Working Memory for Natural Scenes

Visual working memory plays a fundamental role in visual perception and visually guided behavior, and much has been learned about the nature of this memory system by studies using arrays of artificial but easily controlled stimuli (e.g., arrays of colored squares). Indeed, task performance with these artificial stimuli is strongly predictive of overall cognitive ability and is markedly reduced in schizophrenia and other disorders. However, current quantitative models of visual working memory based on these artificial stimuli cannot be readily extended to the kinds of complex, structured scenes that humans face in daily life. In this talk, I will describe a new model of the representation of natural visual scenes called the population vector model.

In our model, a scene is represented in visual working memory as a noisier version of the pattern of activation that was produced during the perception of that scene. We model this by feeding the scene into a deep neural network model of the ventral object recognition pathway and using the resulting pattern of activation across the population of units (the population vector) as a model of the working memory representation. We have tested this model using both behavioral and electrophysiological paradigms, and we find that it is remarkably successful at predicting both behavioral performance and scalp EEG activity. Although the model is far from perfect, it is a first step toward understanding how natural scenes are stored in working memory.

Dr. Steve Luck is a Distinguished Professor at the University of California, Davis.
 

NACS Seminars are free and open to the public.