mjg24@columbia.edu Todd R. Johnson , School of Health Information Sciences, University of Texas Health Sciences Center at Houston Todd.R.Johnson@uth.tmc.edu Jiajie Zhang , School of Health Information Sciences, University of Texas Health Sciences Center at Houston Jianji.Zhang@uth.tmc.edu Injuries resulting from medical device use errors far exceed injuries arising from device failures. Until recently, human factors issues have received relatively little attention in medicine. Although the situation is gradually changing, many devices that are currently on the market are sub-optimal from the human factors perspective. This situation places significant responsibility for the device interface quality on the purchasers. Research on the influence of usability considerations on device purchasers’ decision making could provide valuable theoretical foundation for designing medical device selection guidelines. Medical device selection is a complex team decision making process that involves individuals with varying levels of specialized knowledge, and may reflect a number of individual and group biases. Klein and colleagues describe several aspects of team cognition and metacognition that distinguish successful teams (for review, see Klein, 1998). They suggest that successful teams are characterized by experience, stability and coherence. Members of such teams have common goals and share understanding of the situation. They use their “collective intelligence” to monitor their performance. In medicine, success of team functioning is related to similar decision-making characteristics (Patel, et al. 2002). This paper describes a retrospective analysis of an infusion pumps purchase in a large urban hospital system and focuses on cognitive and organizational factors in the decision making process. The study involved a) semi-structured interviews with nine participants in the latest infusion pumps purchase in a major urban hospital system, and b) analysis of documents relevant to the purchase. The data analysis was based on several formal qualitative analytic methods, including thematic coding and semantic analysis.|../mac/prof295.html",
    "A comparative analysis of learning in multitrial free and serial recall...||Member Abstract Track A comparative analysis of learning in multitrial free and serial recall Krystal A. Klein , Volen Center for Complex Systems Brandeis University krklein@indiana.edu Kelly M. Addis , Volen Center for Complex Systems Brandeis University addis@brandeis.edu Michael J. Kahana , Volen Center for Complex Systems Brandeis University kahana@brandeis.edu Two of the most widely used paradigms in mem- ory research examine the ability of participants to study lists of words and subsequently recall them. In the free recall paradigm, participants may recall the studied items from a list in whatever order they choose. In the serial recall paradigm, however, par- ticipants are instructed to recall the items from a list in the presented order. In addition to di®ering in re- call instruction, free and serial recall tasks di®er in item presentation order across trials of the same list, which typically varies from trial to trial in free recall and remains constant across trials in serial recall. Waugh (1961) compared these paradigms to a new condition, free recall with constant presentation or- der. She concluded on the basis of the average num- ber of items recalled that di®erences between free and serial recall were due only to recall instruc- tion and not presentation order. As learning curves provide an incomplete picture of the nature of re- call (Addis & Kahana, in press), the present study reevaluated the relationship between serial and free recall using Waugh's three conditions, through the analysis of recall transitions and organization of in- formation retained from trial to trial, in addition to overall recall performance. Twelve participants learned 21 19-item lists in each of three conditions: free recall with varied pre- sentation order, free recall with constant presenta- tion order, and serial recall. Learning took place over ¯ve study-test trials for each list. Results indicated that performance in the free re- call constant condition shares some commonalities with each of the other conditions. Overall item learning was found to be greater in serial recall and free recall constant than in free recall varied. Order learning was highest in serial recall, but signi¯cantly higher in free recall constant than in free recall var- ied. Analysis of item-to-item recall tendencies indi- cated that in free recall constant, participants make transitions that resemble a loose serial organization: they allow themselves to make backward transitions more frequently than in serial recall, but predomi- nantly recall in the forward direction. and are more likely to make transitions to nearby serial positions than to further positions. Finally, analysis of list organization across trials showed that level of orga- nization holds a consistent relationship with overall recall within each of the three conditions. However, e±ciency at retaining items from trial to trial is con- sistently high during all trials of free recall constant and serial recall, but begins low and gradually in- creases over trials in free recall varied. Contrary to Waugh (1961), we conclude that pre- sentation order has an e®ect on recall, causing some aspects of performance in free recall constant to di- verge from performance in free recalled varied as list trials proceed and to gain resemblance to per- formance in serial recall. Thus, we suggest that pre- sentation order and recall instructions have separate e®ects on learning performance in the recall tasks.|../mac/prof296.html",
    "The Role of Associations in Recognition of Mimetic and Imitative Words...||Member Abstract Track The Role of Associations in Recognition of Mimetic and Imitative Words Yuki Kobayashi , Department of Psychology, Kawamura Gakuen Woman’s University yuki_kobayashi@nifty.com This study investigated how associations affect the recognition of mimetic and imitative words. Associations play a very important role in word recognition, and mimetic / imitative words become a strong cue to recall target words. When we recall mimetic / imitative words in themselves, however, these associations rather interfere with recalling target words. For example, when “pouring” which associates “rain” is target, we would recall “misty” which also associates with “rain” instead of the target. Association would distort memory of mimetic / imitative words. Experiment 1 investigated how the association distorted the memory of mimetic / imitative words. It was predicted that participants would recall non-target words which have same association with target words. Experiment 2 investigated whether the distortion of memory in Experiment 1 affected productions of mimetic / imitative words. It was predicted that the larger distortion of memory would affect productions of mimetic / imitative words.|../mac/prof297.html",
    "Method to Test Spatial Structures of Spatial Terms...||Member Abstract Track Method to Test Spatial Structures of Spatial Terms Takatsugu KOJIMA , Graduate school of Education, Kyoto University kojima@cpsy.mbox.media.kyoto-u.ac.jp When the structure of a spatial term; above, below, left, right, etc, is tested, a rating task is used (Crawford, Regier, and Huttenlocher, 2000; Hayward and Tarr, 1995). This task is very easy to perform and less tiring for participants. In addition, it is suitable for examining areas or locations of boundaries of spatial terms. Kojima and Kusumi (2002), however, pointed out that the task does not entail relative relations between spatial points. Thus they used Thurstone's law of comparative judgment (case V) instead. This method still does not imply the locations of boundaries of spatial terms by itself, i.e. appropriate fit patterns of spatial terms have been kept intangible. In this study, I propose a systematical combination of both the methods in order to investigate spatial structures of spatial terms and make fit patterns of them. relations between the locations.|../mac/prof298.html",
    "Problem-solving Strategies that Distinguish Creative Artists...||Member Abstract Track Problem-solving Strategies that Distinguish Creative Artists Aaron Kozbelt , Psychology Department, Brooklyn College AaronK@brooklyn.cuny.edu Artistic creativity is of considerable general interest, yet it has not often been studied by psychologists and remains a mysterious phenomenon. For instance, it is unclear how to rigorously characterize the psychological processes leading to creative artistic products, or the relation (if any) between such processes and the creativity ofthe final product. One distinctive characteristic of highly creative artists may be that they have more effective problem solving strategies for creating original drawings than do their less creative peers. Such strategies have been explored to some extent in visual artists. Getzels and Csikszentmihalyi (1976) observed 31 college student artists creating original drawings. They found that artists who engaged in problem finding behaviors (e.g., handling more objects before drawing or including more abstract elements) produced more original drawings. They viewed problem finding processes as distinct from problem solving processes. However, this claim is controversial, as is the meaning of problem finding generally (Dudek and C6t6, 1994). The present study views the artistic process as goaldirected problem solving (Newell & Simon, 1972). ln solving the problem of creating a satisfying drawing, an artist's subgoals can unfold in either a straightforward, preplanned way, or in a more opportunistic way (Suwa, Gero, & Purcell, 1999). The purpose of the study is to examine artistic creation from a problem solving perspective and to examine closely the strategies that distinguish artists who produce very creative work from those who produce less creative work. The hypothesis is that more opportunistic strategies in the process of creation will be associated with higher level of final product quality and creativity.|../mac/prof299.html",
    "How to Plan and Run Online Seminars...||Tutorials Track How to Plan and Run Online Seminars Lisa Neal , eLearn Magazine lisa@acm.org Donna Anastasi , Aptima, Inc. anastasi@aptima.com The Cognitive Science Society, with funding from the ONR, started an online seminar series in 2002. This seminar series was initiated by Art Markman, with the assistance of the authors, who had previously conducted a seminar series for a cognitive task analysis community of practice, CTAResource.com. The Cognitive Science Society seminar series has been successful at promoting the society and reaching a global audience who might not otherwise have the opportunity to hear well-known cognitive scientists speak. Through the archives, people who were not able to attend due to conflicts or time zone constraints are able to re-play the sessions at a more convenient time. Online seminars are an increasingly common way to reach a broader audience, as exemplified by the Cognitive Science Society. This tutorial is designed to help people who wish to set up their own seminars to reach students outside of the classroom, to bring in the expertise of speakers who could not travel to their location, to promote innovative research from their labs, or for meetings of groups that cannot be physically together.|../mac/prof3.html",
    "Computational Explorations of Cognitive Development...||Publication-Based Talk Track Computational Explorations of Cognitive Development Thomas R. Shultz , Department of Psychology and School of Computer Science, McGill University thomas.shultz@mcgill.ca Many of the problems of children's cognitive development have been difficult and longstanding. Among these are issues of how children represent knowledge, whether stages exist, and how children achieve transitions between stages. Despite a century of scientific evidence on child development, comprehensive theoretical understanding of these issues has remained elusive. Part of the reason is that the problems of psychological development are too complex for traditional verbal theories of development. Nonetheless, it has become clear in recent years that considerable leverage on these problems can be gained by applying computational modeling, particularly using artificial neural networks. This is because computational modeling is a good way to capture complex processes, neural networks capture developmental phenomena in a natural way, and successful models can be examined in detail to discover insights into the phenomena that they simulate (Shultz, 2001).|../mac/prof30.html",
    "Minding the Gap: Understanding the Difference in Performance on Opacity and...||Member Abstract Track Minding the Gap: Understanding the Difference in Performance on Opacity and False Belief Tasks Carla S. Krachun , Institute of Interdisciplinary Studies (IIS), Carleton University ckrachun@ccs.carleton.ca Deepthi Kamawar , Department of Psychology and IIS, Carleton University deepthi_kamawar@carleton.ca Two tasks have commonly been used to investigate the developing representations of children: false belief (FB) and opacity tasks (described in more detail below). While both tasks require tracking of another’s representations, children consistently find opacity tasks harder. One possible reason for this—that the forced-choice nature of FB tasks makes them easier than opacity tasks—is the focus of the current study. Our research derives from a broader body of interdisciplinary research that includes issues in cognitive development (e.g., representational development), language development (e.g., syntax of complement clauses), and philosophy of language (e.g., propositional attitudes). In false belief tasks participants must keep track of another’s representations in order to predict her actions. For example, if Anna places an object in location A (cupboard), and then is absent while the object is moved to location B (fridge), she will represent the object as being in location A. Children who can monitor Anna’s representations will correctly predict she will look for the object in location A. Opacity tasks, like false belief tasks, also require participants to track another’s representations, but in opacity tasks the same referent has more than one term. For example, a woman may be both Sue’s mom and also a police officer. Each is correct, but replacing one term with its co-referential term can affect the truth value of the resulting sentence, setting up what is known as an ‘opaque context’ (Quine, 1995). Even if it is true, for instance, that “Mark believes the police officer rescued the cat from the tree”, it may be false that “Mark believes Sue’s mom rescued the cat from the tree” because Mark may not represent the woman as both Sue’s mom and as a police officer. To succeed in an opacity task, participants must recognize that Mark’s representation of the woman may not take in all possible representations.|../mac/prof300.html",
    "Human Multitasking: Towards an ACT-R Task-Independent General Executive...||Member Abstract Track Human Multitasking: Towards an ACT-R Task-Independent General Executive Yelena Kushleyeva , Department of Computer Science, Drexel University yk45@drexel.edu Dario D. Salvucci , Department of Computer Science, Drexel University salvucci@cs.drexel.edu One of the questions that arises in investigation of human multitasking is whether it is governed by general or taskspecific executive processes (see, e.g., Kieras et al., 2000). On one hand, specialized training in specific sets of concurrent tasks clearly leads to improved performance, thus implying involvement of task-specific mechanisms. On the other hand, people are generally able to multitask relatively successfully when given new tasks or new task combinations. This ability to apply time-sharing skills to novel settings favors the idea of generic executive mechanisms involved in human multitasking. To further understand the nature of general executives we modeled them within the ACT-R cognitive architecture (Anderson & Lebiere, 1998), thus relying on the architecture to inform us which potential scenarios fit into this theory of human cognition.|../mac/prof301.html",
    "Measuring Category Learning One Step at a Time...||Member Abstract Track Measuring Category Learning One Step at a Time Jonathan M. Kwasny , Department of Psychology, Binghamton University jkwasny@binghamton.edu Kenneth J. Kurtz , Department of Psychology, Binghamton University kkurtz@binghamton.edu) Traditionally, researchers have studied category acquisition by evaluating whether and how quickly learners reach performance criterion on a classification task and by conducting subsequent transfer tests. With some exceptions, research carried out within this paradigm makes little use of rich sources of trial-by-trial information about the time course of acquisition and response latency. In order to track the acquisition process, we present an experimental paradigm based on the inclusion of additional on-line behavioral measures within each classification learning trial. We believe the approach of investigating the microstructure of category learning will inform theoretical discussion of the representations and processes underlying categorization. In order to study category acquisition ‘as it happens’ a set of novel categories was designed for participants to learn. The stimuli were three types of grid patterns derived by introducing a level of distortion to source prototypes never seen by participants. This approach follows from Posner and Keele’s (1968) use of dot pattern stimuli. Each stimulus item was a grid pattern consisting of a 4 x 4 matrix composed of individual cells colored either black or white. Category instances were created by applying a level of distortion to the prototype by reversing the color of a number of cells. The prototypes themselves are much like everyday tile patterns with diagonal symmetry and a 50-50 ratio of black-to-white cells. Several design features bear further emphasis. The studies use a three-way classification task as opposed to the commonly used binary choice to promote the learning of a positively defined set of categories rather than merely learning members versus non-members of a single category. The studies have been designed to maximize naturalism by using large category sets (16 items per category) rather than small sets that easily become a ‘toy problem’. In addition the stimuli are designed to be convenient for experimental control, but not overly artificial as clear instantiations of a small set of binary feature values. The grid patterns invite individual interpretation and construction of higher order features.|../mac/prof302.html",
    "Which image processing algorithms best describe the minimal amount of visua...||Member Abstract Track Which image processing algorithms best describe the minimal amount of visual information required for image recognition? Cédric Laloyaux , Cognitive Science Research Unit, Université Libre de Bruxelles claloyau@ulb.ac.be Cédric Schmitt , LBUM, CHUM, Hôpital Notre-Dame cedric.schmitt@umontreal.ca In the framework of visual prosthesis, most methods use a video-camera in order to record visual information (Margalit et al., 2002). Determining the relevant level of information (e.g. contours, junctions) to be send to the retina, the optic nerve or a sensory substitution device for object recognition purpose, is a crucial aspect of raw images preprocessing (Boyle, 2002). For example, in Delbeke et al. (2002), a subject, with a four-contact optic nerve electrode, was able to discriminate simple patterns like “L” or “+” but nothing is said about the image processing algorithm to use in realworld conditions. In this research, we aim to investigate the minimal amount of information that different image processing algorithms provide, for object and scene recognition tasks.|../mac/prof303.html",
    "Hide and Seek: Using Computational Cognitive Models to Develop and Test Aut...||Member Abstract Track Hide and Seek: Using Computational Cognitive Models to Develop and Test Autonomous Cognitive Agents for Complex and Dynamic Tasks Frank J. Lee , Department of Cognitive Science, Rensselaer Polytechnic Institute fjl@rpi.edu Stéphane J. Gamard , Department of Cognitive Science, Rensselaer Polytechnic Institute gamars@rpi.edu Computational cognitive modeling refers to the study of human cognition through the development of computer programs to simulate cognitive processes. It is rooted in Artificial Intelligence and Cognitive Psychology and is being used successfully in many basic and applied areas including user modeling in Human-Computer Interaction, student modeling in Intelligent Tutoring Systems, and agent modeling in Computer-Generated Forces. In this abstract we describe the COGBOT project, an attempt to use computational cognitive modeling as a cognition engine to develop autonomous agents for complex and dynamic tasks.|../mac/prof304.html",
    "Human Supervisory Control: A Cognitive Perspective...||Member Abstract Track Human Supervisory Control: A Cognitive Perspective Simon Y. W. Li S. Li, UCLIC, University College London simon.li@ucl.ac.uk Christopher W. Myers , Cognitive Science Department Rensselaer Polytechnic Institute myersc@rpi.edu Michael J. Schoelles , Cognitive Science Department Rensselaer Polytechnic Institute schoem@rpi.edu Wayne D. Gray , Cognitive Science Department Rensselaer Polytechnic Institute grayw@rpi.edu Skill-based tasks in supervisory control systems often require the human operator to allocate attention to, for example, plan what to do next, monitor the system, or intervene when required to make adjustments (National Academy of Sciences, 1983). In this pilot study, we attempted to reproduce the supervisory control phenomenon in a simple simulated task environment. The basic supervisory control effect was replicated; namely, over checking the least prioritised item while under checking the most prioritised item. Results from the current study provide empirical against which future computational cognitive models will be compared.|../mac/prof305.html",
    "Does Orthography Influence Location Learning of Menu-Driven Computer Interf...||Member Abstract Track Does Orthography Influence Location Learning of Menu-Driven Computer Interfaces? Kin Pou Lie K. Lie, Department of Psychology, The University of Hong Kong kplie@hkusua.hku.hk To date most studies on location learning of menu-driven computer interfaces have been focused on the influences of label representativeness and the influences of practice. Little attention has been given to the potential influences of the orthographical characteristics of text labels. This research aims to investigate the issue by comparing location learning of Chinese menu items and location learning of English menu items. Chinese was singled out to be compared with English, not only because Chinese is a widely used language, but also because there are two pertinent orthographical differences between Chinese and English: 1. Written English consists of word strings that differ in form, in height, and in length, whereas written Chinese consists of characters constructed from strokes in a uniformly square-shaped area. It is conjectured that the varying word shapes of written English act as contextual cues (Chun, 2000) that facilitate location learning. In contrast, the unvarying word shapes of written Chinese provide little or no such facilitation. 2. Written Chinese is, in general, visually more complex and spatially denser than written English. It is conjectured that, compared with written English, the complexity and density of written Chinese imposes a higher perceptual load (Lavie, 1995), which in turn uses more attentional capacity and restrains location learning to a greater extent.|../mac/prof306.html",
