Neural Mechanisms for Combinatorial Semantics in Language and Vision: Evidence From FMRI, Patients, and Brain Stimulation

Loading...
Thumbnail Image
Degree type
Doctor of Philosophy (PhD)
Graduate group
Neuroscience
Discipline
Subject
brain imaging
brain stimulation
compositionality
conceptual combination
semantic integration
semantic memory
Neuroscience and Neurobiology
Psychology
Funder
Grant number
License
Copyright date
2016-11-29T00:00:00-08:00
Distributor
Related resources
Contributor
Abstract

Throughout our daily experience, humans make nearly constant use of semantic knowledge. Over the last 20-30 years, the majority of work on the neural basis of semantic memory has examined the representation of semantic categories (e.g., animate versus inanimate). However, a defining aspect of human cognition is the ability to integrate this stored semantic information to form complex combinations of concepts. For example, humans can comprehend “plaid” and “jacket” as separate concepts, but can also effortlessly integrate this information to create the idea of a “plaid jacket.” This process is essential to human cognition, but little work has examined the neural regions that underlie conceptual combination. Many models of semantic memory have proposed that convergence zones, or neural hubs, help to integrate the semantic features of word meaning to form coherent representations from stored semantic knowledge. However, few studies have specifically examined the integrative semantic functions that these high-level hub regions carry out. This thesis presents three experiments that examine lexical-semantic combinatorial processing (as in the “plaid jacket” example above): 1) a study in healthy adults using fMRI, 2) a study in healthy adults using brain stimulation, and 3) a study examining impairments of lexical-semantic integration in patients with neurodegenerative disease. The fourth and final experiment of this thesis examines semantic aspects of combinatorial codes for visual-object representation. This study identifies neural regions that encode the feature combinations that define an object’s meaning. The findings from these four experiments elucidate specific cortical hubs for semantic-feature integration during language comprehension and visual-object processing, and they advance our understanding of the role of heteromodal brain regions in semantic memory.

Advisor
Murray Grossman
Date of degree
2016-01-01
Date Range for Data Collection (Start Date)
Date Range for Data Collection (End Date)
Digital Object Identifier
Series name and number
Volume number
Issue number
Publisher
Publisher DOI
Journal Issue
Comments
Recommended citation