Back to search

University of British Columbia

Two-component somatotopic template for speech in electrocorticography

Abstract

dc:description

The precise spatiotemporal organization of β (~12–35 Hz), γ (~35–70 Hz), and higher-frequency Γ (~70–140 Hz) oscillations in the sensorimotor cortex (SMC) during speech remains unclear, particularly their movement activation and inhibition roles. Using multi-session high-density electrocorticography (ECoG) from four subjects performing a consonant-vowel task, I compare SMC β, γ, and Γ bandpowers over time. Peak and SMC bandpowers are compared as energies during speech (activation) and rest (inhibition) to assess similarity and overlap. I apply principal component analysis (PCA) to identify a reduced template of β/γ/Γ activity, correlating principal components with cortical speech activity and comparing low-dimensional reconstructions to original observations. I show somatotopic differences between SMC β/γ/Γ oscillations, observing frequency-specific changes local to distinct areas during speech. Similarity/overlap differ significantly between β and Γ (DSC ≈ 10–14%); γ shows mixed results, conditionally resembling higher or lower frequencies. PCA reveals a significant two-component speech model in β/γ/Γ. The first γ/Γ principal component spatially localizes to somatotopic speech areas during vocalization, suggesting movement activation. Simultaneously, the first β component displays larger, non-overlapping, and temporally anti-correlated activity, suggesting movement inhibition. PCA also identifies moderate-to-strong correlations between the second β and Γ components (r ≈ 0.70 ± 0.20), indicating novel β–Γ cross-frequency coupling. For all subjects, spatiotemporal reconstructions necessitate only two principal components (r ≈ 0.95; NRMSE ≈ 7%). Adding a third component makes insignificant improvements (∆r ≈ 0.01; ∆NRMSE ≈ 1%), showing negligible correlation with speech-SMC activity; explained variance and SMC information are also insignificantly improved. My analysis identifies a spatially distinct two-component SMC speech movement template. I demonstrate the sufficiency of two components to represent speech-SMC dynamics, paralleling the dual activation-inhibition roles of β and γ/Γ. Finally, I discuss bridging observed cortical activity to emerging whole-body motor integration models, contrasting traditional views of the SMC, and I discuss the potential to detect similar templates with electroencephalography (EEG).

Degree

thesis:*
Name thesis:degree_name
Master of Applied Science - MASc
Level thesis:degree_level
master's
Discipline thesis:degree_discipline
Electrical and Computer Engineering
Grantor dc:publisher
University of British Columbia
Year dc:date
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Easthope, Eric
Contributors dc:contributor
  • Fels, Sidney

Rights

dc:rights
Statement dc:rights
  • Attribution-NonCommercial-NoDerivatives 4.0 International
Language dc:language
eng

Identifiers

dc:identifier.*
Handle dc:identifier
http://hdl.handle.net/2429/93484

Chain of custody

source
Harvested from
University of British Columbia
Base URL
circle.library.ubc.ca/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Easthope, Eric. Two-component somatotopic template for speech in electrocorticography. master's thesis, University of British Columbia, 2025. http://hdl.handle.net/2429/93484