2014 — 2016 |
Cohen, Michael [⬀] |
F32Activity Code Description: To provide postdoctoral research training to individuals to broaden their scientific background and extend their potential for research in specified health-related areas. |
Anatomical Constraints On Cognition and How They Develop @ Massachusetts Institute of Technology
DESCRIPTION (provided by applicant): Human visual cognition is known to have strict capacity limitations: we can only perceive and remember a few items at a time. What is the nature of these limitations? For several years, researchers have attempted to answer this question using a variety of well-known behavioral paradigms (change blindness, inattentional blindness, etc.). While this work has produced a variety of important findings, surprisingly little work has been done to relate the capacity limits of perception and memory to the underlying neural architecture of the visual regions involved in representing visual objects (e.g ventral visual cortex). Several decades of neuroscience research have shown that high-level objects (e.g. faces, bodies, scenes, and objects) are represented in the ventral visual cortex by distinct neural networks that are often non-overlapping. Thus, it is natural to wonder if these dissociable neural structures correspond to distinct pools of cognitive resources. Recently, as part of my doctoral thesis, I found evidence in support of this idea by showing that people can store more information in visual working memory if that information comes from different categories (e.g. faces and scenes) rather than the same category (e.g. only faces). Moreover, the size of the benefit from presenting different categories was correlated with the extent to which the neural regions involved in representing those categories overlapped in ventral visual cortex. The goals of the proposed research are 1) to more thoroughly investigate the relationship between neural organization and perceptual abilities and 2) to examine how this relationship between anatomy and behavior develops. In Aim 1, we will first measure the representational structure of different regions within the visual system (e.g. occipitotemporal cortex, early visual cortex, etc.) in response to a wide range of stimulus categories (faces, bodies, phones, hammers, cats, etc.). Then we will see if performance with multiple perceptual tasks (e.g. object categorization and visual search) can be predicted by the organization of this structure. In Aim 2, we will repeat the same procedures as those in Aim 1, but will do so with adults and children ages 5-7. From this, we will ask if the representational structure of the visual system is the same in children and adults, and if that structure imposes the same constrain on perception across the two groups. This work will provide insight into the limitations of human cognition and shed light on how those limitations develop. Furthermore, understanding of the developmental time course of the visual system and its relation to behavior in children could serve as useful benchmarks that could be used in clinical settings to understand and diagnose children with developmental disorders that are known to affect the visual system (e.g. autism, dyslexia, etc.).
|
0.915 |