    "Viewpoints in Embodied Objectivity...||Member Abstract Track Viewpoints in Embodied Objectivity Daniel Hsi-wen Liu , Division of Humanities, Providence University hwliu@pu.edu.tw This research explores a novel sense of objectivity for understanding perception and perceptual imagery, as a continuing endeavor after Grush (2000) on the topic of objectivity in the embodied cognitive science. The profile of building this sense of objectivity uses the embodied nature of deictic codes as source materials of consideration, as discussed in Ballard et al. (1997). They see deictic codes as (visual or haptic) frames set up on different points in the scene; an agent can continually fixate on various such points. A frame consists of several inter-related (visual or handmotor) routines, thereby the points on which frames are set up can be regarded as viewpoints of perception and imagery.|../mac/prof307.html",
    "Separating the Effects of Duration and Neighborhood Density in Nonword Repe...||Member Abstract Track Separating the Effects of Duration and Neighborhood Density in Nonword Repetition Latency John Lipinski , Department of Psychology, University of Iowa john-lipinski@uiowa.edu Prahlad Gupta , Department of Psychology, University of Iowa prahlad-gupta@uiowa.edu One of the most interesting results in the study of neighborhood effects on nonword repetition has been Vitevitch & Luce’s (1998) finding that increased neighborhood density has a facilitatory effect on nonword repetition latency. However, analysis of the stimuli used by Vitevitch and Luce (1998) reveals a mean stimulus duration difference of 16 ms that may have contributed to the observed difference. Experiment 1 presents a replication of the basic response latency effect and Experiment 2 presents a reduction of the effect to nonsignificance after controlling for stimulus duration differences through stimulus extension and compression.|../mac/prof308.html",
    "Teleological Explanation: Why We Answer “Why?” the Way We Do...||Member Abstract Track Teleological Explanation: Why We Answer “Why?” the Way We Do Tania Lombrozo , Department of Psychology, Harvard University lombrozo@wjh.harvard.edu Susan Carey , Department of Psychology, Harvard University scarey@wjh.harvard.edu Seeking explanations is a fundamental part of everyday cognitive life. Many accounts of concepts and naïve theories have correspondingly placed explanation at the center of conceptual representation. Concepts for which we have an underlying explanation are learned more efficiently (e.g. Spalding & Murphy, 1999) and said to gain conceptual coherence (Murphy & Medin, 1985) and embody systematic sets of beliefs (Keil, 1989). Similarly, causal-explanatory principles are thought to ground our naïve theories of physics, psychology, and biology (e.g. Carey, 1985). Despite the prevalence of explanation as both an everyday behavior and a theoretical construct, there are many open questions about its cognitive basis. For example, why do we apply certain kinds of explanations selectively? Here we summarize three experiments investigating the conditions under which adults accept teleological explanations (TEs) — explanations in terms of a function or goal. Specifically, we examine whether the domain of what’s being explained, the causal role of the function being invoked in the explanation, and the familiarity of the causal process influence TE acceptance. While previous work suggests that adults restrict TEs on the basis of domain (Kelemen, 1999), we predicted that TEs would be accepted whenever the function invoked in the explanation played a causal role in bringing about what was being explained, consistent with Wright’s (1976) philosophical analysis.|../mac/prof309.html",
    "How space binds objects and words...||Publication-Based Talk Track How space binds objects and words Linda Smith , Indiana University The real world contexts in which children learn object names is one in which multiple objects compete for attention, in which attention is often shifted from one object to another in rapid succession, and in which objects often move in an out of sight. Yet children seem to keep track of individual objects and intended referents. This paper presents evidence that young word learners (15 month olds) keep track of objects through a system of implicit deictic reference whereby attention to particular locations in space is used to define and index individual objects. Our task is borrowed from a prior study by Baldwin. In our version, the child is first presented with a novel object at the A location (e.g., on the left) and attention is directed to that object without naming. The child is then presented with a different object at B (e.g.,to the right) and attention is directed to that object. Each object is then hidden: the A object in a bucket at location A and the B object in a bucket at location B. A delay period (3 to 5 sec) then intervenes. The experimenter then directs attention to one bucket (e.g, A) and offers a name (e.g., “There is a dax “). In Experiment 1 (as in Baldwin’s study), the experimenter then immediately shifts attention to the other bucket (at location B) and pulls object B from the bucket and into view such that the naming event is most closely linked by space to one object (A) but by time to the other (B). Children were subsequently presented with both objects in a neutral location and asked to get one by name. They consistently chose the object most closely linked to the name by location, not by time. Four subsequent experiments using a similar procedure (except children were never shown which objects had been hidden in which buckets) showed that location is a strong force binding objects and intended referents: Without a spatial link, that is attention to the same location in space, 15 month olds cannot map names to objects. Moreover, children take the name as referring to any (and all) objects coherently linked to the attended location to which the name is offered. Finally, children can track an object (an intended referent) over multiple locations if the path of motion is spatially and temporally coherent. The structure of the task and children’s interpreted referents are highly reminiscent of infants perseverative errors in the A not-B task. However, here, spatial coherence organizes intended referents (rather than generating errors). In the discussion, we note that the use of space to bind objects to intended referents is characteristic of gestures and also American Sign Language. We suggest that this is because attention provides a momentary reference to an object and a location (or path) in space and in so doing binds individual objects to cognitive processes.|../mac/prof31.html",
    "Dynamics of Applause: Modeling Group Phenomena Through Agent Interaction...||Member Abstract Track Dynamics of Applause: Modeling Group Phenomena Through Agent Interaction Gary Lupyan , Center for Neural Basis of Cognition glupyan@cnbc.cmu.edu Ilya Rifkin , Department of Computer Science. Cornell University ir23@cornell.edu Unlike many social conventions that proved to be passing fads, applause appears to be a remarkably stable facet of human culture. Although its mani- festations have varied from the °uttering of fans of Victorian ladies, to hearty overhead clapping and cat- calls of less \\re¯ned concert-goers, the desire to ex- press approval through a culturally-speci¯ed display has been in existence for millennia. Demonstrating the perceived importance of applause, French theater and opera companies of the 1800s hired claques, pro- fessionals who led others in applause to secure encores and educate audiences on the ¯ner points of the per- formance. Using a dynamic game theory model, the present work looks at how applause dynamics emerge from the interaction of individual desires. What are the bene¯ts that drive applause? What are the costs that work to end it? How do the costs and bene¯ts of an individual interact with those of others in the au- dience? Despite some simplifying assumptions made in the present model, it o®ers robust predictions of applause in audiences of varying sizes responding to events of varying levels of \\impressiveness. More gen- erally, this work demonstrates how a complex social behavior can be modeled with few assumptions and free parameters.|../mac/prof310.html",
    "Multiple Category Use in Spatial Memory: Issues of Category Salience...||Member Abstract Track Multiple Category Use in Spatial Memory: Issues of Category Salience Keith Maddox , Department of Psychology, Tufts University keith.maddox@tufts.edu David N. Rapp , Department of Educational Psychology, University of Minnesota rappx009@umn.edu Holly A. Taylor , Department of Psychology, Tufts University holly.taylor@tufts.edu In the most general sense, categories help people manage an unlimited amount of available information. When learning an environment, people hierarchically organize spatial information, grouped by spatial categories (e.g., McNamara, 1986; Stevens & Coup, 1978). Further, non-spatial information appears to be associated with its respective location in memory, whereby non-spatial information primes locations (McNamara, Halpin, & Hardy, 1992). What happens when the non-spatial information can be categorized separately from the locations? Our previous work (Maddox, et al., 2002) examined use of the non-spatial category of race in a map learning task. We proposed two ways that non-spatial category information might be used. The Selective Category Application Theory suggests that people use category information only when it is functional or salient. The Generalized Category Application Theory proposes that people use category information when available, regardless of applicability or salience. Our previous findings support selective application, driven by both function and salience. Distance estimations showed influences of both spatial and social category information, whereby the spatial category had functional relevance and the social (race) category may have been particularly salient for our college-aged population. Matching of spatial to non-spatial information showed influences of only the social category, which had functional relevance and high salience for the task. The present restudy examined spatial and social categorization in map learning, using a social category with less salience than race—political affiliation.|../mac/prof311.html",
    "A Long Interval affects the Mere Exposure Effect for the Prototypes...||Member Abstract Track A Long Interval affects the Mere Exposure Effect for the Prototypes Ken Matsuda , Faculty of Education, Kyoto University ken@p01.mbox.media.kyoto-u.ac.jp Takashi Kusumi , Faculty of Education, Kyoto University n50609@sakura.kudpc.kyoto-u.ac.jp We examined how a long interval influences concept formation and affective judgment following a mere exposure. Using a mere exposure (e.g., Zajonc, 1968) and the concept formation paradigm (e.g., Barsalou et al., 1999), Matsuda & Kusumi (2002) discovered three points. First, concept formation with repeated exposure was based on the event, and the concept builds a prototype. Second, the judgment of like/dislike is affected by stimulus typicality and exposure frequency. Third, prototypical stimuli that integrate the dimensions of each individual are preferred if the value of that dimension is weighted. While Matsuda & Kusumi (2002) used a short retention interval (5 min), we used a longer interval because it is important to explore the long-term process of concept formation and affective judgment, since implicit memory influences the mere exposure effect (e.g., Schacter, 1987; Squire, 1992; Seamon et al., 1995) and the retention curve is U-shaped (Matsuda, 2000).|../mac/prof312.html",
    "Stochastic learning in neural network models of categorization...||Member Abstract Track Stochastic learning in neural network models of categorization Toshihiko Matsuka , Department of Psychology, Rutgers University - Newark matsuka@psychology.rutgers.edu James E. Corter , Department of Human Development, Teachers College, Columbia University jec34@columbia.edu A number of neural network models of categorization have been proposed. The models differ notably in the types of internal representation used (e.g. exemplars vs. prototypes; hyperplane vs. hypersphere activation regions). However, many of these NN models of categorization (e.g., ALCOVE) use some form of gradient method for learning. These methods have been successful in reproducing group learning curves, but tend to underpredict variability in individual-level data, for both accuracy and attention allocation measures (Matsuka, 2002). Here, we show that use of a different learning algorithm with a given model can result in different learning trajectories and more realistic variability in individual learning curves, especially for attention allocation. Our proposed algorithm is a form of constrained simulated annealing (Ingber, 1989). Initial parameter sets (dimensional attention weights and network connection weights) are randomly selected. At the beginning of each training epoch, a hypothetical “move” in the parameter space is computed by adjusting each parameter by an independently sampled term. These adjustment terms are drawn from a prespecified distribution (e.g., a Cauchy distribution). The move (i.e., the set of new parameter values) are accepted or rejected, based on the computed relative fit of the new values. Specifically, if the new parameter values result in better fit, they are accepted. If they result in worse fit, they are accepted with some probability P. The adjustment in parameters is very rapid initially, and it gradually decreases over learning blocks.|../mac/prof313.html",
    "The Implicit Influence of Idioms...||Member Abstract Track The Implicit Influence of Idioms Matthew S. McGlone , Department of Psychology, Lafayette College mcglonem@lafayette.edu Heather Bortfeld , Department of Psychology, Texas A & M University bortfeld@psyc.tamu.edu Communicators use analogies in strategic discourse to invite inferences about ambiguous situations that reinforce their own construals of these situations. For example, in the short-lived political debate preceding the U.S.’s entry into the 1991 Persian Gulf conflict, supporters of our involvement likened it to WW II (a war most Americans believe we “won”), whereas opponents called attention to its similarities with the Vietnam War (which many consider a “defeat”). Several years earlier, Gilovich (1981) found that political science students were far more likely to recommend intervention in a hypothetical foreign policy crisis when irrelevant features of the scenario (e.g., the location used for press briefings) called to mind WW II (Winston Churchill Hall) rather than Vietnam (e.g., Dean Rusk Hall). The presence of analogical cues did not, however, lead students to judge the scenario as being more similar to one of the previous conflicts than the other. Subsequent reasoning research has also found that seemingly trivial cues can compel people to unwittingly employ historical analogies in their judgments and decisions. The reported research explores idioms’ potential as an unobtrusive means of introducing analogies into decision scenarios. Many of the conventional figurative expressions we use to describe abstract concepts have a common analogical derivation (e.g., Lakoff & Johnson, 1980). For example, the idioms we use to describe corporate organizations reflect analogies to families (e.g., parent company), sports (sales team), ecological systems (e.g., business climate), and other source domains. In business correspondence, corporate executives tend to favor organizational idioms derived from analogies that cohere with their own beliefs about management (Morgan, 1997). How might this preference influence their correspondents’ perceptions of the firm?|../mac/prof314.html",
    "Cognition in Jazz Improvisation...||Member Abstract Track Cognition in Jazz Improvisation David Mendonça , Information Systems Department mendonca@njit.edu W.A. Wallace , Department of Decision Sciences and Engineering Systems wallaw@rpi.edu This research investigates thinking processes of duos of jazz improvisers in performance. Of particular interest are cognitive processes related to creativity and to reasoning about time, since both activities are fundamental to improvisation. Data sources are the retrospective verbal protocols of individual members of these duos. The protocols were collected from each performer as he watched and listened to a recording of the performance and said out loud what he had been thinking while taking part in it. One result of this work is that cognition related to reasoning about time and creativity varied little either within or across groups, regardless of the type of tune being played. This result is further investigated by examining some of the statements from the protocols themselves.|../mac/prof315.html",
    "Teaching Cognitive Science through Collaborative Reflection (1): Overview...||Member Abstract Track Teaching Cognitive Science through Collaborative Reflection (1): Overview Naomi Miyake , School of Computer and Cognitive Sciences, Chukyo University nmiyake@sccs.chukyo-u.ac.jp Hajime Shirouzu , School of Computer and Cognitive Sciences, Chukyo University shirouzu@sccs.chukyo-u.ac.jp Yoshio Miyake , School of Computer and Cognitive Sciences, Chukyo University ymiyake@sccs.chukyo-u.ac.jp We have been developing and testing an undergraduate curriculum set to teach cognitive science [1][2][3]. To do so we rely on cognitive scientific research findings and ways of thinking. At the same time we try to feedback our findings to strengthen our understandings of how people learn, to implement information environment suitable for learning. The target is general, because we believe the knowledge of cognitive science has pragmatic value for most of what we do in everyday lives.|../mac/prof316.html",
    "Strategically Logical: Experience-Related Changes in Strategy Use on Deduct...||Member Abstract Track Strategically Logical: Experience-Related Changes in Strategy Use on Deductive Problems Bradley J. Morris , Grand Valley State University morrisb@gvsu.edu Christian Schunn , University of Pittsburgh schunn@pitt.edu We propose a conceptual framework for explaining logical reasoning in terms of competing strategies. The Logical Strategy Model (LSM; Morris & Schunn, in press). describes a series of strategies that differ in their functionality. Algorithmic strategies (e.g., token- based, verbal) are more costly (i.e., longer processing time) but more accurate, while heuristic strategies (e.g., analogies, matching rules, knowledge-based rules) are less costly but less accurate. The LSM was tested with 45 undergraduates, 23 graduate students, and 15 children ages 8-11. Each subject was given 24 deductive problems and asked to reflect on how they solved a problem by selecting one of five descriptions of a strategy. We conducted this experiment to test three predictions of the LSM: (1) subjects should report using a variety of strategies across the problem set, (2) the use of algorithmic strategies should be associated with more correct responses while the converse should be true for heuristic strategies, and (3) increasing experience should be associated with greater use of algorithmic strategies.|../mac/prof317.html",
    "Knowledge Representation in Engineering Design: An Initial Investigation...||Member Abstract Track Knowledge Representation in Engineering Design: An Initial Investigation Jarrod Moss , Department of Psychology, Carnegie Mellon University jarrodm@cmu.edu Kenneth Kotovsky , Department of Psychology, Carnegie Mellon University kotovsky@cmu.edu Jonathan Cagan , Department of Mechanical Engineering, Carnegie Mellon University cagan@cmu.edu Engineering design is a domain in which a number of complex problem solving activities occur. As in all such tasks, cognitive processes operate upon the internal representations of the task as well as upon other relevant knowledge. These representations can change over the course of experience in order to enable a person to better respond to the problems and challenges of a domain. The time course of these representation changes is a reflection of the structure and content of a domain as well as the cognitive learning mechanisms responsible for the changes. The goal of the work presented here is to identify some of the characteristic differences that distinguish experts and novices in this domain. While these studies only include freshman and senior engineering students, there are a number of interesting findings, and there are plans to extend this work to professional engineers. The first study utilizes a recall paradigm that has been employed by a number of researchers looking at expert/novice differences (e.g., Chase & Simon, 1973). The second study uses Latent Semantic Analysis (Deerwester, Dumais, Furnas, & Landauer, 1990) as a methodological tool to aid in exploring and analyzing the content of students’ representations.|../mac/prof318.html",
    "Measuring Inquiry Cycles in Simulation-Based Leaning Environments...||Member Abstract Track Measuring Inquiry Cycles in Simulation-Based Leaning Environments Tom Murray , Hampshire College School of Cognitive Science tmurray@hampshire.edu Larry Winship , Hampshire College School of Cognitive Science lwinship@hampshire.edu Neil Stillings , Hampshire College School of Cognitive Science nstillings@hampshire.edu The project An Inquiry-Based Simulation Learning Environment for the Ecology of Forest Growth consisted of three main stages of approximately one year each: software development, curriculum development, and classroom implementation. We evaluated the software, called SimForest, in clinical and college classroom settings, then ran a professional development program to support eight secondary school teachers in incorporating the software and curriculum into their classes. This paper focuses on our results from the clinical and classroom trials in which we observed students working in pairs using the software. SimForest is a simulation-based learning environment in the domain of forest ecology that simulates tree and forest growth, the succession of tree species over time, and the effects of environmental and man made disturbances on forest growth (see Figure 1). In the simulation students set environmental parameters such as rainfall, temperature, soil fertility, soil texture, and soil depth; they plant a plot of trees from a list of over 30 species; and they run the simulation and observer the trees as they grow and the forest evolves. A forest plot's sensitivity to natural and man-made disturbances can be evaluated, and emergent properties such as species succession can be observed.|../mac/prof319.html",
