Date of Award

1-1-2025

Degree Type

Dissertation

Degree Name

Doctor of Philosophy in Electrical Engineering

Specialization

Biomedical Engineering

Department

Electrical, Computer, and Biomedical Engineering

First Advisor

Valda Shahriari

Abstract

Electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) are complementary noninvasive neuroimaging modalities that have contributed greatly to our understanding of the mechanisms of auditory processing. As an emerging technology, fNIRS has produced promising insights into the mechanisms of auditory processing, although considerable work remains to characterize hemodynamic responses in auditory tasks. In addition, the complementary nature of EEG and fNIRS and the relationships between the signals they record have been underexplored. In particular, nonlinear interactions between these signals remain under-characterized, despite the potential benefit to future multimodal neuroimaging studies. A challenge in understanding these interactions is the influence of systemic physiology on the recorded signals, especially fNIRS, which can independently interact with electrocortical signal components. Addressing these challenges would provide new insights in the context of mulitmodal neuroimaging of auditory processing.

One potential application of state-of-the-art multimodal neuroimaging approaches is the characterization of variations in auditory-related electro-vascular interactions along the schizotypy continuum. Schizotypy is a personality construct consisting of traits similar to the symptoms of schizophrenia, and its study has provided insights into the etiology of schizophrenia, the risk of development of psychosis, and the healthy expression of related traits. Developing a deeper understanding of electro-vascular interactions and their variation along the schizotypy continuum could contribute to our understanding of auditory processing and inform parameterization of noninvasive neuromodulation approaches, such as transcranial direct current stimulation (tDCS).

In this study, we aim to 1) characterize individual-specific changes in hemodynamic signals measured using fNIRS to auditory stimuli; 2) explore the interactions between EEG and fNIRS in healthy populations in the context of auditory processing, and their mutual interactions with systemic physiology; 3) identify the cross-modal causal interactions in signatures of auditory processing as measured by EEG and fNIRS and 4) investigate the use of tDCS to modulate responses related to auditory processing.

Our current findings have demonstrated that hemodynamic signal slope as measured by fNIRS is correlated with task information in an auditory oddball task, and that this correlation can be improved by using an independent-component analysis (ICA) -based spatial filtering approach. We have also demonstrated global phase-amplitude coupling (PAC) between respiration effort signals and spectral features of EEG signals in multiple frequency bands (α, β, γ), highlighting the role of systemic physiology in cortical oscillations. We extend these findings by measuring PAC between fNIRS and EEG before and after applying a temporally embedded canonical correlation analysis (tCCA)-generalized linear modeling (GLM) approach to correct fNIRS signals using independent measurements of systemic physiology and demonstrating PAC in both cases. We additionally explored causal interactions between EEG and fNIRS signals during an auditory processing task, finding significantly elevated causal connections from the hemodynamic signals in the right auditory region to the 40 Hz EEG amplitude envelope during the 40 Hz auditory steady-state response (ASSR) than at rest. Findings from the same analysis identified heterogeneity in causal interactions across fNIRS signals from different regions and EEG oscillatory activity in the frontocentral region potentially attributable both to systemic and cortical hemodynamic components, with stronger interactions in the direction originating from fNIRS signals to EEG signals. We additionally observed no association between schizotypy scores on ASSR evoked power, inter-trial coherence, or task-related regressor fit in a general linear model (GLM) fNIRS analysis pipeline. Finally, we did not observe any modulation of EEG or fNIRS signals recorded during the ASSR task using tDCS, nor differences in responses resulting from different stimulation configurations.

These results reflect the utility of innovative nonlinear multimodal approaches to noninvasive neuroimaging data recorded during auditory processing, as well as novel characterizations of auditory neural responses of interest. Though no differences were observed along the schizotypy continuum in the metrics reported, these results are informative regarding the extension of disruptions of auditory processing to sub-clinical scales. Finally, the lack of modulation of the ASSR using tDCS could indicate that other neuromodulatory approaches/configurations may be more effective.

Download

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.