The ultimate goal of the proposed research is to better understand the neural mechanisms underlying organization and storage of semantic memory. Elucidating the development and organization of semantic memory remains a critical problem in cognitive neuroscience, and various theories attempt to capture the mapping of conceptual representations for everyday objects onto their neural correlates. One prevalent model of semantic memory suggests that distributed object representations correspond to modality-specific sensorimotor brain areas. Numerous behavioral, neuropsychological, and neuroimaging studies have supported this model, but it remains an open question as to how varied sensory experience helps to shape long-term memory of objects. Specifically, feature diagnosticity may play an important role in forming the richness of an object's representational structure. This proposal seeks to investigate the relationship between experience and memory by training subjects on a set of novel objects, thus building controlled representations of the objects. Subjects will either learn a set of objects for which both color and shape are diagnostic, or for which only shape is a diagnostic feature. In Experiment 1,1 will examine behavioral changes in novel object representations through an object recognition task. I will also see how these changes vary as a function of learning style. Experiment 2 offers a direct test (through traditional univariate analyses) for neural correlates of feature diagnosticity by employing neuroimaging (fMRI) to investigate how diagnostic features alter neural representations of learned objects. Here, the extent to which diagnostic features modulate activity in brain regions responsible for perception of these features will be examined. Experiment 3 probes behavioral (multidimensional scaling) and neural (multi-voxel pattern analysis) representations of the similarity space and gradient of the newly learned objects, predicting significant correlations for the two representations in brain regions responsible for perceiving the diagnostic features. This proposal explores the role of feature diagnosticity in interactions of experience and knowledge to both shed light on the neural bases for visual knowledge organization and aid in understanding the nature of semantic knowledge deficits. Deficits in conceptual knowledge are associated with several disorders, including Alzheimer's Disease and Frontotemporal Dementia. A deeper knowledge of the neural bases underlying the organization and storage of conceptual knowledge will allow us to better understand the manifestation of specific conceptual knowledge deficits, which will in turn yield insight in informing patient treatments.

Agency
National Institute of Health (NIH)
Institute
National Institute on Aging (NIA)
Type
Predoctoral Individual National Research Service Award (F31)
Project #
5F31AG034743-03
Application #
8128429
Study Section
Special Emphasis Panel (ZRG1-F12A-E (20))
Program Officer
Wagster, Molly V
Project Start
2009-09-25
Project End
2012-05-31
Budget Start
2011-09-25
Budget End
2012-05-31
Support Year
3
Fiscal Year
2011
Total Cost
$17,734
Indirect Cost
Name
University of Pennsylvania
Department
Neurosciences
Type
Schools of Medicine
DUNS #
042250712
City
Philadelphia
State
PA
Country
United States
Zip Code
19104
Hsu, Nina S; Schlichting, Margaret L; Thompson-Schill, Sharon L (2014) Feature diagnosticity affects representations of novel and familiar objects. J Cogn Neurosci 26:2735-49