    "On the Continuity of Mind...||Publication-Based Talk Track On the Continuity of Mind Michael J. Spivey , Department of Psychology, Cornell University spivey@cornell.edu A recent movement in the cognitive sciences is encouraging that we discard the computer metaphor of the mind in favor of a continuous (both in time and in feature-space) dynamical framework for describing cognition (e.g., Port & Van Gelder, 1995; Spivey, in preparation; Thelen & Smith, 1993; Van Orden, Holden, & Turvey, in press). As much of the advancement of this metatheoretical framework has taken place in motor movement research (e.g., Kelso, 1995), a significant proportion of the cognitive science community has conveniently been able to ignore it, more or less. However, as contemporary theorists (e.g., Ballard, Hayhoe, Pook & Rao, 1997; Barsalou, 1999; Glenberg, 1997) renew an emphasis on the physical embodiment of cognition, and the pivotal role of action in all thought, it becomes clear that motor movement -- and especially the theoretical advances of dynamical systems that have come with it -- need to figure prominently in our treatment of mind. In this talk, I will briefly touch on a number of experimental findings, from a few different cognitive psychology laboratories, that appear more consistent with a dynamical-systems perspective on cognition than an information-processing one. These findings come from some of the core areas in traditional cognitive psychology, including categorical perception, visual attention, spoken word recognition, and sentence processing. When placed in the context of neurophysiological evidence for distributed neuronal population codes coalescing over time, and computational demonstrations of attractor network dynamics, these findings converge on a description of the mind as a graded, probabilistic, continuously flowing “event”, rather than a discrete logical stage-based “object”. Without abandoning the vast empirical database produced by decades of traditional cognitive psychology, the new framework encourages an extension of these inquiries using continuous on-line experimental measures that can reveal the real-time dynamical nature of cognition, perception, and action. Additionally, a computational characterization of those temporal dynamics can be provided by attractor networks, which loosely approximate both the neurophysiological properties and the temporally continuous nature of real biological neural networks. In a dynamical (as well as ecological) psychology, we are compelled to treat mind as a continuous nonlinear trajectory through a high-dimensional state-space; not as a box full of boxes full of rules and symbols. As the debate continues (cf. Dietrich & Markman, 2000), the benefits of this new perspective will be witnessed in the decades to come.|../mac/prof32.html",
    "Toward Characterizing Best-Practice Pedagogy for Inquiry in Simulation-Base...||Member Abstract Track Toward Characterizing Best-Practice Pedagogy for Inquiry in Simulation-Based Leaning Environments Tom Murray , Hampshire College School of Cognitive Science tmurray@hampshire.edu , Hampshire College School of Cognitive Science lwinship@hampshire.edu , Hampshire College School of Cognitive Science nstillings@hampshire.edu The project An Inquiry-Based Simulation Learning Environment for the Ecology of Forest Growth consisted of three main stages of approximately one year each: software development, curriculum development, and classroom implementation. We evaluated the software, SimForest, in clinical and college classroom settings, then ran a professional development program to support eight secondary school teachers in incorporating the software and curriculum into their classes. This paper focuses on our results from the clinical trials in which we analyzed the strategies used by an expert teacher. A primary goal of this phase of our research was to document a variety of successful inquiry-based methods for using simulations in the classroom. SimForest is a simulation-based learning environment in the domain of forest ecology that simulates tree and forest growth, the succession of tree species over time, and the effects of environmental and man made disturbances on forest growth. In the simulation students set environmental parameters such as rainfall, temperature, soil fertility, soil texture, and soil depth; they plant (or load in from a file) a plot of trees from a list of over 30 species; and they run the simulation and observer the trees as they grow and the forest evolves. A forest plot's sensitivity to natural and man-made disturbances can be evaluated, and emergent properties such as species succession can be observed.|../mac/prof320.html",
    "Toward an Operational Definition of the Zone of Proximal Development for Ad...||Member Abstract Track Toward an Operational Definition of the Zone of Proximal Development for Adaptive Instructional Software Tom Murray , School of Cognitive Science, Hampshire College tmurray@hampshire.edu Ivon Arroyo , Computer Science, University of Massachusetts ivon@cs.umass.edu Measuring and comparing student learning in adaptive computer assisted learning (ACAL) systems is problematic because the system is trying to both model and change the user, and in this sense is chasing a moving target. Processoriented metrics for measuring learning, such as the zone of proximal development (ZPD) can be more robust in such situations. Though the concept of the ZPD is often invoked in the context of instructional systems, it has not been operationalized in a manner that allows it to be used in ACAL. We propose a straight-forward method for measuring ZPD-learning that focuses on the ongoing amount of hints or help that learners need as they solve problems. The ZPD is commonly used to articulate apprenticeship learning approaches, scaffolding and fading (note: references removed from this extended abstract, available from the authors), and authentic (situated) learning tasks. The ZPD describes a zone within which tasks are too difficult to accomplish without assistance, but which can be accomplished with some help. The ZPD in terms of a student's readiness to learn a new skill in terms of the assessment of learning potential or learnability. These descriptions of the ZPD are useful for framing certain educational issues, but they are not defined in an operational way. We argue that keeping the learner within this optimal zone could be described in several compatible ways: ß Putting a greater emphasis on monitoring learning processes variables and maintaining efficient as well as effective learning; ß Cognitively there is a goal to presenting material that is neither too easy nor too difficult; ß Affectively there is a goal of avoiding the extremes of boredom and confusion (being overwhelmed); ß This can also be seen as maintaining a constant level of challenge (and support), or a constant rate of learning.|../mac/prof321.html",
    "This Way or That: Determining Where to Look First...||Member Abstract Track This Way or That: Determining Where to Look First Christopher W. Myers , Cognitive Science Department, Rensselaer Polytechnic Institute myersc@rpi.edu Wayne D. Gray , Cognitive Science Department, Rensselaer Polytechnic Institute grayw@rpi.edu Michael J. Schoelles , Cognitive Science Department, Rensselaer Polytechnic Institute schoem@rpi.edu Although there are many studies of how distractor-target differences influence visual search (e.g., see Shen, Reingold, & Pomplun 2000), there are no studies on how configural or historic factors influence the initial saccade of a search. The current study suggests that both the current configuration and history are important and have an additive effect on saccadic selectivity.|../mac/prof322.html",
    "The Automaticity of Image Schema Function in Metaphor Comprehension...||Member Abstract Track The Automaticity of Image Schema Function in Metaphor Comprehension Keiko Nakamoto , Faculty of Education, Kyoto University kenakamoto@nifty.com The purpose of this study was to provide experimental evidence for orientational metaphors on the basis of the VERTICAL image schema (e.g., Lakoff & Johnson, 1980). One of the hypotheses implied by the conceptual metaphor view assumes that metaphorical thought functions automatically in people’s online understanding of linguistic meaning. If this is correct, we should always access the image schema whenever we understand the meaning of a word that is believed structured by orientational metaphor. To test the hypothesis, the Stroop-like spatial judgment task was employed (Nakamoto, 2000). In the task, the directionality on the VERTICAL schema of target words was introduced as the irrelevant dimension to the spatial judgment. If the target words, expressed using verticalityrelated words, are understood through the image schema, then directionality on the schema should affect judgment.|../mac/prof323.html",
    "Phonological Memory Predicts Second Language Oral Gains in Adults...||Member Abstract Track Phonological Memory Predicts Second Language Oral Gains in Adults Irena O’Brien , Département de psychologie, Université du Québec à Montréal irena.obrien@internet.uqam.ca Norman Segalowitz , Department of Psychology and the Centre for the Study of Learning and Performance, Concordia University segalow@vax2.concordia.ca Recent research has stressed the importance of phonological short-term memory (PSTM) in children’s first (L1) and second (L2) language acquisition (Baddeley, Gathercole, & Papagno, 1998). Indeed, Baddeley et al. (1998) have called PSTM a language learning device. Although PSTM has also been shown to be implicated in L2 learning by adults, adult studies have tended to look at experimental word learning in the laboratory over one or two sessions (see Baddeley et al., 1998). The goal of the present study was to examine the role of PSTM in other areas of L2 learning, namely gains in oral proficiency and fluency made by adults outside the laboratory and over a longer time period.|../mac/prof324.html",
    "A Word-Net Vector Space Frequency Semantic Link Distance Model of Word-Mean...||Member Abstract Track A Word-Net Vector Space Frequency Semantic Link Distance Model of Word-Meaning Equivalence Kunal Patel , School of Computer Science and Electrical Engineering and School of Behavioral and Brain Sciences, University of Texas kunaliitb@hotmail.com Richard M. Golden , School of Behavioral and Brain Sciences, University of Texas golden@utdallas.edu The ability to identify indices of semantic similarity among words is an important problem which has important applications in the fields of both artificial intelligence and experimental psychology. Unfortunately, the problem of rank-ordering words according to their semantic similarity measure is not a straightforward one even with powerful semantic network models of the lexicon such as WORDNET (Fellbaum, 1998). The problem arises because “word-meaning equivalence” is not some objective quantity but is a subjective context-dependent property. Two words that seem very different in one context might be judged semantically similar within another context (and viceversa). For example, a model of the lexicon such as WORDNET would predict the superordinates of CUP: “CROCKERY” or “DISHWARE” to be more consistent with the meaning of the word “CUP” than the subordinates of “CUP”: “TEA CUP” or “COFFEE CUP”. But such predictions might not be consistent with human performance. That is, people might simply be more likely to use the phrase “COFFEE CUP” instead of the word “CROCKERY” when they want to express the meaning “CUP”. The goal of this research is to compare the standard semantic similarity measure of distance in WORDNET which is based upon the number of links separating two words in WORDNET with two new algorithms for computing semantic distance.|../mac/prof325.html",
    "Left of Zero: Representing Negative Numbers on the Mental Number line...||Member Abstract Track Left of Zero: Representing Negative Numbers on the Mental Number line Richard W Prather , University of Wisconsin-Madison, Department of Psychology rwprather@wisc.edu Lera Boroditsky , Department of Brain & Cognitive Sciences, MIT lera@mit.edu How do people represent and use abstract mathematical notions that have few or no real-world counterparts? For example, how are we able to think of imaginary numbers, surreal numbers, or even negative numbers? One proposal is that abstract knowledge domains (such as number, time, ideas, ect.) are represented through analogical extensions from more concrete or more experience based domains (e.g., Lakoff & Johnson, 1980, Boroditsky 2000). Lakoff and Nunez (2001) have suggested that the same kinds of experience based may be used for all manner of mathematical concepts. This paper focuses on the domain of negative numbers. Previous studies of positive number representations have found evidence for the psychological reality of a mental number-line (e.g., Dehaene, Bossini, & Giraux, 1993; Reynvoet & Brysbaert 1999). In these studies subjects were shown positive integers one at a time and asked to indicate whether the number was odd or even. The subjects made their response by pressing the ‘odd’ and ‘even’ buttons that were spaced far apart on the keyboard. These studies found that people made an implicit association between spatial location and number magnitude called the Spatial Numerical Association Response Codes or SNARC effect. We consider the new case of negative numbers. How will people represent numbers less than zero on the odd/even task described above? Does the SNARC effect hold only for positive numbers that can correspond to actual numbers of objects in the world? If negative numbers are represented on the mental number-line, will they be laid out according to magnitude or absolute value?|../mac/prof326.html",
    "Learning By Doing: Cognitive Apprenticeship in Chemical Research...||Member Abstract Track Learning By Doing: Cognitive Apprenticeship in Chemical Research Amy B. Preece , Department of Chemistry, University of Kansas apreece@ku.edu Janet Bond-Robinson , Department of Chemistry, University of Kansas jrobinso@ku.edu Studies of science-as-practice increase robustness of epistemology while assuming that scientists are simply cognitive agents in specialized communities in which (a) human judgment and representational use are exercised within research (Giere, l992), (b) scientists’ problem-solving strategies and representations, refined over the history of science, are sophisticated outgrowths of ordinary reasoning and representations (Nersessian, 1999), and (c) mental models are integral to reasoning and capable of illustrating dynamic cause and effect in real-world systems.|../mac/prof327.html",
    "Women's Sexual Strategies: More Common (and Diverse) than We Think?...||Member Abstract Track Women's Sexual Strategies: More Common (and Diverse) than We Think? Holly Ramsawh , Department of Psychology, Boston University hramsawh@bu.edu Catherine Harris , Department of Psychology, Boston University charris@bu.edu The literature on human mating strategies emphasizes gender differences in the pursuit of short-term sexual strategies. In particular, women are believed to be most concerned with pursuing long-term sexual partners, because of the greater parental investment that women make in their offspring (Buss & Schmitt, 1993; Laumann et al., 1994; Oliver & Hyde, 1993). We accept the premise of Parental Investment Theory (Travis & Yeagers, 1991; Trivers, 1972), but take issue with the empirical studies that are used to support Buss and Schmitt's (1993) Sexual Strategies Theory (SST), and with the extent of the purported difference between the genders. All empirical studies of gender differences in preferences for casual sex have been conducted with college students. Even in the current era, young women reach sexual maturity under the backdrop of the sexual double standard, in which sex outside of a committed relationship is stigmatized. For men, losing their virginity is often a positive event (Holland, Ramazanoglu, Sharpe, & Thomson, 2000), but for women the initiation of sexual activity may be associated with loss, guilt, regret, and other negative affective experiences (Moore & Davidson, 1997; Resneck-Sannes, 1991; Schwartz, 1993; Travis & White, 2000). College-aged may need to protect their reputation, and may be realistically concerned about the high prevalence of STDs within their cohort (Aral, 2001; Bustamante, 1993), to which women are more susceptible than men. They may therefore be more conservative in their actual or reported behavior. Women are more at risk than men for childhood sexual abuse (CSA), sexual assault, and intimate partner violence, with estimates ranging between 15% and 34% for CSA alone (Spitzberg, 1999; Vogeltanz et al., 1999). These events often have a detrimental impact on women?s sexuality and mental health (Coker et al., 2002; Cloutier, Martin, & Poole, 2002).|../mac/prof328.html",
    "Sociocultural Factors in Conceptual Change: Tracking Mental Models in A Com...||Member Abstract Track Sociocultural Factors in Conceptual Change: Tracking Mental Models in A Community-Based Parent Education Program L. Fernando Romero , Division of Psychology in Education, Arizona State University fer@asu.edu Sarah K. Brem , Division of Psychology in Education, Arizona State University sarah.brem@asu.edu The National Academy of Sciences and the Institute of Medicine (NAS/IOM, 2000) argue for disseminating neuroscientific and psychological research to caregivers, citing misconceptions, debates raging without evidence, and a tendency to treat “nature” and “nurture” as mutually exclusive. Altering caregiver reasoning and behavior requires creating new mental models of the brain and mind. NAS/IOM (2000) reported that existing interventions fail in accommodating sociocultural factors, and in empowering family decision making. Furthermore, the clearest benefits of caregiver knowledge and skills have been to children disadvantaged by such events as brain injury (Kinsella, Ong, Murtagh, Prior & Sawyer, 1999), and neglect or abuse (Bakermans-Kranenburg, Ijzendoorn, & Juffer, 2003); Given that such trauma is often emotionally challenging, the most important caregiver models may be constructed in the context of strong social and affective pressures. We examined the mental models of caregivers participating in a program on infant brain development, representing not only the domain content of their models, but also sociocultural and affective factors. Affect and cognition interact in the experience of regret (Roese, 1997; Gilovich & Medvec, 1995), and dissonance (Elliot & Devine, 1994). Sociocultural factors can affect inquiry (Brodie, Kjellson, Hoff & Parker, 1999), attribution and emotional regulation (Chavira, Lopez, Blacher & Shapiro, 2000), and reasoning (Nisbett, Peng, Choi & Norenzayan, 2001). Our goal is to enhance theoretical models of conceptual change, while tracking family outcomes.|../mac/prof329.html",
    "Construction and use of cultural environment structure through simple decis...||Publication-Based Talk Track Construction and use of cultural environment structure through simple decisions P.M. Todd Cognitive mechanisms are shaped by selection to match their environments. But through their use, these mechanisms exert a shaping force on their surroundings as well. These forces are particularly evident when a population of decision-makers create, through their combined decisions, the cultural environment that they share. What happens to the usefulness of these cognitive mechanisms when they alter the environment they are applied to? Here we explore this cognition-environment interaction by looking at how using a very simple cognitive mechanism, the recognition heuristic for making choices, can result in some objects in the constructed cultural environment being much more often recognized and “talked about” than others. An analytic model and an agent-based simulation are used to show what behavioral factors affect the emergence of this environmental structure. We start with a very simple cognitive mechanism that aims to select the “bigger” of two options, where “bigger” is defined on some unobservable criterion dimension. This decision can be made even simpler when the only information available is whether or not each option has ever been encountered before. In this case, the decision maker can do little other than to rely on his or her own partial ignorance, e.g. choosing recognized options over unrecognized ones. This kind of “ignorance-based reasoning” is embodied in the recognition heuristic: When choosing between two objects (on some criterion), if one is recognized and the other is not, then select the former. This heuristic works well—that is, it is ecologically rational—when the environment is structured in such a way that “bigger” objects (that is, those that score higher on the criterion in question) are more recognized that “smaller” ones. It turns out that many environments are structured this way—people tend to talk about prominent landmarks, individuals, and cultural events (e.g., the tallest mountains, richest billionaires, or biggest-selling movies) and hence recognize these more than other, “smaller,” ones—which in turn means that they can use the recognition heuristic to accurately judge which things are bigger. What happens when a population of decision-makers uses the recognition heuristic to choose items in their world, and talks to each other about what they have chosen? We have created both analytical models and agent-based simulations to find out. Over time, recognition of some objects spreads disproportionately through the population, resulting in the emergence of a J-shaped distribution of which objects are selected most often. Such J-shaped distributions (where a very few objects have very high rates of being chosen, and most objects have quite low rates) appear commonly in real-world domains that people think about, talk about, and make decisions about, such as records sold, movies watched, brands purchased, and the like. Such environment structure can in turn make the recognition heuristic ecologically rational, that is, able to make beneficial choices. In this talk, I will present the models we have developed and use them to show how salient cultural environment structure can emerge from the interactions of simple decision makers, and show how this structure in turn impacts the decisions that are made in this environment.|../mac/prof33.html",
    "Supporting Learning from Worked-Out Examples in Computer-Based Learning Env...||Member Abstract Track Supporting Learning from Worked-Out Examples in Computer-Based Learning Environments Julia Schuh , Virtual Ph.D. Program: Knowledge Acquisition and Knowledge Exchange with New Media j.schuh@iwm-kmrc.de Peter Gerjets , Multimedia and Hypermedia Research Unit, Knowledge Media Research Center p.gerjets@iwm-kmrc.de Katharina Scheiter , Department of Applied Cognitive Psychology and Media Psychology, University of Tuebingen k.scheiter@iwm-kmrc.de Gerjets, Scheiter, and Tack (2000) demonstrated that learners experience serious difficulties in utilizing instructional examples according to their profitability when learning with a nonlinear hypertext environment. In two experimental studies we examined two possible causes of these difficulties and investigated different instructional methods for improving learners’ example utilization when interacting with a learning and problem-solving environment in the domain of combinatorics. (Incomplete Examples with Feedback) A second possible cause of insufficient example utilization refers to the fact that learners may suffer from illusions of understanding when learning from worked-out examples – resulting in scarce example retrieval in hypertext environments. To prevent learners from such illusions we introduced an instructional condition with fragmentized example solutions and allowed learners to complete the gaps by selecting|../mac/prof330.html",
    "Cortical Oscillations During Memory Encoding Predict Successful Retrieval...||Member Abstract Track Cortical Oscillations During Memory Encoding Predict Successful Retrieval Per Sederberg , Volen Center for Complex Systems; Brandeis University sederberg@brandeis.edu Joseph Madsen , Department of Neurosurgery; Childrens' Hospital Boston joseph.madsen@TCH.Harvard.edu Michael Kahana , Volen Center for Complex Systems; Brandeis University kahana@brandeis.edu We examined the neurophysiological correlates of memory encoding that lead to successful retrieval. Previous work has demonstrated that changes in in- tracranial ERPs and fMRI BOLD signals predict subsequent recall in episodic memory tasks (Fer- nandez et al., 1999; Wagner et al., 1998). Recent work has shown that in free recall, 32-48Hz (gamma) phase synchrony in the medial temporal lobe (MTL) during study predicts subsequent recall (Fell et al., 2001). To explore the role of a wide range of oscilla- tions outside of the MTL, we tested whether changes in oscillatory activity at various frequencies and at widespread cortical sites during encoding predict successful episodic recall. Recording from 675 widespread cortical sites in 7 patients undergoing treatment for medically resis- tive epilepsy, we examined oscillatory power between 2 and 64Hz as participants studied lists of com- mon nouns. At many frequencies we found oscilla- tions that predicted subsequent recall. We observed increased 32-64Hz (gamma) power associated with successful encoding of list items at 31 sites. We also observed decreased 9-16Hz (alpha) power predictive of subsequent recall at 46 sites. Sites exhibiting these two patterns at di erent frequency bands often appeared in the same brain region. Within a single frequency band, however, electrodes exhibiting in- creases and decreases in power that predicted subse- quent memory clustered in topographically distinct regions. Electrodes exhibiting increases in gamma oscillations that positively correlated with subse- quent recall were found at many cortical locations, but especially in the temporal lobe and subtempo- ral occipital region. Electrodes exhibiting decreases in alpha oscillations that predicted successful recall localized to the left and inferior regions of both the temporal and occipital lobes. These ndings point to a crucial role of brain oscillations in episodic mem- ory function.|../mac/prof331.html",
