Visual working memory (VWM) underlies humans'ability to perceive the world as a coherent visual experience rather than a disjointed series of isolated views. Decades of research has established a hallmark of the VWM system is its limited capacity-the ability to maintain only 3-5 items at once. The nature of VWM capacity limitations is currently under debate in the literature, and no clearly dominant account of how capacity might be realized in the human brain has emerged. The objective of this proposal is to provide such an account by investigating development of the VWM system in early childhood, paired with implementation in a neurally- grounded computational model. The proposal tests two specific predictions of the model: first, that memory for colors becomes more precise over development;second, that metric interactions between items held in memory will show a u-shaped developmental trajectory. These predictions will be tested in two computerized tasks designed for use with 3- to 6-year-old children. A color discrimination task will test the precision of children's memory, and a multi-item change detection task will test for interactions between similar colors held simultaneously in memory. The behavioral and computational work in this proposal will provide a new explanation for adults'capacity limitations in VWM, offering the first comprehensive developmental account of how capacity limits arise and change in early childhood. The long-term goal of this research is to uncover causal mechanisms of developmental change in working memory and to use this knowledge to inform research on related neurological and/or behavioral impairments. A number of disorders-including autism, Williams Syndrome, schizophrenia, attention deficit hyperactivity disorder, Alzheimer's and other dementias-lead to divergence from typical patterns of performance in laboratory tasks tapping visual and spatial working memory. In order to understand neurological differences in these disordered populations, we must first understand how the typically-developing brain perceives, remembers, and uses visual information in service of behavior.
The proposed research studies the visual memory in early childhood within typically-developing populations. By understanding this system better in the majority of the population, future research may address how this system differs with disorders-including autism, Williams Syndrome, schizophrenia, attention deficit hyperactivity disorder, Alzheimer's and other dementias-that lead to atypical patterns of performance in related tasks.