Perceptual Science Series
Many kinds of visually-guided behavior must integrate perception, working memory, and long-term memory. For example, selecting the ripest tomato requires an observer to compare the color of a tomato in view to the memory of recently seen tomatoes and the stored memory of ripe tomato color. We argue that it is fruitful to understand color perception and memory in one probabilistic estimation framework. In such frameworks, observers arrive at estimates of object properties by combining sensory information about object properties with prior knowledge about what object properties are likely to be present. Results from several studies are consistent with the idea that both working and long-term memory act to increase sensory uncertainty; color estimates in memory exhibit biases that can be explained by the influence of short-term priors, such as characteristics of the experimental stimulus ensemble, and long-term priors, such as color categories.