    "Extending the Response-Time “Guilty Knowledge” Test...||Member Abstract Track Extending the Response-Time “Guilty Knowledge” Test Travis L. Seymour , Department of Psychology, University of California Santa Cruz Jess Kerlin , Department of Psychology, University of California Santa Cruz Anna Kurtz , Department of Psychology, University of California Santa Cruz Sometimes it is important to determine another’s familiarity with privileged information despite their intention to conceal this knowledge. Following previous work showing that recognition memory can be used to index knowledge activation (Farwell & Donchin, 1991; Lykken, 1981), Seymour, Seifert, Mosmann, & Shafto (2000) proposed a new “Guilty Knowledge” Test (GKT) based on the response-time (RT) and accuracy to critical information. In the GKT paradigm, participants are first exposed to critical “Probe” or crime information. Following a delay, and in an ostensibly novel task, participants learn a new set of “Target” phrases (e.g., “Blue Coat”) and are given a speeded recognition task. When presented with Targets, participants are to press a button indicating “Old.” When presented with new “Filler” phrases as well as familiar Probes, participants are to respond with a button indicating “New.” If Probes are familiar, participants may respond differentially to Probes and Fillers. However, when Probes are unfamiliar no differential response is expected. Despite being motivated to respond identically to Probes and Filler, and thus masking their knowledge of the Probes, participants’ responses to Probes were reliably slower and less accurate than to Filler items. This “Guilty Knowledge Effect” (GKE) did not occur when participants responded during a block where Probes are unfamiliar (Seymour et al., 2000). This differential pattern of results for Probes and Fillers led to a high hit rate and low false alarm rate when compared to other test using physiological measures such as heart rate, or more direct measures such as EEG (Farwell & Donchin, 1991).|../mac/prof332.html",
    "Teaching Cognitive Science through Collaborative Reflection (2): A Case of ...||Member Abstract Track Teaching Cognitive Science through Collaborative Reflection (2): A Case of Learning Semantic Net Representation Hajime Shirouzu , School of Computer and Cognitive Sciences, Chukyo University shirouzu@sccs.chukyo-u.ac.jp Naomi Miyake , School of Computer and Cognitive Sciences, Chukyo University nmiyake@sccs.chukyo-u.ac.jp We have been developing and testing a set of undergraduate curriculum to teach cognitive science [1]. Here we report one particular case of our approach where students learned some basic constructs of human memory. Students gradually integrated pieces of research findings on the “semantic net representation” by the “jigsaw” method, by tying the understanding to their previous experience of analyzing data from classic psychology experiment [2][3]. Learning through collaborative reflection enabled them to clearly understand the reasons why people remember semantic aspects of sentences better than their superficial features and the significance of the results when they were to apply it in real-world problem solving. Students also gained meta-cognitive experiences of becoming an expert on a piece of literature assigned to them and of actively grasping its main points. These experiences provided the students with the base to engage in more rigorous constructive interaction in the latter phase of the curriculum.|../mac/prof333.html",
    "Analogy as a Tool to Enhance Innovative Problem Solving...||Member Abstract Track Analogy as a Tool to Enhance Innovative Problem Solving Cynthia M. Sifonis , Oakland University sifonis@oakland.edu Francis H. Chen , General Motors R&D Center, General Motors Corporation francis.h.chen@gm.com Dianne Bommarito , General Motors R&D Center, General Motors Corporation dianne.bommarito@gm.com Current advances in markets, technology, and information occur at an exponential rate and require adjustments at all levels of an organization. This is requiring industry veterans to “think outside the box.” General Motors (GM) has tackled the problem of thinking innovatively by applying knowledge gained from the research in psychology to develop innovative problem solving workshops. Though relatively little empirical research has focused on the use of analogy to enhance creativity, quite a bit is known about the processes involved in analogical problem solving. Consequently, GM was able to use this knowledge to develop Cross Domain Analogical Analysis (CDAA) workshops. In the workshops, participants receive detailed information about multiple source domains. They are encouraged to map between the source and target domains and to list any insights they have generated. A case study of this approach reveals that workshop participants generate multiple insights about the target domain. The greater the number of source domains to which a person has been exposed, the greater the number of solutions they generate to the target problem. Of specific interest to the current study is the degree to which the methods used in the CDAA workshops can be investigated in a more controlled environment. Will university students taught to use analogical problem solving and given specific source domains with which to work perform similarly to industry veterans using the same technique? We will also be investigating the effect of conceptual distance between the source and target domains on the solutions generated. Based on the existing literature (Dahl & Moreau, 2002), we predict that conceptually distant source and target domains will result in more creative solutions to the target problem.|../mac/prof334.html",
    "Multiple Goal Facilitation in a Rule Discovery Task...||Member Abstract Track Multiple Goal Facilitation in a Rule Discovery Task Devon Skelton , Department of Psychology, Bowling Green State University tussin21@hotmail.com Ryan D. Tweney , Department of Psychology, Bowling Green State University tweney@bgnet.bgsu.edu he present study sought to expand previous research on Wason?s 2-4-6 task by exploring the limits of the multiple goal facilitation effect. In the original task subjects were given an initial triple (set of three numbers), 2-4-6, and told to generate their own sets of triples to aid them in finding the experimenter?s rule, ?any ascending sequence.? While the task seems simple, on a first announcement only about 20% of subjects found the correct rule. Paradoxically, instructions requiring subjects to find two rules (?ascending sequence? = ?DAX?, ?anything else? = ?MED?) resulted in a dramatic increase in success rate; 60% of the subjects solved the rule on a first announcement (Tweney, Doherty, Worner, Pliske, Mynatt, Gross, & Arrkelin, 1980). Gale and Ball (2002) reviewed two possible explanations; first, that the increase in success rate could be attributed to the mutually exclusive relation between the two rules, or, second, that the modified task helps avoid ?positivity-bias? in that subjects focus on the positive label of the second rule rather than on the negative label ?does not fit?. Gale and Ball tested the contrasting explanations and found that while feedback labeling had little or no effect on the likelihood of success, presence of multiple goals helped as in the original ?DAX-MED? findings. The results supported the ?goal-complementarity? explanation. The initial finding, that success rates increase with the addition of a second labeled rule, is seemingly paradoxical because increasing task load seems to facilitate performance. Shouldn?t the greater complexity of a multiple goal task cause a decline in performance? To test this, we conducted three experiments in which subjects had to seek three goals. We expected to find a decline in performance when a third rule, ?ZIF?, was introduced.|../mac/prof335.html",
    "The Role of Perceptual Concreteness in Analogical Mapping and Transfer...||Member Abstract Track The Role of Perceptual Concreteness in Analogical Mapping and Transfer Ji Y. Son J. Son, Department of Psychology, Indiana University jys@indiana.edu Robert L. Goldstone , Department of Psychology, Indiana University rgoldsto@indiana.edu Abstract ideas are often grounded in concrete representations such as contextualized examples or visual aids. Relatively subtle aspects of these concrete representations affect how the underlying abstraction is learned and transferred to other domains (DeLoache, 1995; Kotovsky & Gentner, 1996). In a study by Goldstone and Sakamoto (in press), participants showed better understanding of a superficially dissimilar transfer situation when the initial learning situation was presented with idealized drawings rather than concrete pictures. Concreteness seems to bind the abstract concepts more tightly to the domain being learned, thereby making transfer difficult. This study examined how concreteness interacts with analogical mapping and transfer. Participants saw a tutorial on signal detection theory (SDT) embedded in a story about a doctor who is diagnosing patients as sick or “healthy by looking at their cell distortion level. Later on, he finds out which of his patients are actually sick or healthy. In traditional SDT terms, the cell distortion is the stimulus, the diagnosis is the response, and sick and healthy people are signal and noise respectively. Participants saw a presentation of the doctor story with either richly detailed pictures of cells (rich condition) or simple circles representing cells (idealized condition). After taking a tutorial quiz about the doctor situation, the participants were told about a university admissions committee deciding which students to accept or reject on the basis of their SAT scores. Later on, the committee receives feedback on whether these students went on to have good GPAs or bad GPAs at their respective colleges. A transfer quiz was administered to test participants’ understanding of the underlying mechanisms of SDT in the university admissions context. After the transfer quiz, the students were told that the two situations are analogous and were asked about their relationship with an analogy quiz.|../mac/prof336.html",
    "Imitation, Autism & ToM...||Member Abstract Track Imitation, Autism & ToM Francys Subiaul , Columbia University Herbert Lurie , Columbia University David Holmes , The Eden Institute, Princeton University Tovah Klein , Barnard College of Columbia University Ralph Holloway , Columbia University Herbert Terrace , Columbia University The ability to learn from others is probably one of the most important cognitive adaptations of our species. Children 72 hours old can copy specific facial expressions (Meltzoff and Moore, 1977) and one-year olds can replicate simple sequences (Bauer, 1996). But children with autism may represent a special case. Autism spectrum disorder (ASD) is characterized by linguistic and social impairments (Happé, 1994), which may deprive these children of the kinds of benefits healthy children derive from imitation.|../mac/prof337.html",
    "Reconfiguration and Inertial Processes in Attention Switching during Readin...||Member Abstract Track Reconfiguration and Inertial Processes in Attention Switching during Reading Marlene R. Taube-Schiff , Department of Psychology, and the Centre for the Study of Learning and Performance, Concordia University marlene_taubeschiff@yahoo.ca Norman Segalowitz , Department of Psychology, and the Centre for the Study of Learning and Performance, Concordia University segalow@vax2.concordia.ca Shifting or switching attention from one task to another involves a cost in cognitive processing (Rogers & Monsell, 1995). Different paradigms have been developed to investigate the mechanisms that may underlie such switch costs. For example, Rogers and Monsell (1995) developed the alternating runs paradigm requiring responses to two different tasks that alternate (…AABBAA…) so as to create a predictable sequence of repeat trials and switch trials. They found reaction times on switch trials were slower than on repeat trials, even with a long delay between a response and the upcoming stimulus (residual switch cost). This result was hypothesized to be compatible with the existence of an endogenous “reconfiguration” process, a stage of preparation for the predictable, upcoming trial. Wylie and Allport (2000) created a 3-phase version of the alternating runs paradigm. Phases varied according to whether stimuli were monovalent (affording performance of only one task) or bivalent (affording performance on either task even though only one was appropriate). In Phase 1, all trials were monovalent, in Phase 2 half were bivalent and half monovalent, and in Phase 3 all were bivalent. This design allowed comparison of performance on dissimilar switch trials following similar repeat trials (switch to monovalent versus bivalent trials from monovalent repeat trial), and performance on similar switch trials following dissimilar repeat trials (switch to bivalent from a monovalent versus bivalent repeat trial). Wylie and Allport found that performance on switch trials depended on whether the preceding repeat trial was monovalent or bivalent, and that performance on monovalent versus bivalent switch trials did not vary when the nature of the preceding repeat trial was held constant. They concluded that switch costs reflect inertial inhibitory processes persisting from the previous trial that interfere with and hence slow processing needed for the current trial. Recently, Carrière (2002) in our lab found that both reconfiguration and inertial mechanisms may be implicated in switch costs involving simple, decontextualized stimuli. It is not known, however, whether these processes play a role in more complex activities. We studied this question using a reading activity involving sentence-like materials.|../mac/prof338.html",
    "Evaluating Geographic Visualization Tools and Learning about User Tasks...||Member Abstract Track Evaluating Geographic Visualization Tools and Learning about User Tasks Carolina Tobón , Department of Geography, University College London c.tobon@ucl.ac.uk Computer environments for visual data exploration are commonly developed to investigate plausible scenarios or questions that prompt the discovery of relations or patterns that are useful. The goal of simultaneously representing data (geospatial or otherwise) in a number of graphical forms is typically to support a process of hypothesis formation and knowledge construction (Gahegan 2001). However, a clear definition of users’ goals and tasks that facilitates the evaluation of software tools or concepts is not always possible due to the exploratory nature of visualization. In addition, visualization techniques can be pertinent for a wide range of applications, users and data types. At present, this implies that evaluations must be custom tailored to each system and the domain dependent tasks it is designed to support, which makes the generalization of results about (geo)visualization techniques difficult in many cases.|../mac/prof339.html",
    "On the Perceptual and Neural Basis of Event...||Publication-Based Talk Track The Role of Motion in Segmenting Simple Events Jeffrey M. Zacks , Psychology Department, Washington University jzacks@artsci.wustl.edu How do we represent what is happening around us? In particular, how do observers perceive and understand the temporal organization of everyday, goal-directed activities? From making coffee to performing a tonsillectomy, people seem to talk about continuous activity in terms of discrete parts that are hierarchically organized. Evidence from four sources indicates that such talk is no accident; rather, people integrate bottom-up perceptual cues with top-down information about intentions to understand events. First, studies of perception indicate that everyday events such as making the bed or doing the dishes are encoded in terms of hierarchical part-subpart relations: When observers are asked to segment movies of these activities into parts at coarse and fine temporal grains, they spontaneously do so in terms of hierarchical relationships between parts and sub-parts. This perceptual structure appears to be preserved in memory. Recent data strengthen the view that motion cues play a key role in identifying event segment boundaries. Second, neuroimaging studies suggest that event segmentation is an ongoing component of perception, subserved by specialized neural substrates. In one study, participants passively viewed movies of everyday activities during functional MRI scanning and later segmented the same movies into meaningful coarse and fine units. A network of regions including posterior extrastriate cortex and precentral cortex showed transient increases at those moments later identified as perceptual event boundaries. Notable in the posterior regions was the MT complex, a region known to be specialized for processing motion information. Throughout the network, increases were larger for coarse boundaries than fine boundaries, and increased when participants deliberately segmented the activity. Third, psycholinguistic production data suggest that information about event structure is spontaneously encoded into language. Descriptions of coarse-grained and fine-grained event parts differ systematically in their syntax: Coarse-grained descriptions pick out objects more precisely, whereas fine-grained descriptions are more precise regarding actions. Within descriptions of fine-grained units, those that occur near coarse boundaries take on some of the syntactic properties of coarse-grained descriptions. Thus, speakers implicitly convey information about hierarchical event structure within running descriptions of fine-grained events. Finally, studies of narrative understanding demonstrate that event boundaries can guide working memory updating. Narrative time shifts such as “an hour later” serve as cues to readers that an event boundary has been encountered. After such phrases, anaphors to information presented before the event boundary are processed more slowly, and recognition memory for nouns presented before the boundary is less accurate. Together, these results support the view that we automatically and actively encode events in terms of structured representations that capture recurring features of activity such as goals, roles and causal influence. The human perceptual system appears to include specialized routines for processing such information, possibly implemented by specialized neural systems, leading to active structuring of the perceptual information. This perceptual structure appears to guide downstream processing, including linguistic production and narrative understanding.|../mac/prof34.html",
    "How Novices Reason About Anomalies...||Member Abstract Track How Novices Reason About Anomalies Susan B. Trickett , Department of Psychology, George Mason University stricket@gmu.edu Thomas Watkins , Department of Psychology, George Mason University twatkin1@gmu.edu J. Gregory Trafton , Naval Research Laboratory trafton@itd.nrl.navy.mil Theories that address how people respond to anomalous data either explicitly state or imply that anomalies are processed differently from non-anomalous data. (Chinn & Brewer, 1992; Alberdi, Sleeman, & Korpi, 2000; Trickett, Trafton, Schunn, & Harrison, 2001). However, studies of reasoning about anomalies do not provide an experimental basis for comparison between responses to anomalous versus non-anomalous data. We investigated differential reasoning processing by manipulating whether data met (non-anomaly) or violated (anomaly) expectations. to differentially process large and small anomalies, or indeed, between anomalous and non-anomalous results. We collapsed the results across condition for the remaining analyses (see Table 1). Participants made little use of the information in the story, either at prediction or at explanation, relying on PE (especially at prediction) and CS (especially at explanation). Table 1 shows interesting shifts in strategy use in different phases of reasoning. Initial heavy reliance on PE at prediction decreased significantly at explanation, C2(1) = 3.8, p = .05. Use of CS increased|../mac/prof340.html",
    "When Problem Solving Leads to Impaired Structural Knowledge...||Member Abstract Track When Problem Solving Leads to Impaired Structural Knowledge David L. Trumpower , University of New Mexico, Department of Psychology dtrumpow@unm.edu Timothy E. Goldsmith , University of New Mexico, Department of Psychology gold@unm.edu Bryan J. Williams , University of New Mexico, Department of Psychology asnjoe@unm.edu Structural, or conceptual, knowledge is defined as the knowledge of concepts and their interrelationships with one another. Although it has been shown that structural knowledge is predictive of expertise (e.g., Goldsmith, Johnson, & Acton, 1991), questions remain about how it develops, especially with respect to procedural knowledge. For example, Rittle-Johnson, Siegler, and Alibali (2001) propose that procedural and structural knowledge develop in an iterative fashion, with increases in one leading to increases in the other. The current studies, however, explore a situation in which procedural training leads to better problem solving performance but impaired structural knowledge.|../mac/prof341.html",
    "Collaborative learning from an analogy...||Member Abstract Track Collaborative learning from an analogy Michael Tscholl , Department of Computer Science, University College London m.tscholl@cs.ucl.ac.uk John Dowell , Department of Computer Science, University College London j.dowell@cs.ucl.ac.uk Analogy is an important mechanism of learning: by implicitly or explicitly comparing examples, people are able to abstract common structural features (cf. Gentner, 1989) or generalize over them (Ross & Kennedy, 1990). Our study investigated how analogical learning occurs in a group using complex material. We identify a correlation of roles with specific analogical learning processes, and related to that a repeated cycle of analogical learning with apparently varying learning outcomes. The study shows that individual contributions can be differentially associated with the component processes of analogical reasoning in a group. A schema abstracted by one individual was subsequently exploited by the others. Further, a variability in learning outcomes over the different phases of the exercise is evident: learning from an analogy will be more proficient when, as in the early phases, transfer makes use of an abstracted schema.|../mac/prof342.html",
