In this paper, we present a novel interpretation of the role of forgetfulness (i.e., memory impairment) in understanding game play in primates. Specifically, we examine how two primate species play a variant of the Wisconsin Card Sorting Task (WCST), a widely used clinical assessment game for measuring neurological and cognitive function. Our goal is to understand the role that memory plays in both learning and subsequently playing this game in humans and rhesus monkeys (Macaca mulatta). We are also interested in clustering these two populations based on their forgetfulness. This enables establishing baseline correspondences for cross-species comparison of memory function between different age groups, with the intention of enabling translational clinical treatments for a host of pathologies that involve memory dysfunction, such as senile dementia and Alzheimers disease. Doing so requires a more in-depth understanding of the role memory plays in cognitive tests like the WCST, which we provide here through computational modeling. We also show this model surprisingly provides a clear indication that learning of an unknown game has actually occurred. It thereby disputes earlier monkey studies on a variant of the WCST by providing evidence their subjects never actually learned to play the game on which they were being evaluated. The model also demonstrates that the effect of memory impairment on game performance is highly non-linear. We find memory degradation has little gradual effect; rather, it shows a steep response past a threshold value, which has strong implications for understanding the dynamics of human aging.