Modelling the effects of transcranial alternating current stimulation on the neural encoding of speech in noise
File(s)
Published version
Author(s)
Kegler, Mikolaj
Reichenbach, Johann
Type
Journal Article
Abstract
Transcranial alternating current stimulation (tACS) can non-invasively modulate neuronal activity in the cerebral cortex, in particular at the frequency of the applied stimulation. Such modulation can matter for speech processing, since the latter involves the tracking of slow amplitude fluctuations in speech by cortical activity. tACS with a current signal that follows the envelope of a speech stimulus has indeed been found to influence the cortical tracking and to modulate the comprehension of the speech in background noise. However, how exactly tACS influences the speech-related cortical activity, and how it causes the observed effects on speech comprehension, remains poorly understood. A computational model for cortical speech processing in a biophysically plausible spiking neural network has recently been proposed. Here we extended the model to investigate the effects of different types of stimulation waveforms, similar to those previously applied in experimental studies, on the processing of speech in noise. We assessed in particular how well speech could be decoded from the neural network activity when paired with the exogenous stimulation. We found that, in the absence of current stimulation, the speech-in-noise decoding accuracy was comparable to the comprehension of speech in background noise of human listeners. We further found that current stimulation could alter the speech decoding accuracy by a few percent, comparable to the effects of tACS on speech-in-noise comprehension. Our simulations further allowed us to identify the parameters for the stimulation waveforms that yielded the largest enhancement of speech-in-noise encoding. Our model thereby provides insight into the potential neural mechanisms by which weak alternating current stimulation may influence speech comprehension and allows to screen a large range of stimulation waveforms for their effect on speech processing.
Date Issued
2021-01-01
Date Acceptance
2020-10-01
Citation
NeuroImage, 2021, 224
ISSN
1053-8119
Publisher
Elsevier
Journal / Book Title
NeuroImage
Volume
224
Copyright Statement
©2020 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/ )
License URL
Sponsor
Engineering & Physical Science Research Council (EPSRC)
Engineering & Physical Science Research Council (E
Wellcome Trust
National Institute for Health Research
US Army (US)
Engineering & Physical Science Research Council (EPSRC)
Grant Number
EP/R032602/1
EP/M026728/1
108295/Z/15/Z
RDA26
W911NF1910396
EP/T020970/1
Subjects
Computational modelling
Speech processing
Spiking neural networks
Transcranial alternating current stimulation
Neurology & Neurosurgery
11 Medical and Health Sciences
17 Psychology and Cognitive Sciences
Publication Status
Published
Article Number
ARTN 117427
Date Publish Online
2020-10-07