    "Artifacts of Memory: Michael Faraday and the Construction of Meaning...||Member Abstract Track Artifacts of Memory: Michael Faraday and the Construction of Meaning Ryan D. Tweney , Department of Psychology, Bowling Green State University tweney@bgnet.bgsu.edu Christopher D. Ayala , Department of Psychology, Bowling Green State University ayalac@bgnet.bgsu.edu In a series of recent papers, our research group has been examining a large number of surviving microscope slides and other specimens made in 1856 by Michael Faraday (1791-1867). The research has centered on the way Faraday used these specimens to formulate and guide his research endeavors (Tweney, Mears, & Spitzmüller, in press). Faraday’s microscope slides were shown to be epistemic artifacts; “agentive” participants in his research (Tweney, 2002). Such a finding extends recent research on cognitive artifacts (e.g., Zhang & Norman, 1994), as well as existing cognitive accounts of Faraday’s experimental practices (e.g., Gooding, 1990). The present study re-examines an aspect of the research diaries kept by Faraday throughout his career. These diaries became so extensive (recording literally tens of thousands of experiments, speculations, and so on) that sophisticated finding aids were necessary. We emphasize the role of these aids as epistemic artifacts in their own right. That is, we seek to demonstrate that laboratory notebooks, like Faraday’s microscope slides, were agentive devices whose use went beyond the stereotypical role of notebooks as passive repositories of factual information.|../mac/prof343.html",
    "The Use of General and Specific Counterexamples in Conditional Reasoning...||Member Abstract Track The Use of General and Specific Counterexamples in Conditional Reasoning Niki Verschueren , Laboratory of Experimental Psychology University of Leuven Niki.Verschueren@psy.kuleuven.ac.be Walter Schaeken , Laboratory of Experimental Psychology University of Leuven Walter.Schaeken@psy.kuleuven.ac.be Géry d’Ydewalle , Laboratory of Experimental Psychology University of Leuven Géry.Dydewalle@psy.kuleuven.ac.be Conditional inferences are simple problems based on ‘ifthen’- rules. We will focus on everyday causal conditionals of the form ‘if cause, then effect’. For instance: 1. If you flip the light switch, then the light goes on. You flip the switch. Does the light go on? 2. If a dog has fleas, then it will scratch itself. A dog scratches itself. Does it have fleas? It is repeatedly found that reasoners base their conclusion on the retrieval of information from memory rather than on the formal/logical structure of the argument. Reasoners often use content-information -such as counterexamples- to draw inferences. A counterexample (CE) provides an alternative conclusion for the suggested one. If reasoners think of a CE, they do not accept the given conclusion. A CE can be formulated in a general or specific way. For instance: 1. General: the switch is flipped without the light going on Specific: there can be a short-circuit 2. General: there are other reasons for scratching Specific: the dog can have a skin disease The CE retrieval mechanism is mainly based on semantic association. De Neys, Schaeken and d’Ydewalle (2002) showed that the retrieval process does not stop after retrieval of a single CE. Markovits and Barrouillet (2002) state that when multiple specific CE’s are retrieved, reasoners will summarize them in a general CE for reasons of cognitive economy. Hence we should expect that it is more likely to observe general CE’s on sentences for which it is easy to retrieve CE’s, then for sentences with few CE’s. We will verify this assumption by means of a verbal protocol study.|../mac/prof344.html",
    "Multilevel Computational Modeling of Human Attentional Networks...||Member Abstract Track Multilevel Computational Modeling of Human Attentional Networks Hongbin Wang , School of Health Information Sciences, University of Texas Health Science Center at Houston Hongbin.Wang@uth.tmc.edu Jin Fan J. Fan, Sackler Institute for Developmental Psychobiology, Weill Medical College of Cornell University jif2004@med.cornell.edu Suppose a student S was asked to solve the equation “2x + 3 = 9”. After 2 seconds, he gave the answer “x = 3”. Both cognitive scientists A and B were interested in understanding how S did it. Scientist A recorded S’s detailed verbal protocol, based on which, and other relevant behavioral measures, A hypothesized the possible knowledge structure underlying S’s problem solving and developed a symbolic computational model that simulated the process. On the other hand, scientist B adopted sophisticated brain imaging techniques including electroencephalograph (EEG) and functional Magnetic Resonance Imaging (fMRI) and acquired a high-resolution recording of S’s brain dynamics during problem solving. Based on some well-supported neural computing principles, B then developed a biologically realistic connectionist model to simulate the brain activities underlying S’s performance. Though both models fitted the data well, the two models are clearly different. While the symbolic model offers a description of the process with psychological plausibility and high behavioral relevance, the connectionist model emphasizes the process’ biological realism and brain foundations. One question is, do we, cognitive scientists who endeavor to discover unified theories of cognition, have justifiable reasons to prefer one to another?|../mac/prof345.html",
    "Exemplar-Based Visual Discrimination and Categorization in Chickens and Imp...||Member Abstract Track Exemplar-Based Visual Discrimination and Categorization in Chickens and Implementation of an Autonomous Agent Model Christian W. Werner , C. und O. Vogt Institut für Hirnforschung, Universitätsstr. wernerc@uni-duesseldorf.de Roul Sebastian John , Institute of Cognitive Science, Katharinenstr. rjohn@uos.de Generally, processing of visual stimuli is explained by decomposition of stimuli into characteristic and/or defining elements or features (feature analysis). An animal, when learning relations between stimuli and responses, is assumed to form associations between elements of the stimuli and the responses. In contrast, an exemplar-based model may propose the representation of stimulus situations as whole configurations, stored in so called exemplars. These exemplars then may serve as units for association with a response. does not need to assume cognitive processes like “feature analysis” and “selective attention”. The autonomous agent model In addition, we are implementing these theoretical assumptions into an autonomous agent model. The advantage of this modeling approach is that one can directly test the adequacy of theoretical assumptions by measuring the performance of the agent in the same way as was done with the chickens before (same dependent variable, same|../mac/prof346.html",
    "Learning Free Word Associations from Texts...||Member Abstract Track Learning Free Word Associations from Texts Manfred Wettler , University of Paderborn wettler@psycho.upb.de Petra Seidensticker , University of Paderborn karas@psycho.upb.de|../mac/prof347.html",
    "SIAM-LSA: An interactive activation model of sentence similarity...||Member Abstract Track SIAM-LSA: An interactive activation model of sentence similarity Peter Wiemer-Hastings , DePaul University, School of Computer Science peterwh@cti.depaul.edu The ability to assess the similarity of objects in the world is fundamentally important to our survival. A variety of theories have been proposed for modeling human similarity judgments. Most of these theories involve comparing the sets of features of di erent objects to determine the overlap between them. But most of the theories also ignore the structure of the objects and relationships between the parts. Goldstone's SIAM system uses a connectionist architecture to create correspondences between objects and their features in di erent scenes (Goldstone, 1994). Excitatory connections reinforce coherent mappings between objects (e.g. ObjectA to ObjectC and ObjectB to ObjectD). Inhibitory connections ght against redundant or contradictory mappings (e.g. ObjectA to ObjectC and Object A to ObjectD). Likewise, connections between the features of objects either support or inhibit each other and the corresponding object-object connections. Siam's connectionist architecture allows it to take into account the structure of the scenes and the objects as well as the similarity of the features. Recent language processing research has used similarity judgments between sentences as a method for \\understanding student utterances in an intelligent tutoring system (Wiemer-Hastings, Wiemer- Hastings, & Graesser, 1999). The system matches what the student says to its set of expected answers for the current question. A signi cantly high match indicates that the student has addressed that particular answer. Latent semantic analysis (LSA) uses a corpus-derived vector representation to compare texts. LSA takes no account for word order or sentence structure. By combining a simple surface parse with comparison of components, structured LSA (SLSA) has produced improved similarity judgments (Wiemer-Hastings & Zipitria, 2001). The current research merges the SIAM approach with SLSA to compare texts based on a connectionist representation of the correspondences between their components.|../mac/prof348.html",
    "Analogical Transfer in the Perspective of Insight Problem Solving: A Study ...||Member Abstract Track Analogical Transfer in the Perspective of Insight Problem Solving: A Study on Duncker’s radiation problem Tsunhin J. Wong , Department of Psychology, The University of Hong Kong thjwong@hkucs.org Albert W. L. Chau , Department of Psychology, The University of Hong Kong awlchau@hkucc.hku.hk The primary aim of this study is to provide a new perspective to look at the transfer of analogical problem solving in terms of the theory of contemporary studies on insight (Knoblich, 1999). Many research in analogical problem solving made use of insight problems, for example Duncker’s radiation problem and weight-the-elephant problem (Gick & Holyoak, 1980; Chen, 2002). However, most of the research was on the similarity between base problem and target problem in terms of structure and superficial features; few had taken an approach to address these problems and the transfer between them in terms of insight problem solving theories. In this study, transfer was defined in terms of constraint relaxation and chunk decomposition, the processes that Knoblich et al (1999) used to describe the problem solving performance of matchstick arithmetic problems. Subjects were predicted to be more able to solve the target problem if they were exposed, in the base problem, the same type of constraint to be relaxed and/or the same type of chunk to be decomposed that were necessary for solving the target problem.|../mac/prof349.html",
    "The Interaction between Informational and Computational Properties of Exter...||Paper Track The Interaction between Informational and Computational Properties of External Representations on Problem-Solving and Learning Shaaron Ainsworth , School of Psychology, University of Nottingham Shaaron.Ainsworth@nottingham.ac.uk Gareth Peevers , School of Psychology, University of Nottingham gjp@psychology.nottinghm.ac.uk This paper reports an experiment that explores if the way instructions for operating a complex device are represented influences problem-solving and learning about the task. Instructions were presented in one complex representation or in multiple simpler ones. The form this information took was tabular, diagrammatic or textual. Participants found the optimal solution more often when given instructions in text representations or in a single representation. However, the single text representation was associated with significantly slower performance. Participants recalled more about the task with text representations, irrespective of how the information was distributed. This experiment confirmed that representations that display instructions in such a way as to increase the cost of operating with them can paradoxically lead to better performance.|../mac/prof35.html",
    "Processing of proper names in Mandarin Chinese...||Member Abstract Track Processing of proper names in Mandarin Chinese Huei-Ling Yen , Experimental Neurolinguistics Group, Faculty of Linguistics University of Bielefeld huei-ling.yen@uni-bielefeld.de Horst M. Müller , Experimental Neurolinguistics Group, Faculty of Linguistics University of Bielefeld hmueller@uni-bielefeld.de The distinction between concrete and abstract nouns has been postulated since several thousand years in ancient Greek and in Chinese language philosophy, (e.g., Kripke, 1972; Jiang, 1993). Furthermore, within the category of concrete nouns the proper names are considered as a distinct subgroup. Such linguistic categories may reflect cognitive structures since they are common in all languages - or they may be completely artificial (Müller & Kutas, 1996). The theoretical assumptions that proper names have a cognitive reality are supported by recent case studies (e.g., Fukatsu et al., 1999) and experimental studies in English and German (Schuth, Werner & Müller, 2002; Weiss & Müller, 2003). To investigate processing of proper names and common nouns in a non Indo-European language, we conducted an experiment with Mandarin Chinese (lexical decision task).|../mac/prof350.html",
    "The possible-word constraint in Cantonese speech segmentation...||Member Abstract Track The possible-word constraint in Cantonese speech segmentation Michael C. W. YIP , School of Arts & Social Sciences, The Open University of Hong Kong myip@ouhk.edu.hk Speech segmentation is a central issue of spoken language comprehension research (Cutler, 2001). And, recently, one important solution comes from the discovery of a mechanism operated in our lexical system, the Possible-Word Constraint (PWC). In their word-spotting experiments, Norris, McQueen, Cutler and Butterfield (1997) observed that listeners usually found it more difficult to spot the real word apple in the nonsense word string fapple than in vuffapple. Both f and vuff are not words in English, but only the latter one could be a possible word in the sense that the word vuff generally satisfied all the phonological conditions for becoming a word. The consonant of f would never been a possible word under any circumstances. Hence, Norris et al. (1997) proposed that our lexical system would be sensitive to this kind of discrepancy and inhibit the activation of those words that stranded an “impossible word candidate” residue in the speech signal during speech processing. This procedure was called the Possible Word Constraint (PWC). However, realization of this mechanism has been done in English (Norris, McQueen, Cutler, Butterfield & Kearns, 2001); Japanese (McQueen, Otake, & Cutler, 2001) and Sesotho (Cutler, Demuth & McQueen, 2002) so far. Hence, the present study aims to further examine the efficiency of the PWC in Cantonese, a language that differs significantly from most Indo-European languages (English, Dutch, French) in its use of lexical tones, phonotactic structure and its special morphemic nature of words.|../mac/prof351.html",
    "The Use of Pronominal Case Information in Sentence Interpretation...||Member Abstract Track The Use of Pronominal Case Information in Sentence Interpretation Yuki Yoshimura , Department of Modern Languages, Carnegie Mellon University yyuki@cmu.edu Brian MacWhinney , Department of Psychology, Carnegie Mellon University macw@cmu.edu Cross-linguistic research in the Competition Model framework (MacWhinney & Bates, 1989) has shown that sentence interpretation is driven by surface grammatical markers in accord with their relative cue strength. For example, word order is the strongest cue in English sentence processing, because it is also the most reliable and available cue in English corpora. On the other hand, word order is not a strong cue in Spanish and Japanese, because the reliability of this cue in those languages is comparatively low (Kail, 1989, Kilborn & Ito, 1989, Sasaki, 1991). Studies have repeatedly shown that subjects rely the most on cues that have high statistically validity in their language (McDonald, 1987). Bates and MacWhinney (1989) have reported that word order is the strongest cue in English, and that subject-verb agreement and animacy are much weaker. However, prior research has not yet determined the relative strength of the case-marking cue for English, since case-marking is only available for pronouns in English. Given the high reliance on case-marking that is found in languages such as Hungarian and Japanese, evidence about eventual domination of case over word order in Dutch, and the high reliability of case-marking, we might expect that case-marking could dominate over word order in English. On the other hand, word order is much more generally available than pronominal case. If availability is a factor determining the relative strength of two reliable cues, then we might expect word order to dominate over case. To date, no study has examined this theoretically important issue in cue competition.|../mac/prof352.html",
    "Squashing, Rotating, Seeing, and Going: On Visual Knowledge in fMRI Resear...||Paper Track Squashing, Rotating, Seeing, and Going: On Visual Knowledge in fMRI Research itself? Morana Alac , Department of Cognitive Science malac@cogsci.ucsd.edu Squashing, Rotating, Seeing, and Going: Role of Visual Knowledge in fMRI Research Morana Alac (malac@cogsci.ucsd.edu) Department of Cognitive Science, 9500 Gilman Drive MC 0515 La Jolla, Ca 92093-0515 Abstract In order to assess the role that visual representations play in brain mapping research, two concrete instances of fMRI practice are analyzed. The primarily focus is on the conceptual structure of the participants in action. The analysis shows that comprehension is organized through complex conceptual structure that transforms abstract concepts into manipulable, object-like entities. It is also observed that such structures employ fictive motion (Talmy, 2000) in which a static entity is construed as being in motion. For fMRI researchers, visual representations of the brain play a crucial role in these cognitive processes. In addition, I will show how the meaning of brain images is constructed in coordination with other representations composed of linguistic utterances, as well as gestural representations, and the appropriation of material structures environment (Hutchins, 1995). Introduction practice century (Foucault, 1963, 1973: xiii).|../mac/prof36.html",
    "Conceptual Maps: a metacognitive strategy to learn concepts...||Paper Track Conceptual Maps: a metacognitive strategy to learn concepts Suzana Maria Modern educational pratice has been greatly influenced by trends in Cognitive Science. Recent studies have shown that conceptual maps are an effective metacognitive strategy. The conceptual maps use improve students’s learning in actual classrooms and also in the distance learning classrooms. I have made claims about knowledge and learning that are substantially different from behaviorism, because knowledge is “the ability to demonstrate what you can do” (Schank, 1999). Research investigates the performance of adult categorizes has shown that the most effective categorization is often associated with the dynamic memory and the notion of scenes and scripts (Schank, 1999). This kind of categorization is spontaneous and the work with conceptual maps is in contrast of this process of categorization, in wich students have consciouness of their own knowledge segmentation. I investigated the process of conceptual maps construction and I will analyze the finally maps produced in two classrooms experiments: a class where students are presents and a virtual class in a distance learning course and interpret these results. This work intends to investigate the basic cognitive characteristics of conceptual maps with the intention of better applying and exploring them to modelling of those concepts by the students – either in a real classroom situation or on a distance learning basis – using the software Cmap Tools, developed by the Institute for Human Cognition of the West Florida University as a pedagogical tool. In the identificztion of the main cognitive processes which are implied in the modelling of the concepts, also including the interaction facilited by the use of the software, which allows students to participate collaborativelly in the construction of each other’s conceptual maps. Additionally, the study of the production of students’ conceptual maps gives the recognition of the concepts underlying common strutural properties of the individuals and their community as well the perception of the opresence of the cognitive bypass wich itself constitute a significative obstacle to the learning of the concepts. The knowledge structures were part of the identity of a cultural group. Thus, a conceptual map should be considered as an active memory organizer, like a script (scripts have a stronger role than previously supposed and they have a primary function as organizers of information in memory). So, conceptual maps afford the observations about culture and cultural learning apply to concepts. Given what we know about the ideological similarity (Stich, 1982, 1983) and the diversity of cultures.|../mac/prof37.html",
    "Distributed Metacognition during Peer Collaboration...||Paper Track Distributed Metacognition during Peer Collaboration Christopher Andersen , School of Teaching and Learning, The Ohio State University andersen.18@osu.edu Current educational reform efforts stress the need for educators to teach critical thinking skills and use group-based pedagogical methods. However, the empirical literature is still unclear under which circumstances they are best fostered. The present study examined scientific reasoning as participants worked on a causal reasoning task alone versus with a partner. In previous work, individual working alone readily acquire new strategies and add them to their existing repertoire but do not abandon their less-effective strategies. The challenge of cognitive development lies not in the construction of new strategies, but rather in the consolidation of metastrategic knowledge that allows individuals to effectively evaluate and select strategies from their repertoire. The results of this research suggest that special emphasis should be placed in peer collaboration’s role in fostering metastrategic understanding rather than focusing only on the development of strategies. Further, it suggests a benefit of collaboration not previously explored; the mutual reinforcement of each other’s unstable metastrategic control through a process of distributed metacognition. Instead of deep collaboration about simple thought (theories and strategies), the benefits of collaboration for these participants lie in simple collaboration about deep thought (metastrategic knowledge).|../mac/prof38.html",
    "Sex and Seniority: The Effects of Linguistic Categories on Conceptual Judgm...||Paper Track Sex and Seniority: The Effects of Linguistic Categories on Conceptual Judgments and Memory Florencia Anggoro , Department of Psychology, Northwestern University f-anggoro@northwestern.edu Dedre Gentner , Department of Psychology, Northwestern University gentner@northwestern.edu The current study explored the effects of different semantic categories in kinship terms on similarity judgments, word extensions, and recognition memory. We compared Indonesian – in which sibling terms are based on relative age – with English – in which sibling terms are based on gender. In Experiment 1, participants saw triads of pictures of scenes involving kinship relations and were asked to make similarity judgments and to extend novel labels from the standards to the variants. The variants each resembled the standard along one dimension and differed along the other. In Experiment 2, other participants were asked to remember the standard pictures and were later tested on their recognition memory using the variants. Results from both experiments converged to suggest that participants’ judgments, word extensions, and memory were influenced by their semantic categories.|../mac/prof39.html",
