Murdoch University Research Repository

Welcome to the Murdoch University Research Repository

The Murdoch University Research Repository is an open access digital collection of research
created by Murdoch University staff, researchers and postgraduate students.

Learn more

Toward an individualized neural assessment of receptive language in children

Petit, S., Badcock, N.A., Grootswagers, T., Rich, A.N., Brock, J., Nickels, L., Moerel, D., Dermody, N., Yau, S., Schmidt, E. and Woolgar, A. (2020) Toward an individualized neural assessment of receptive language in children. Journal of Speech, Language, and Hearing Research, 63 (7). pp. 2361-2385.

[img]
Preview
PDF - Published Version
Download (2MB) | Preview
Free to read: https://doi.org/10.1044/2020_JSLHR-19-00313
*No subscription required

Abstract

Purpose

We aimed to develop a noninvasive neural test of language comprehension to use with nonspeaking children for whom standard behavioral testing is unreliable (e.g., minimally verbal autism). Our aims were threefold. First, we sought to establish the sensitivity of two auditory paradigms to elicit neural responses in individual neurotypical children. Second, we aimed to validate the use of a portable and accessible electroencephalography (EEG) system, by comparing its recordings to those of a research-grade system. Third, in light of substantial interindividual variability in individuals' neural responses, we assessed whether multivariate decoding methods could improve sensitivity.

Method

We tested the sensitivity of two child-friendly covert N400 paradigms. Thirty-one typically developing children listened to identical spoken words that were either strongly predicted by the preceding context or violated lexical–semantic expectations. Context was given by a cue word (Experiment 1) or sentence frame (Experiment 2), and participants either made an overall judgment on word relatedness or counted lexical–semantic violations. We measured EEG concurrently from a research-grade system, Neuroscan's SynAmps2, and an adapted gaming system, Emotiv's EPOC+.

Results

We found substantial interindividual variability in the timing and topology of N400-like effects. For both paradigms and EEG systems, traditional N400 effects at the expected sensors and time points were statistically significant in around 50% of individuals. Using multivariate analyses, detection rate increased to 88% of individuals for the research-grade system in the sentences paradigm, illustrating the robustness of this method in the face of interindividual variations in topography.

Conclusions

There was large interindividual variability in neural responses, suggesting interindividual variation in either the cognitive response to lexical–semantic violations and/or the neural substrate of that response. Around half of our neurotypical participants showed the expected N400 effect at the expected location and time points. A low-cost, accessible EEG system provided comparable data for univariate analysis but was not well suited to multivariate decoding. However, multivariate analyses with a research-grade EEG system increased our detection rate to 88% of individuals. This approach provides a strong foundation to establish a neural index of language comprehension in children with limited communication

Item Type: Journal Article
Murdoch Affiliation: School of Psychology and Exercise Science
Publisher: American Speech-Language-Hearing Association
Copyright: © 2020 American Speech-Language-Hearing Association
URI: http://researchrepository.murdoch.edu.au/id/eprint/57065
Item Control Page Item Control Page

Downloads

Downloads per month over past year