Access Type

Open Access Dissertation

Date of Award

January 2013

Degree Type

Dissertation

Degree Name

Ph.D.

Department

Psychiatry and Behavioral Neurosciences

First Advisor

Eishi Asano

Abstract

The work presented as part of this dissertation represents a multi-modality study of language structure and function. The primary functional modality employed is task-related electrocorticography (ECoG). This is complemented by discussion and evaluation of previously published functional magnetic resonance imaging (fMRI) data. Language-related structure is explored using diffusion-weighted magnetic resonance imaging in conjunction with ECoG data. The scientific questions pursued are broad and include reevaluation of previously proposed theories.

We start by taking the first steps in validating our naming-related ECoG approach by comparing our results from a small cohort of patients to the clinical gold-standard technique of electrical brain stimulation. This evaluation begins to address a clinical problem involving the insensitivity of electrical brain stimulation in language mapping of young children. Thus, our patients across all studies are a mixture of children, adolescents, and adults. Combining data presented within this thesis, data from other members of our team, and published data from teams at other institutions, evidence suggests that language-related ECoG mapping is a powerful language mapping tool when it is employed with an appropriate task. The task employed here is a now well-studied auditory descriptive naming task.

Language-related ECoG is then utilized to dissect language function mechanistically employing contrast tasks alongside the descriptive naming task. Working memory and language functions of the frontal lobe are dissected and conclusions are drawn to shed light on their degree of overlap and interaction during ongoing language processing. Evidence of secondary auditory processing and language comprehension gained from other modalities is reevaluated. In particular, reverse speech and signal-correlated noise are employed and evaluated as control tasks for non-language-specific auditory function. A discrepancy between language-related ECoG and language-related fMRI is discovered in regards to the use of reverse speech as such a control task. It is found that signal correlated noise may be more reliable in identifying non-language auditory functions of the temporal lobe.

Age-old questions of language-related connectivity are explored by combining diffusion-weighted magnetic resonance imaging tractography with language-related ECoG findings to evaluate terminations of the arcuate fasciculus. Results support recent evidence suggesting that the precentral gyrus is an important termination of this language-related white matter pathway. New models modifying century-old, entrenched models are evaluated in light of these findings and proposals for follow-up work that may create further clarity are provided.

Finally, the thesis rounds out with a study exploring the effects of focal interictal epileptiform activity upon ongoing language processes; contributing beyond the neuroscience of language to the epilepsy literature, in honor of the patients providing the data for these studies. Our data demonstrates that such localized pathological activity can have clinically imperceptible effects upon language functions, suggesting one possible mechanism toward cognitive deficits frequently reported in such patients.

Share

COinS