    "Using Eye Movements to Study Cognitive Processes...||Tutorials Track Using Eye Movements to Study Cognitive Processes Keith Rayner , University of Massachusetts rayner@psych.umass.edu For the past twenty-five years, many researchers have been using eye movement data to investigate various issues related to cognitive processing. Recording of eye movements provides a relatively unobtrusive moment-to-moment indicator of processing in a number of tasks of interest to cognitive scientists. For example, a great deal of research on reading, scene perception, and visual search has used eye movement data. More recently, eye movement data have been used in a number of other information processing tasks (see Rayner, 1998). In this presentation, an overview will be provided of the many different types of eye-tracking systems that are available on the market. These systems typically vary along a number of dimensions including cost and how easy they are to use. Some systems can be mastered in a short time period whereas others require a considerable amount of practice before the operator feels comfortable with the system. More critically, there are important trade-offs between temporal and spatial resolution that are inherent in the differing systems. These tradeoffs will be discussed. An important point is that the specific research application typically dictates how important each factor is. It is also important to note that the temporal and spatial resolution of the system is also related to cost and ease of use issues. Eye movement recording typically results in a vast amount of data and it is essential that researches have well-formulated plans regarding how to analyze the data. The issue of how to best analyze eye movement data will be discussed and some representative examples of eye-tracking applications will be provided. Finally, some specific examples of how eye movement data can be used to answer certain research questions will be given.|../mac/prof4.html",
    "Implications of Distributed Representations for Semantic Processing: Eviden...||Paper Track Implications of Distributed Representations for Semantic Processing: Evidence from Alzheimer’s Disease Justin M. Aronoff , Departments of Linguistics Laura M. Gonnerman , Department of Psychology, Lehigh University, lag5@lehigh.edu Elaine S. Andersen , Departments of Linguistics, Psychology, and Program in Neuroscience, University of Southern California, eanderse@usc.edu Daniel Kempler , Comunication Sciences and Disorders, Emerson College, Daniel_Kempler@emerson.edu Amit Almor , Department of Psychology almor@sc.edu Prior work by Gonnerman and colleagues presented a theory of semantic processing in normal and impaired populations. Their account incorporates distributed representations and predicts a complex relationship between semantic knowledge and naming ability. According to this account, during the course of progressive brain damage, one should observe different relationships between damage to semantic knowledge and naming ability for natural kinds versus artifacts. For artifacts, the theory predicts that naming ability will not be strongly correlated with damage to semantic category structure, whereas for natural kinds the nature of the relationship will change as damage to the system progresses. To test this theory, young and elderly participants and patients with Alzheimer’s disease named a series of pictures and completed a board sorting task, in which they placed words from a semantic category on a two dimensional grid in a way that represented their inter-similarities, thus reflecting the nature of their semantic knowledge. Results confirmed the prediction that a strong relationship between picture naming and disrupted category structure is evident only for natural kinds categories at later stages of damage. For natural kinds in earlier stages and artifacts throughout the progression of the disease, disrupted category structure is not directly reflected in naming performance. These data point to a complex relationship between the underlying category structure and its realization in naming ability.|../mac/prof40.html",
    "Instructional Focus Does Not Effect Implicit Pattern Learning...||Paper Track Instructional Focus Does Not Effect Implicit Pattern Learning Ivan K. Ash , Department of Psychology,University of Illinois at Chicago iash1@uic.edu Timothy J. Nokes , Department of Psychology,University of Illinois at Chicago tnokes@uic.edu Participants performed a dual-component training procedure that combined a serial reaction time task and an artificial grammar learning task under two instructional conditions. Participants given memory-focused instructions performed at the same level as participants given motor-focused instructions on the serial reaction time test and grammar sorting task. Both groups performed better than a control (no training) group. Results suggest that only a minimal amount of attentional focus on aspects of the stimuli relevant to the pattern is needed to acquire implicit pattern knowledge.|../mac/prof41.html",
    "There Is No Naked Eye: Higher-Order Social Concepts Clothe Visual Perceptio...||Paper Track There Is No Naked Eye: Higher-Order Social Concepts Clothe Visual Perception Emily Balcetis , Department of Psychology, Uris Hall, Cornell University eeb29@cornell.edu Rick Dale , Department of Psychology, Uris Hall, Cornell University rad28@cornell.edu Vision researchers have investigated many sources of information that assist perception. Although basic visual properties of stimuli can alter interpretation, the following 5 studies contend that implicit, complex, social information significantly influences basic visual perception. Study 1 employed a scrambled sentence priming procedure used to activate concepts without participants’ awareness. Studies 2, 4, and 5 used less contrived priming techniques that required participants to read paragraphs related to the intended prime. Study 3 utilized self-generated primes created through minimal prompting. All priming procedures resulted in an increase in the proportion of initial percepts that were related to the primed concept. These priming procedures not only demonstrate the influence of complex information on perception of ambiguous figures but of indistinct natural scenes as well.|../mac/prof42.html",
    "The Use of Mass Nouns to Quantify Over Individuals...||Paper Track The Use of Mass Nouns to Quantify Over Individuals David Barner , Department of Psychology, William James Hall, Harvard barner@fas.harvard.edu Jesse Snedeker , Department of Psychology, William James Hall, Harvard snedeker@wjh.harvard.edu Theories of the mass-count distinction in linguistics, philosophy and psychology commonly argue that count nouns are distinguished from mass nouns by their reference to, and quantification over, individuals (e.g., Bloom, 1999; Bunt, 1985; Wisniewski, Imai & Casey, 1996). We present experimental evidence that both children and adults interpret some mass nouns as quantifying over individuals, and suggest a model of the mass-count distinction that includes these cases. discrete and bounded, and which are subject to being counted. This entailment persists even in cases where particular lexico-semantic knowledge is absent. Talk of three blickets, while not terribly informative, does nevertheless imply that three individual things, whatever they may be, are being referred to. Such examples suggest that speakers might “conceptualize the referents of count nouns as distinct, countable, individuated things and those|../mac/prof43.html",
    "A Negative Effect of Evaluation Upon Analogical Problem Solving...||Paper Track A Negative Effect of Evaluation Upon Analogical Problem Solving Christopher R. Bearman , Psychology Department, Lancaster University C.Bearman@lancaster.ac.uk Thomas C. Ormerod , Psychology Department, Lancaster University T.Ormerod@lancaster.ac.uk Linden J. Ball , Psychology Department, Lancaster University L.Ball@lancaster.ac.uk Evaluation is generally considered to enhance problem solving and is strongly correlated with increasing expertise. Moreover, manipulations that increase the active processing of source problems generally promote analogical transfer of solution principles. Therefore, we expected that an instruction to evaluate the information given in problem and solution exemplars would enhance analogical problem solving. However, in Experiment 1, evaluation was found to have a detrimental effect on transfer compared with control groups instructed to summarize source problems and solutions, even when participants received additional instructions to memorize source problems for later recall. In Experiment 2, the impairing effects of instructions to evaluate were not reduced by making participants engage in evaluation when solving the target problem, a test of a ‘transfer appropriate processing’ explanation. We propose that instructions to evaluate lead participants to focus upon some elements of source problems and solutions at the expense of other elements required for effective transfer. Evaluation, the set of processes involved in judging problem-related information under dimensions such as importance, value, appropriateness and scope, is generally seen as fundamental to human intellectual performance. It is central to the TOTE (Test-Operate-Test-Exit) unit proposed by Miller, Galanter and Pribram (1960) as the basis of exploratory problem-solving behaviour, and is considered to be beneficial when seeking solutions to complex real-world problems (Klein, 1999). For example, Ashton and Kennedy (2002) found that evaluation instructions led to a reduction in the use of temporally salient but otherwise inappropriate data when making auditing decisions. Evaluation is also invoked in accounts of expertise in domains such as management (Easton, 1992), medicine (Nezu & Nezu, 1995), and design (Darses, 2002; Ormerod & Ridgeway, 1999). Indeed, the proportion of problem-solving time devoted to evaluative activity has been shown to increase with domain expertise (Chi, Feltovich & Glaser, 1981; Schoenfeld, 1992).|../mac/prof44.html",
    "The Flexible Use of Deontic Mental Models...||Paper Track The Flexible Use of Deontic Mental Models Sieghard Beller , Department of Psychology, University of Freiburg beller@psychologie.uni-freiburg.de Vision researchers have investigated many sources of information that assist perception. Although basic visual properties of stimuli can alter interpretation, the following 5 studies contend that implicit, complex, social information significantly influences basic visual perception. Study 1 employed a scrambled sentence priming procedure used to activate concepts without participants’ awareness. Studies 2, 4, and 5 used less contrived priming techniques that required participants to read paragraphs related to the intended prime. Study 3 utilized self-generated primes created through minimal prompting. All priming procedures resulted in an increase in the proportion of initial percepts that were related to the primed concept. These priming procedures not only demonstrate the influence of complex information on perception of ambiguous figures but of indistinct natural scenes as well.|../mac/prof45.html",
    "Polynesian Tapu in the ‘Deontic Square’: A Cognitive Concept, its Linguisti...||Paper Track Polynesian Tapu in the ‘Deontic Square’: A Cognitive Concept, its Linguistic Expression and Cultural Context Andrea Bender , Department of Psychology, University of Freiburg bender@psychologie.uni-freiburg.de Sieghard Beller , Department of Psychology, University of Freiburg beller@psychologie.uni-freiburg.de The Polynesian concept of tapu, introduced into European languages as ‘taboo’, is the most salient example of a deontic concept in Tonga. In order to examine whether tapu and its related concepts reflect cross-cultural equivalences, we pursue a comparative approach that is conceptually grounded in the ‘deontic square’. By analyzing preferences for linguistic expressions, implications of grammar, and the traditional cultural context, we found remarkable similarities on a structural level. The most significant difference lies in the concept of agency, which is more prevalent in ‘Western’ than in Tongan culture.|../mac/prof46.html",
    "Embodied Verbal Semantics: Evidence from an Image-Verb Matching Task...||Paper Track Embodied Verbal Semantics: Evidence from an Image-Verb Matching Task Benjamin Bergen , Department of Linguistics bergen@hawaii.edu Shweta Narayan , International Computer Science Institute shweta@icsi.berkeley.edu Jerome Feldman , International Computer Science Institute jfeldman@icsi.berkeley.edu It has recently been demonstrated that certain neural circuitry involved in the execution of specific motor actions is also used when the very same motor actions are observed or when language describing those actions is perceived. In humans, the pre-motor cortex is organized into regions that are involved in the execution and observation of actions performed by at least the following three general areas: the mouth, the hand, and the leg. The discovery of this “mirror system”, involved in production and perception of motor behavior, leads to a viable hypothesis about the processing of linguistic units that refer to these actions. It could be that understanding a verb describing an action involves the activation of the very same mirror circuitry involved in performing and recognizing that action. This hypothesis is tested in a matching task, in which subjects were presented first with an image depicting some action, followed by a verb that either described that action or did not. They were asked to decide as quickly as possible whether the verb was appropriate to the image. It was reasoned that if the verbs and images for particular actions recruited the same mirror circuitry, then there should be interference in those cases where the actions described by the verb and image were not the same but used the same effector. The results showed that it took subjects significantly longer to reject nonmatching verbs and images when the two shared an effector than when they did not. These results support the hypothesis that understanding action language requires the activation of effector-specific neural circuitry in the human mirror system.|../mac/prof47.html",
    "Temporal Processing for Syntax Acquisition: A simulation study...||Paper Track Temporal Processing for Syntax Acquisition: A simulation study Jean-Marc Blanc , Institut des Sciences Cognitives UMR 5015 CNRS-Université Claude Bernard Lyon 1 blanc@isc.cnrs.fr Christelle Dodane , Institut des Sciences Cognitives UMR 5015 CNRS-Université Claude Bernard Lyon 1 dodane@isc.cnrs.fr Peter Ford Dominey , Institut des Sciences Cognitives UMR 5015 CNRS-Université Claude Bernard Lyon 1 dominey@isc.cnrs.fr Early perceptual processing capabilities are likely to contribute to categorization of lexical vs. grammatical words by newborns. This lexical categorization could be performed by detecting differences in the prosodic structure of these word categories. Here we demonstrate that a Temporal Recurrent Network (TRN) that allows realistic treatment of the dynamic temporal aspect of prosody performs this lexical categorization task on French and English. We then examine the functional relation between this capability, and non-linguistic temporal discrimination. We reduce sensitivity to temporal structure in the TRN by increasing the network time constants. This yields (1) a reduction in performance in the lexical categorization task, and (2) a deficit in the processing of brief auditory events similar to that observed for children with SLI. While our principle result is that the TRN can perform lexical categorization based on prosodic structure, it is of interest that the impaired TRN suggests a functional link between impaired temporal processing, and impaired lexical categorization in SLI.|../mac/prof48.html",
    "The effect of emotion on conditional reasoning...||Paper Track The effect of emotion on conditional reasoning Isabelle Blanchette , Department of Psychology, University of Manchester i.blanchette@psychology.bbk.ac.uk Anne Richards , Department of Psychology, Birkbeck College, University of London a.richards@bbk.ac.uk In two experiments, we explore whether logical reasoning abilities are affected by emotion. In both experiments, we compared participants’ performance on a conditional reasoning task when the content was emotional and neutral. In Experiment 1, conditional statements included either emotional words or neutral words. In Experiment 2, we manipulated the emotional connotation of initially neutral words using a conditioning procedure. Words were repeatedly paired with either positive, negative, or neutral images. These words were then used in a conditional reasoning task. In both experiments, participants’ performance was more likely to deviate from prescriptions of normative logic when the content was emotional compared to when it was neutral.|../mac/prof49.html",
    "Complicate Locally, Simplify Globally: Starting With Complex Primitives Pay...||Rumelhart Prize Talk Complicate Locally, Simplify Globally: Starting With Complex Primitives Pays Off Aravind K. Joshi , Department of Computer and Information Science and Institute for Research in Cognitive Science University of Pennsylvania In setting up a formal system to specify a grammar formalism, the conventional (mathematical) wisdom is to start with primitives (basic primitive structures) as simple as possible and then to introduce various operations for constructing more complex structures. An alternate approach is to start with more complex primitives, which directly capture some crucial linguistic properties and then to introduce some universal operations for composing these complex structures. These two approaches provide different domains of locality, i.e., the domains over which various types of linguistic dependencies can be specified. This latter approach, characterized as complicate locally, simplify globally (CLSG), pushes non-local dependencies to become local, i.e., they arise in the basic primitive structures to start with. I will explore the CLSG approach in the context of some mathematical, linguistic, computational, statistical, and psycholinguistic properties, all these with possible relevance to the cognitive architecture for language. I will describe these results in an introductory manner.|../mac/prof5.html",
    "Content Effects in Conditional Reasoning: Evaluating the Container Schema...||Paper Track Content Effects in Conditional Reasoning: Evaluating the Container Schema Amber N. Bloomfield , Department of Psychology, Northwestern University a-bloomfield@northwestern.edu Lance J. Rips , Department of Psychology, Northwestern University rips@northwestern.edu This study presented participants with conditional statements with content that was group-related, container-related, and a combination of group and container. Participants received modus ponens, modus tollens, denying the antecedent, affirming the consequent, and inconsistent forms of conditional arguments. Results demonstrated a significant advantage of the group over the container wording for some of the arguments. These results suggest that content that is specifically container-related does not provide a consistent advantage over other types of content, contrary to some theories of embodied mathematics.|../mac/prof50.html",
    "Literal meaning and context categories in the attribution of communicative ...||Paper Track Literal meaning and context categories in the attribution of communicative intentions: A developmental study Francesca M. Bosco , Centro di Scienza Cognitiva, Università di Torino bosco@psych.unito.it Monica Bucciarelli , Centro di Scienza Cognitiva, Università di Torino monica@psych.unito.it Bruno G. Bara , Centro di Scienza Cognitiva, Università di Torino bara@psych.unito.it Studies in developmental literature claim that for young children context plays a more important role than literal meaning in comprehending a speaker's communicative intention. The present study evaluates this claim for different categories of context in children aged 3 to 7 years. In particular, we analyze contexts pertaining to the categories we define as follows: Access, Space, Time, Discourse, Extra-linguistic-behavior and Status. The results of the study show that, for all children, the contexts investigated within the categories Space, Time and Status play a more important role than literal meaning and, on the contrary, the literal meaning has a weight major than the context within the category Discourse. We discuss the results in terms of different roles played in different contexts in the reconstruction of a speaker’s communicative intention in children of different ages.|../mac/prof51.html",
    "Analysis of Adaptive Dynamical Systems for Eating Regulation Disorders...||Paper Track Analysis of Adaptive Dynamical Systems for Eating Regulation Disorders Tibor Bosse , Vrije Universiteit Amsterdam, Department of Artificial Intelligence tbosse@cs.vu.nl Martine F. Delfos , PICOWO, Psychological Institute for Consultancy, Education and Research mfdelfos@wanadoo.nl Catholijn M. Jonker , Vrije Universiteit Amsterdam, Department of Artificial Intelligence jonker@cs.vu.nl Jan Treur , Vrije Universiteit Amsterdam, Department of Artificial Intelligence, Universiteit Utrecht, Department of Philosophy treur@cs.vu.nl To analyse a subject’s mental processes, psychotherapists often face nontrivial properties of adaptive dynamical systems. Analysis of dynamical systems usually is performed using mathematical techniques. Such an analysis is not precisely the type of reasoning performed in psychotherapy practice. In this paper it is shown how practical reasoning about dynamic properties of adaptive dynamical systems within psychotherapy can be described using dynamical logical methods and a high-level language to describe dynamics.|../mac/prof52.html",
    "Deference and Essentialism in the Categorization of Chemical Kinds...||Paper Track Deference and Essentialism in the Categorization of Chemical Kinds Nick Braisby , Department of Psychology, The Open University, Walton Hall N.R.Braisby@open.ac.uk Psychological essentialism has been subject to much debate. Yet a key implication – that people should defer to experts in categorizing natural kinds – has not been widely examined. Three experiments examine deference in the categorization of chemical kinds. The first establishes borderline cases used in the second and third. These latter show limited deference to experts, and some deference to non-experts. These data are consistent with a perspectival framework for concepts in which categorization is sometimes based on micro-structural properties and sometimes on appearance and function.|../mac/prof53.html",
    "Something Old, Something New: Addressee Knowledge and the Given-New Contrac...||Paper Track Something Old, Something New: Addressee Knowledge and the Given-New Contract Holly P. Branigan , Department of Psychology Holly.Branigan@ed.ac.uk Janet F. McLean , Department of Psychology Janet.McLean@ed.ac.uk Hannah Reeve , Department of Psychology Although speakers in a dialogue are known to design utterances cooperatively with respect to meaning, less is known about audience design with respect to syntax. We report two picture verification experiments that investigated the production of Given-New ordering when speakers' and addressees' knowledge differed. In both experiments, speakers produced word orders that reflected their own but not their addressees' knowledge states. We suggest that speakers do not engage in audience design for aspects of their utterances which they do not consider to be necessary for adequate communication.|../mac/prof54.html",
