Novel Insights into Statistical Learning

Abstract

The brain has a powerful capacity for statistical learning–an ability to detect statistical regularities in the environment in order to make predictions and guide behaviour–often without conscious awareness. It has been claimed that statistical learning plays a key role in a range of everyday perceptual and cognitive processes including those associated with language. There is demonstration that both spoken and written forms of natural language contain a rich statistical structure and there is demonstration that artificial languages are easier to learn when they contain statistical regularities. And yet, to date, there is very little empirical data showing a direct link between a capacity for statistical learning and proficiency with natural language. Here we report on a study of visual statistical learning that provides two important insights in the quest for such empirical evidence: this capacity follows a developmental trajectory in typically developing children, and it is correlated with reading ability.


Back to Thursday Papers