    "Syntactic Alignment Between Computers and People: The Role of Belief about ...||Paper Track Syntactic Alignment Between Computers and People: The Role of Belief about Mental States Holly P. Branigan Holly.Branigan@ed.ac.uk Martin J. Pickering , Department of Psychology, Edinburgh Martin.Pickering@ed.ac.uk Jamie Pearson , Department of Psychology, Edinburgh Jamie.Pearson@ed.ac.uk Janet F. McLean , Department of Psychology, Edinburgh Janet.McLean@ed.ac.uk Clifford Nass , Department of Communication, Stanford University nass@stanford.edu People tend to mirror the syntax used by their interlocutors in dialogue. Given that people treat computers as “social actors” in many ways, we might expect them to mirror computers’ syntax as well. We report an experiment in which naïve participants played a dialogue game in which they believed a pre-programmed interlocutor was a person or a computer. In both cases, there was a very strong tendency to repeat syntactic form. It does not appear that beliefs about the mental states of one’s interlocutor mediate between perception and production.|../mac/prof55.html",
    "When it is Adaptive to Follow Streaks: Variability and Stocks...||Paper Track When it is Adaptive to Follow Streaks: Variability and Stocks Bruce D. Burns , Department of Psychology; Michigan State University burnsbr@msu.edu Streaks of events are ubiquitous yet understanding the behavioral effects of them has been restricted by the lack of testable hypotheses concerning the most basic question: When do we tend to follow streaks (positive recency), and when do we tend to go against streaks (negative recency)? From an analysis of positive recency in terms of adaptivity, I develop two elements of people's representation of the process generating a sequence which should be predictive of people's use of positive recency. The first factor is randomness, which has already been tested empirically, and the second is variability which is tested here. The context used is stocks because not only do people seem to give weight to streaks in the stockmarket, but recent evidence suggests that it may be beneficial to do so. As predicted, participants were more likely to predict that a small company (more variable price) than a large company (less variable price) would continue a streak of six months of increased/decreased stock prices. However, regardless of the initial streak, participants strongly tended to switch which company would do better between six months and ten years. The results show that there are many interesting behavioral phenomena associated with streaks, and that Burns' (2001, 2003) analysis provides some useful tools for exploring them.|../mac/prof56.html",
    "Causality and Reasoning: The Monty Hall Dilemma...||Paper Track Causality and Reasoning: The Monty Hall Dilemma Bruce D. Burns , Department of Psychology; Michigan State University burnsbr@msu.edu Mareike Wieth , Department of Psychology; Michigan State University wiethmar@msu.edu In the Monty Hall Dilemma (MHD) contestants try to choose which of three doors conceals a prize. After selecting a door, one of the other doors is opened by a host who knows where the prize is, but always reveals a dud. Contestants are then asked if they want to stay with their first choice, or switch to the other unopened door? Switching yields a two-thirds chance of winning, but most people have difficulty accepting this answer. Glymour (2001) points out that central to the MHD is a particular causal structure, the collider principle, that people often experience difficulty with. We hypothesized that an isomorphic version of the MHD that would help participants understand its causal structure would improve their performance. Making the MHD a form of competition should be a way to achieve this. We confirmed this by showing that in a competition version of the MHD participants were much more likely to solve the problem, and more likely to answer a counterfactual question indicating a correct understanding of the problem's causal structure. Furthermore, regardless of MHD version, participants who solved the problem were more likely to also answer the counterfactual correctly. Thus the MHD can be seen as an example of people's difficulties with understanding even relative simple causal structures.|../mac/prof57.html",
    "A Mechanism-Based Framework for Predicting Routine Procedural Errors...||Paper Track A Mechanism-Based Framework for Predicting Routine Procedural Errors Michael D. Byrne , Department of Psychology, Rice University byrne@acm.org Routine procedural errors are facts of everyday life but have received little empirical study and have eluded prediction. Leading frameworks for thinking about such errors have not been successful in generating predictions, either. This paper describes the desiderata for error prediction, and notes that the MHP was a step in the right direction. Because it generally meets the criteria, ACT-R is proposed as a simulation framework for making predictions about routine procedural errors, and some of the critical mechanisms explored.|../mac/prof58.html",
    "The Statistical Brain: Reply to Marcus’ The Algebraic Mind...||Paper Track The Statistical Brain: Reply to Marcus’ The Algebraic Mind Francisco Calvo , Department of Philosophy, University of Murcia fjcalvo@um.es Eliana Colunga , Department of Psychology, University of Colorado colunga@psych.colorado.edu Marcus (2001) argues that only those connectionist models that incorporate (classical) rules can account for the phenomenon of transfer of learning in infants. Seidenberg and Elman (1999) have tried to counter to Marcus by means of a simple recurrent network (SRN) trained on a categorization task. In this paper we show how a prediction-SRN, trained on a simple but structured pre-training set, can preserve its computational equivalency with respect to classical counterparts while eschewing the need to posit rule-governed underlying mechanisms; a criticism that has been raised against Seidenberg and Elman’s categorization-based reply.|../mac/prof59.html",
    "The Convergence of Language and Computation: A Symposium in Honor of Aravin...||Rumelhart Plenary The Convergence of Language and Computation: A Symposium in Honor of Aravind K. Joshi Fernanda Ferreira , Psychology, Michigan State University Robert Frank , Cognitive Science, Johns Hopkins University rfrank@jhu.edu Edward Stabler , Linguistics, University of California, Los Angeles stabler@ucla.edu Bonnie Webber , Informatics, University of Edinburgh bonnie@inf.ed.ac.uk A hallmark of the research of Aravind K. Joshi is its elegant fusion of ideas from the domains of mathematics and computation with concerns from the cognitive science of language. Joshi's work has time and again demonstrated the important role that formally constrained systems can play in providing simple explanations for apparently complex phenomena in the domains of syntax, discourse, language acquisition, sentence comprehension and generation. This symposium honors Joshi's many contributions, presenting a range of research he has influenced in fundamental ways, both through his own work and through his role in stimulating the work of others.|../mac/prof6.html",
    "Do we think about time in terms of space?...||Paper Track Do we think about time in terms of space? Daniel Casasanto , NE20-457, MIT djc@mit.edu Lera Boroditsky , NE20-456, MIT lera@mit.edu We often talk about time in terms of space, but do we think about time in terms of space? A series of experiments investigated the relationship between mental representations of space and time, using low-level psychophysical tasks with entirely nonlinguistic stimuli and responses. Participants watched lines growing across a computer screen, and estimated either their duration or their displacement. Temporal estimates were strongly modulated by the displacements of the moving stimuli, even when participants were encouraged to selectively attend to temporal information. In contrast, spatial estimates were not significantly modulated by stimulus duration. This behavioral asymmetry reflects the directionality of spatiotemporal metaphors in language. Results support the proposal that whereas space is mentally represented on its own terms, temporal representations depend upon spatial representations.|../mac/prof60.html",
    "The Prescriptive Destiny of Predictive Attitudes: From Expectations to Norm...||Paper Track The Prescriptive Destiny of Predictive Attitudes: From Expectations to Norms via Conventions Cristiano Castelfranchi , Institute of Cognitive Science and Technology – CNR castel@ip.rm.cnr.it Francesca Giardini , University of Siena, Cognitive Science Doctorate giardini@media.unisi.it Emiliano Lorini , University of Siena lorini@media.unisi.it Luca Tummolini , University of Siena, Cognitive Science Doctorate tummoli@media.unisi.it The goal of this paper is to show the normative component of a convention spread in a population of cognitive agents. To address this aim we will defend two distinct thesis. The former is that even simple predictions developed to anticipate future state of affairs have an intrinsic tendency to evolve in full expectations and then in prescriptions. We consider this as a multilevel phenomenon occurring either at the individual psychological level or at the interpersonal one or, finally, at the collective macro social level. The latter thesis is that we consider this tendency as one of the possible paths of the spontaneous emergence of agents’ commitments, of conventions and likely of real social norms: the tacit emergence of a prescriptive character and, then, of obligations and duties. We will examine the constitutive elements –both cognitive and relational – of this process of spontaneous transition from the predictive attitudes to the prescriptive ones and, on this basis, to real normative attitudes. Finally, we will discuss the inevitably normative component of conventions as traditionally described (Lewis 1969). We will argue that this fundamental process is notably left implicit or insufficiently explained.|../mac/prof61.html",
    "Learning Spurious Correlations instead of Deeper Relations...||Paper Track Learning Spurious Correlations instead of Deeper Relations Norma Chang , Department of Psychology, Carnegie Mellon University nchang@andrew.cmu.edu Kenneth R. Koedinger , Human-Computer Interaction Institute, Carnegie Mellon University koedinger@cmu.edu Marsha C. Lovett , Department of Psychology, Carnegie Mellon University lovett@cmu.edu Effective instructional design requires navigating the tradeoff between providing helpful cues to the correct solutions and supplying hints that ultimately detract from what students learn. Manipulating the correlations between superficial features and the correct solutions in a set of training problems resulted in leading students to the incorrect answers on a posttest lacking the same associations. The theroretical and educational implications of the outcome of this practice are discussed.|../mac/prof62.html",
    "Diagrammatic Re-codification of Probability Theory:A Representational Epist...||Paper Track Diagrammatic Re-codification of Probability Theory:A Representational Epistemological Study Peter C-H. Cheng , Department of Informatics/COGS, University of Sussex p.c.h.cheng@sussex.ac.uk It is claimed that the current representations used for probability theory provides a poor codification of that knowledge. The limitations of the representations and how they encode the knowledge causes conceptual difficulties and makes problem solving difficult. Probability Space diagrams constitute a new representational system that provides a simpler and more coherent codification of probability theory, which effectively supports problem solving. The two approaches are contrasted, which demonstrates that the effects of representations on advanced forms of cognition extend beyond problem solving and impacts on the conceptual understanding and learning of bodies of knowledge.|../mac/prof63.html",
    "When A’s and B’s are C’s and D’s: The effect of the cross-classification of...||Paper Track When A’s and B’s are C’s and D’s: The effect of the cross-classification of items on learned concepts Seth Chin-Parker , Department of Psychology,University of Illinois chinpark@s.psych.uiuc.edu Brian H. Ross , Beckman Institute, University of Illinois bross@s.psych.uiuc.edu Category learning plays a crucial role in cognition since the acquired concepts are used later for categorization, explanation, communication, problem solving, and inference, yet there is little information about how category learning can affect previously acquired concepts. This is especially interesting when one considers items that belong to multiple categories, items that can be cross-classified. The current study investigates a learning situation where one classification set is learned and the knowledge gained is assessed. Then a second, orthogonal classification set is learned for the same items. The experiments show that there is an effect of the crossclassification on judgements made about items in terms of the initial classification set. There seem to be two possible effects of learning a cross-classification for items. The secondary learning causes either an intrusion of attribute information critical for the second classification set, or a specification as to what attributes are critical for the first classification set.|../mac/prof64.html",
    "Seven Factors that Make Learning Successful in Networked Collaboration...||Paper Track Seven Factors that Make Learning Successful in Networked Collaboration Kwangsu Cho , Learning Research and Development Center, Univ. of Pittsburgh Kwangsu@Pitt.Edu Christian D. Schunn , Learning Research and Development Center, Univ. of Pittsburgh Schunn@Pitt.Edu This paper examines seven strategic factors that seem to make learning successful in networked collaboration. Through our literature review, it was found that there are at least seven strategic factors: deep thinking, more interaction, cognitive conflict resolution, adaptation over time, constructive use of technology, task coordination between media, and asynchrousness management. It is concluded that the effectiveness of learning and problem solving in networked collaboration is determined by students’ use of those factors in networked collaboration.|../mac/prof65.html",
    "Automaticity of Number Perception...||Paper Track Automaticity of Number Perception Jessica M. Choplin , Vanderbilt University Psychology Department jessica.choplin@vanderbilt.edu Gordon D. Logan , Vanderbilt University Psychology Department gordon.logan@vanderbilt.edu In 2 experiments, the authors investigated whether the attributes “small” and “large” associated with individual digits are responsible for the effects of size congruity on judgments of physical size (Tzelgov, Meyer, & Henik, 1992). In Experiment 1, a size congruity effect was observed when participants judged the relative physical sizes of two digits. However, size congruity effects were just as strong when participants judged the relative physical sizes of small digits (i.e., 1-4) paired with letters. In Experiment 2, a similar size congruity effect was observed when participants judged the sizes of squares within which individual small digits were presented. Consistent with memory-based theories of automaticity, these results suggest that associations between the attribute “small” and individual small digits are sufficient to explain many, if not all, size congruity effects.|../mac/prof66.html",
    "Imagistic Simulation in Scientific Model Construction...||Paper Track Imagistic Simulation in Scientific Model Construction John J. Clement , Scientific Reasoning Research Institute College of Natural Sciences and Mathematics and School of Education clement@srri.umass.edu The role of mental simulation in scientific learning processes is poorly understood. This paper examines case study data concerning: how processes like imagistic simulation can make a schema runnable ; and how assembling a scientific model from simpler runnable schemas can transfer runnability to the model. This has interesting consequences for theories of scientific discovery and for science instruction, and may help us begin to resolve the apparent paradox involved in learning from running a new experiment in one s head.|../mac/prof67.html",
    "PAM: A Cognitive Model of Plausibility...||Paper Track PAM: A Cognitive Model of Plausibility Louise Connell , Department of Computer Science, University College Dublin louise.Connell@ucd.ie Mark Keane , Department of Computer Science, University College Dublin mark.keane@ucd.ie Plausibility has been implicated as playing a critical role in many cognitive phenomena from comprehension to problem solving. Yet, plausibility is usually treated as an operationalized variable (i.e., a plausibility rating) rather than being explained or studied in itself. This paper reports on a new model of plausibility that is aimed at modeling several direct studies of plausibility. This model, the Plausibility Analysis Model (PAM), used distributional knowledge about word co-occurrence (word-coherence) and commonsense knowledge of conceptual structure and relatedness (conceptcoherence) to determine the degree of plausibility of some target description. A detailed simulation of several plausibility findings is reported, which shows a close correspondence between the model and human judgments.|../mac/prof68.html",
    "When Less is Less and When Less is More: Starting Small with Staged Input...||Paper Track When Less is Less and When Less is More: Starting Small with Staged Input Christopher M. Conway , Department of Psychology; Cornell University cmc82@cornell.edu Michelle R. Ellefson , Department of Psychology; University of Warwick M.Ellefson@warwick.ac.uk Morten H. Christiansen , Department of Psychology; Cornell University mhc27@cornell.edu It has been suggested that external and/or internal limitations may paradoxically lead to superior learning (i.e., the concepts of “starting small” and “less is more”; Elman, 1993; Newport, 1990). In this paper, we explore what conditions might lead to a starting small e ect. We report on three artificial grammar learning experiments with human participants. In Experiment 1, we found an e ect of starting small with visual center-embedded, recursive input staged incrementally. Experiment 2 replicated this finding and extended the e ect to right-branching recursive structure. Finally, in Experiment 3 we found no e ect for starting small with auditory center-embedded input. These results suggest that starting small can confer a learning advantage but perhaps only under certain conditions.|../mac/prof69.html",
    "Positioning and Identities in Collaborative Cognition and Learning...||Plenary Track Positioning and Identities in Collaborative Cognition and Learning James G. Greeno , Stanford University Cognitive processes can be considered as functions of activity systems in which individuals participate. Several analyses have shown that successful cognitive functioning depends on ways that individuals are positioned in participation structures of interaction, regarding each other and regarding the subject-matter domain of their activity. Concepts of individuals’ knowledge and cognitive processes have not played significant roles in these analyses. This talk will present findings from research in progress that is focused on continuities of individual students over time in the activity systems of mathematics classrooms, especially students’ and teachers’ constructions of being or not being “good at math.” We are developing a concept of mathematical identity to understand ways in which individual students come to be positioned in classroom interaction, including the knowledge they are understood to acquire and their characteristic ways of participating in classroom activities.|../mac/prof7.html",
    "Hesitation in speech can. . . um. . . help a listener understand...||Paper Track Hesitation in speech can. . . um. . . help a listener understand Martin Corley , School of Philosophy, Psychology and Language Sciences University of Edinburgh Martin.Corley@ed.ac.uk Robert J Hartsuiker , Department of Experimental Psychology, Ghent University Rob.Hartsuiker@rug.ac.be This paper investigates the effect of dis uencies on listeners' on-line processing of speech. More speci cally, it tests the hypothesis that lled pauses like um, which tend to occur before words that are low in accessibility, act as a signal to the listener that a relatively inaccessible word is about to be produced. Two experiments are reported, in which participants followed recorded instructions to press buttons corresponding to images on a computer screen. In 50% of trials, the spoken name of the image was preceded by um. In experiment 1, the intrinsic accessibility of the target items was manipulated (by means of lexical frequency); in experiment 2, the extrinsic (visual) accessibility varied. Both experiments demonstrated that participants were quicker to respond when a target was preceded by um, regardless of whether the item referred to was dif cult to access or not. In addition, in experiment 2 there was a weak interaction between accessibility and presence or absence of an um. We present the data here as early evidence that listeners can bene t from dis uencies in others' speech, and outline some methodological pitfalls and further experiments.|../mac/prof70.html",
    "Strategic, Metacognitive, and Social Aspects in Resource-oriented Knowledge...||Paper Track Strategic, Metacognitive, and Social Aspects in Resource-oriented Knowledge Exchange Dr. Ulrike Cress , Department of Applied Cognitive Psychology and Media Psychology, University of Tuebingen u.cress@iwm-kmrc.de The increasing use of internet and intranet fosters the possibilities for resource-oriented knowledge exchange in large groups of people working in parallel. However, the individual decision to contribute information to a shared pool builds up a public-goods dilemma, and people are often discouraged from sharing knowledge because of strategic reasons. Additionally, the highly anonymous situation where resource-oriented knowledge exchange takes place could further amplify the tendency to withhold knowledge: this situation provides almost no metaknowledge about the importance of one’s information for the others and almost no social cues. In two experiments the effects of metaknowledge and social standards are investigated. Results show that the former influences the quality of the exchanged information, whereas the latter influences the quantity. An example for such a parallel work situation is the following: In a consultancy each consultant’s task is to attend to different companies. To decide which the best strategy is for one of his/her companies, a consultant should know about the efficacy of different possible strategies in other companies which are taken care of by one of her colleagues. To do his/her job in the best way s/he should be able to get information from those other projects. Therefore, knowledge exchange could enhance her/his performance. But the question arises how knowledge can be effectively exchanged in such big organizations. Direct and demandoriented knowledge exchange through asking each other isn’t very efficient when group members do not know which|../mac/prof71.html",
    "A Neurocomputational Account of Short- and Long-Term Recency Effects...||Paper Track A Neurocomputational Account of Short- and Long-Term Recency Effects Eddy J. Davelaar , School of Psychology, Birkbeck College e.davelaar@psychology.bbk.ac.uk Marius Usher , School of Psychology, Birkbeck College m.usher@psychology.bbk.ac.uk A neurocomputational model of free recall is presented that combines an activation-based short-term memory system with a weight-based contextual episodic one. The short-term buffer component adheres to well-established neural principles related to delay activation in the prefrontal cortex and is responsible for recency effects in immediate free recall. The episodic memory system is related to the medial temporal areas and provides a changing context, which underlies recency effects in the continuous distractor task. The model accounts for such basic findings as listlength and proactive interference effects that show a dissociation between primacy and recency effects in immediate free recall. In addition, it also accounts for the dissociation between short- and long-term recency with instructed output order and a dissociation found with amnesic patients. This work provides an initial step towards a fuller understanding of how the underlying neural principles explain patterns of memory performance and can be instrumental in resolving debates in the cognitive/behavioral literature.|../mac/prof72.html",
    "Tick Talk...||Paper Track Tick Talk Douglas J. Davidson , Beckman Institute University of Illinois dvdsn@casper.beckman.uiuc.edu Kathryn Bock , Beckman Institute University of Illinois kbock@s.psych.uiuc.edu David E. Irwin , Beckman Institute University of Illinois dirwin@s.psych.uiuc.edu Three experiments on the production of time expressions are reported. In each of the experiments, speakers produced time expressions to analog clock faces while their eye movements were recorded with a headband-mounted eye tracker. Previous work has shown that speakers can apprehend the visual-conceptual information in a clock display very quickly, and that this is followed by relatively slow, incremental formulation of the time expressions. Analyses of fixation patterns from the present experiments lend additional support to this model.|../mac/prof73.html",
    "Representation Issues in Visual Analogy...||Paper Track Representation Issues in Visual Analogy Jim Davies , Artificial Intelligence Laboratory, College of Computing, Georgia Institute of Technology jim@jimdavies.org Ashok K. Goel , Artificial Intelligence Laboratory, College of Computing, Georgia Institute of Technology goel@cc.gatech.edu|../mac/prof74.html",
    "Analogical Inference in Automatic Interpretation...||Paper Track Analogical Inference in Automatic Interpretation Samuel B. Day , Department of Psychology, Northwestern University s-day2@northwestern.edu Dedre Gentner , Department of Psychology, Northwestern University gentner@northwestern.edu We present findings suggesting that analogical inference can play a role in the fundamental processes involved in automatic comprehension and interpretation. Participants were found to use information from a prior relationally similar example in understanding the content of a currently encoded example. Further, in doing so they were sensitive to structural mappings between the two instances, ruling out explanations based solely on more general kinds of activation and application. Reading speed measures were used to demonstrate that these inferences were taking place during encoding rather than at later retrieval. These findings support the integration of sophisticated processes such as analogical mapping in a wide range of cognitive functions.|../mac/prof75.html",
    "Working Memory Span and Everyday Conditional Reasoning: A Trend Analysis...||Paper Track Working Memory Span and Everyday Conditional Reasoning: A Trend Analysis Wim De Neys , Department of Psychology, K.U.Leuven Wim.Deneys@psy.kuleuven.ac.be Walter Schaeken , Department of Psychology, K.U.Leuven Walter.Schaeken@psy.kuleuven.ac.be Géry d’Ydewalle , Department of Psychology, K.U.Leuven Géry.dYdewalle@psy.kuleuven.ac.be Department of Psychology, This study presents evidence for the role of working memory (WM) capacity in the retrieval and inhibition of counterexamples (alternatives and disablers) during everyday conditional reasoning. A total of 292 university students were given a measure of WM-capacity and a reasoning task with everyday, causal conditionals. Results show that the acceptance ratings of the logically valid Modus Ponens and Modus Tollens inferences follow a quadratic, U-shaped trend in function of WM-capacity, while acceptance ratings of the logically invalid Affirmation of the Consequent and Denial of the Antecedent inferences follow a negative linear trend. Findings support the claim that participants highest in WM-capacity spontaneously inhibit the disabler retrieval process during everyday reasoning.|../mac/prof76.html",
    "“Slow Mapping” in Children’s Learning of Semantic Relations...||Paper Track “Slow Mapping” in Children’s Learning of Semantic Relations Gedeon O. Deák , Department of Cognitive Science, UCSD deak@cogsci.ucsd.edu Jennifer Hughes Wagner , Department of Psychology and Human Development, Vanderbilt University To investigate how young children learn categorical semantic relations between words, 4- to 7-year-olds were taught four labels for novel categories in an “alien” microworld. After two play sessions, where each label was given, with defining information, at least 20 times, comprehension and production were tested. Results of two experiments show that 6-7-yearolds learned more words and correct semantic relations than 4-5-year-olds. The exclusion relation between contrasting category labels was easy to learn, and some findings suggested that hierarchical words are more easily learned than overlapping ones. Both studies showed no advantage to explicitly telling children semantic relations between words (e.g., “All fegs are wuddles.”). The results qualify a common assumption that preschool children have precocious abilities to infer word meaning; such an ability does not seem to extend to semantic relations between words.|../mac/prof77.html",
    "Contingent Behavior, Biases, and Adaptivity in Distributive Negotiation...||Paper Track Contingent Behavior, Biases, and Adaptivity in Distributive Negotiation Fabio Del Missier , ITC-IRST, Trento delmisfa@units.it Danilo Fum , Department of Psychology, University of Trieste fum@units.it Behavioral decision research on negotiation has obtained significant results but has not been able to provide yet a complete picture of the mechanisms underlying the definition, selection and usage of reference points, or to specify the processes responsible for the formulation and evaluation of the offers. We present the results of an experiment, requiring a bargain between a human buyer and a computer-simulated seller, in which we were able to obtain both the notorious biases caused by the initial offer and the framing effect, and an adaptive behavior deriving from a discounting policy. The effects of discount, initial offer and framing have been analyzed as a function of both the reference points and the anchors utilized by the participants in the negotiation phases. The experimental results and the linear models of the buyer support a contingent view of the negotiation behavior that affects the definition, selection and usage of the reference points.|../mac/prof78.html",
    "A Comparison of Statistical Models for the Extraction of Lexical Informatio...||Paper Track A Comparison of Statistical Models for the Extraction of Lexical Information from Text Corpora Simon Dennis , Institute of Cognitive Science, University of Colorado Simon.Dennis@colorado.edu The Syntagmatic Paradigmatic model (SP; Dennis & Harrington 2001, Dennis submitted) and the Pooled Adjacent Context model (PAC; Redington, Chater & Finch 1998) are compared on their ability to extract syntactic, semantic and associative information from a corpus of text. On a measure of syntactic class (and subclass) information based on the WordNet lexical database (Miller 1990), the models performed similarly with a small advantage for the PAC model. On a measure of semantic structure based on the similarities produced by Latent Semantic Analysis (LSA; Landauer & Dumais 1997), the models performed equivalently with a small advantage for the SP model. On a measure of associative information based on the free association norms of Nelson, McEvoy & Schreiber (1999), the SP model shows a substantive advantage over the PAC model producing more than twice as many associates.|../mac/prof79.html",
    "Cognitive Ethnography...||Plenary Track Cognitive Ethnography Edwin Hutchins University of California This talk presents cognitive ethnography as a method of cognitive science having theoretical grounding in distributed, embodied, situated cognition. I will describe cognitive ethnography and consider the role it may play in the development of the science of cognition informed by the themes of this conference.|../mac/prof8.html",
    "An Alignment-based Account of Serial Recall...||Paper Track An Alignment-based Account of Serial Recall Simon Dennis , Institute of Cognitive Science, University of Colorado Simon.Dennis@colorado.edu The task of serial recall has become the touchstone for theories of short term memory. Many simple yet powerful computational models have been proposed to account for performance in the task. However, these models typically make only tangential reference to language processes, despite the fact that a key determinant of performance on the task is the extent to which the to-be-learned list mimics the structure of natural language (Miller & Selfridge 1950). The Syntagmatic Paradigmatic model (SP; Dennis submitted, Dennis & Harrington 2001), a memory-based account of sentence processing, is applied to serial recall. Employing an alignment mechanism derived from String Edit Theory (SET; Sankoff & Kruskal 1983), the SP is able to account for the U shaped serial position curve, patterns of interlist and intralist intrusions and the multiply bowed serial position curve that occurs with grouped study lists.|../mac/prof80.html",
    "Independent Representation of Abstract Arguments and Relations...||Paper Track Independent Representation of Abstract Arguments and Relations Derek Devnich , Department of Psychology, University of California devnich@ucla.edu Greg T. Stevens , Department of Psychology, University of California stevens@psych.ucla.edu John E. Hummel , Department of Psychology, University of California jhummel@psych.ucla.edu Propositions specifying properties of, or relations among, one or more arguments form a central part of human mental representations. Representing a proposition entails binding each relational role to its argument. At the same time, computational considerations suggest that roles and arguments should be represented independently of one another in working memory (WM). We report an experiment using General Recognition Theory (Ashby & Townsend, 1986) to test the independence of relational roles from their arguments in WM. The results suggest that roles and arguments are independent in WM.|../mac/prof81.html",
    "Naïve Meanings of Force: Coherence vs. Fragmentation...||Paper Track Naïve Meanings of Force: Coherence vs. Fragmentation Andrea A. diSessa , Graduate School of Education, UC Berkeley disessa@soe.berkeley.edu Nicole M. Gillespie , Graduate School of Education, UC Berkeley ngillesp@uclink4.berkeley.edu Jennifer Esterly , Department of Psychology, CSU Stanislaus jesterly@csustan.edu This paper contributes to the literature on conceptual change by engaging in direct empirical comparison of contrasting views. We take up the question of whether naïve physics ideas are coherent or fragmented, building specifically on recent work supporting claims of coherence by Ioannides and Vosniadou (2002). A partial replication of the Ioannides and Vosniadou study resulted in radically different results. We analyze several possible reasons for the differences in our results, but find that none can plausibly account for the differences in our data. We argue that the results of our study undermine claims for coherence in naïve conceptualizations of force.|../mac/prof82.html",
    "Learning Grammatical Constructions in a Miniature Language from Narrated Vi...||Paper Track Learning Grammatical Constructions in a Miniature Language from Narrated Video Events Peter Ford Dominey , Institut des Sciences Cognitives, CNRS dominey@isc.cnrs.fr The objective of this research is to develop a system for miniature language learning based on a minimum of prewired language-specific functionality, that is compatible with observations of perceptual and language capabilities in human development. In the proposed system, meaning is extracted from video images based on detection of physical contact and its parameters. Mapping of sentence form to meaning is performed by learning grammatical constructions that are retrieved from a construction inventory based on the constellation of closed class items uniquely identifying the target sentence structure. The resulting system displays robust acquisition behavior that reproduces certain observations from developmental studies, with very modest “innate” language specificity.|../mac/prof83.html",
    "Explaining Color Term Typology as the Product of Explaining Color Te...||Paper Track ExplaiExplaining Color Term Typology as the Product of Explaining Color Term Typology as the Product of Cultural Evolution using a Bayesian Multi-agent Model Mike Dowman , School of Information Technologies, F09, University of Sydney Mike@it.usyd.edu.au An expression-induction model was used to simulate the evolution of basic color terms in order to test Berlin and Kay’s (1969) hypothesis that the typological patterns observed in basic color term systems are produced by a process of cultural evolution under the influence of universal aspects of human neurophysiology. Ten agents were simulated, each of which could learn color term denotations by generalizing from examples using Bayesian inference. Conversations between these agents, in which agents would learn from one-another, were simulated over several generations, and the languages emerging at the end of each simulation were investigated. The proportion of color terms of each type correlated closely with the equivalent frequencies found in the world color survey, and most of the emergent languages could be placed on one of the evolutionary trajectories proposed by Kay and Maffi (1999). The simulation therefore demonstrates how typological patterns can emerge as a result of learning biases acting over a period of time.|../mac/prof84.html",
    "Exploiting Cognitive Psychology Research for Recognizing Intention in Infor...||Paper Track Exploiting Cognitive Psychology Research for Recognizing Intention in Information Graphics Stephanie Elzer , Dept of Computer Science, University of Delaware elzer@cis.udel.edu Nancy Green , Dept of Mathematical Sciences, Univ. of N. Carolina nlgreen@uncg.edu Sandra Carberry , Dept of Computer Science, University of Delaware carberry@cis.udel.edu This paper outlines our approach to a novel application of plan inference, recognizing the intended message of an information graphic, focusing on how results from research by cognitive psychologists have been incorporated into the design of our system. Our work is part of a larger project to develop an interactive natural language system that provides an alternative means for individuals with sight-impairments to access the content of information graphics.|../mac/prof85.html",
    "The Misattribution of Relations in Similarity Judgments of Person Concepts...||Paper Track The Misattribution of Relations in Similarity Judgments of Person Concepts Zachary Estes , Department of Psychology, University of Georgia estes@uga.edu Roger Alix-Gaudreau , Artificial Intelligence Program, University of Georgia rodge@uga.edu We demonstrate an effect of the valence of a relation between person concepts on subsequent similarity judgments. In Experiment 1, the assertion of a positive relation between person concepts (e.g., The plumber protected the housewife.) increased their perceived similarity, whereas the assertion of a negative relation (e.g., The plumber alienated the housewife.) decreased their perceived similarity. In Experiment 2, positive relations (e.g., The plumber protected the housewife.) increased the perceived similarity of two unrelated person concepts (e.g., editors and taxi drivers), and negative relations decreased the similarity of unrelated persons. Experiment 3 ruled out a task demand explanation by showing that the effect is specific to similarity judgments. Results suggest that relations between person concepts are misattributed to the similarity of any two subsequently judged persons.|../mac/prof86.html",
    "“I Can’t See Your Eyes Well ‘Cause Your Nose is Too Short”:An Interactivity...||Paper Track “I Can’t See Your Eyes Well ‘Cause Your Nose is Too Short”:An Interactivity Account of Face Processing Reza Farivar , Dept. of Psychology, McGill University reza.farivar@mail.mcgill.ca Avi Chaudhuri , Dept. of Psychology, McGill University avi@hebb.psych.mcgill.ca The present work utilizes the generalized form of the signal detection theory (the General Recognition Theory) to formally model representation of faces during face perception. We tested the hypothesis that face perception, typically described as a holistic or configural process, can be formally described as an interactive processing of face parts, whereby one component of a face influences perception of other components. We present theoretical and experimental developments on this topic, building on previous work, but utilizing valid stimuli, a powerful mathematical model, and a crucial control condition.|../mac/prof87.html",
    "Understanding Interactive Graphical Communication...||Paper Track Understanding Interactive Graphical Communication Nicolas Fay , ATR Media Information Science Labs, 2-2-2 Hikaridai, Keihanna Science City nfay@atr.co.jp Simon Garrod , Department of Psychology, University of Glasgow simon@psy.gla.ac.uk John Lee , HCRC, University of Edinburgh j.lee@ed.ac.uk Jon Oberlander , HCRC, University of Edinburgh j.oberlander@ed.ac.uk We demonstrate several parallels between interactive verbal communication and graphical communication. Experiment 1 shows that through interaction partners’ graphical representations converge, and are refined, although degree of refinement is dictated by level of interaction. Experiment 2 shows that through interaction graphical representations lose their iconicity, taking on a more symbolic form. Again, this is dictated by the closeness of the interaction. Results are discussed both in terms of the evolution of writing systems and applications that support interactive graphical communication.|../mac/prof88.html",
    "Factors Involved in the Use of In and On...||Paper Track Factors Involved in the Use of In and On Michele I. Feist , Department of Psychology, Northwestern University m-feist@northwestern.edu Dedre Gentner , Department of Psychology, Northwestern University gentner@northwestern.edu What factors influence people’s use of spatial prepositions? In this paper, we examine the influence of four factors – geometry of the Ground, function of the Ground, animacy of the Ground, and animacy of the Figure – on the use of English in and on. We find evidence for all four of these factors. Spatial prepositions appear to involve a complex set of spatial and non-spatial interacting factors.|../mac/prof89.html",
    "The Cultural Origins of Human Cognition...||Plenary Track The Cultural Origins of Human Cognition Michael Tomasello Humans are biologically adapted for culture in ways that other primates are not. Uniquely human forms of social understanding and cultural learning emerge in human ontogeny at around one year of age as infants begin to engage with other persons in various kinds of joint attentional activities, including linguistic communication. Comparisons to the social cognition of apes helps to specify in more detail the nature of uniquely human social cognition and cultural learning.|../mac/prof9.html",
    "Mapping Self-Similar Structure: Commutative Expressions in Structure Mappin...||Paper Track Mapping Self-Similar Structure: Commutative Expressions in Structure Mapping Ronald W. Ferguson , College of Computing, Georgia Institute of Technology rwf@cc.gatech.edu Comparisons over commutative expressions constitute a special problem in analogical mapping. The unique character of these comparisons stems from the nature of commutative expressions themselves, which encapsulate particular dimensions of similarity between their arguments. In this paper, we describe commutative matching mechanisms for two analogical mapping systems (SME and MAGI), and demonstrate how use of these mechanisms can allow analogical mapping to scale over descriptions with lots of self-similar relational structure. We also discuss the cognitive implications of these mechanisms.|../mac/prof90.html",
    "Untitled Document||../mac/prof91.html",
    "Perception and Perspective in Robotics...||Paper Track Perception and Perspective in Robotics Paul Fitzpatrick , MIT Artificial Intelligence Laboratory paulfitz@ai.mit.edu To a robot, the world is a sea of ambiguity, in which it will sink or swim depending on the robustness of its perceptual abilities. But robust machine perception has proven difficult to achieve. This paper argues that robots must be given not just particular perceptual competences, but the tools to forge those competences out of raw physical experiences. Three important tools for extending a robot’s perceptual abilities whose importance have been recognized individually are related and brought together. The first is active perception, where the robot employs motor action to reliably perceive properties of the world that it otherwise could not. The second is development, where experience is used to improve perception. The third is interpersonal influences, where the robot’s percepts are guided by those of an external agent. Examples are given for object segmentation, object recognition, and orientation sensitivity; initial work on action understanding is also described.|../mac/prof92.html",
    "Augmented Naïve Bayesian Model of Classification Learning...||Paper Track Augmented Naïve Bayesian Model of Classification Learning Lewis Frey , Computer Science Department, Vanderbilt University frey@Vuse.Vanderbilt.Edu Douglas Fisher , Computer Science Department, Vanderbilt University dfisher@Vuse.Vanderbilt.Edu The Naï ve Bayesian Classifier and an Augmented Naï ve Bayesian Classifier are applied to human classification tasks. The Naï ve Bayesian Classifier is augmented with feature construction using a Galois lattice. The best features, measured on their within- and between-category overlap, are added to the category’s concept description. The results show that space efficient concept descriptions can predict much of the variance in the classification phenomena.|../mac/prof93.html",
    "A Bayesian Satisficing Model of Human Adaptive Planning...||Paper Track A Bayesian Satisficing Model of Human Adaptive Planning Wai-Tat Fu W. Fu, Department of Psychology, Carnegie Mellon University wfu@gmu.edu This paper presents a Bayesian satisficing model of when a problem-solver stops planning and begins acting. Existing knowledge about the environment is incrementally updated by new observations, and performance improves as a consequence of better knowledge about the environment. The model aims at bridging the gap between machine learning and cognitive science by adopting the rationality framework, which assumes that cognition tends to exploit the characteristics of the environment without engaging in psychologically implausible computations. An empirical study was conducted when human subjects learned to find the fastest path in a simple map. The model fit the human learning and performance well and provided insights into the mechanisms behind learning and performance in problem solving.|../mac/prof94.html",
    "Outcome Evaluation and Procedural Knowledge in Implicit Learning...||Paper Track Outcome Evaluation and Procedural Knowledge in Implicit Learning Danilo Fum , Department of Psychology, University of Trieste fum@units.it Andrea Stocco , Department of Psychology, University of Trieste stocco@units.it Although implicit learning has been considered in recent years as a declarative memory phenomenon, we show that a procedural model can better elucidate some intriguing and unexpected data deriving from experiments carried out with Sugar Factory (Berry & Broadbent, 1984), one of most popular paradigms in this area, and can account for other phenomena reported in the literature that are at o