Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Receptive language proficiency relates to cortical tracking of continuous speech in noise in children
Poster Session D, Saturday, October 26, 10:30 am - 12:00 pm, Great Hall 3 and 4
Amanda Hampton Wray1, Ashley Parker1, Jacie R. McHaney2, Brittany Coleman1, Abby Block1, Bharath Chandrasekaran2; 1University of Pittsburgh, 2Northwestern University
Receptive language refers to an individual’s ability to process and comprehend language. It is one of the earliest skills a child learns, typically emerging in the first year of life, and continuing to develop into adulthood. It is traditionally measured using behavioral (“tabletop”) tasks, where a child may be tested on their ability to understand words and sentences, relationships between words, grammatical rules, and comprehending and following directions. These tests require coordination between receptive language and other cognitive skills, including working memory. Thus, when a child struggles with receptive language, it is not always clear at which processing stage(s) breakdown may occur. Neurophysiology, specifically neural tracking of temporal speech envelopes, is a technique for examining the neural encoding of continuous speech; it can be considered a way of assessing receptive language while limiting some other cognitive demands. Previous work has shown a relationship between neural tracking of continuous speech in quiet and receptive language in preschool-aged children. However, suboptimal, complex listening environments are known to be particularly challenging for children. The degree to which receptive language skills support speech/language perception in complex listening environments in children with intact, though still developing, language skills is not yet clear. In the current study, we used electroencephalography (EEG) to track temporal speech envelopes and administered a battery of speech and language measures to examine the relationships between receptive language proficiency and neural tracking of continuous speech both in quiet and in complex listening environments. Children (n=30, 6-12 years old) with normal audiometric thresholds and no developmental or communication disorders completed subtests of the Clinical Evaluation of Language Fundamentals (CELF-5), Sentence Comprehension, Word Classes, and Following Directions, to derive the Receptive Language Index. Neural tracking of continuous speech was measured using EEG. Participants listened to a target audiobook across two conditions: (1) in quiet, and (2) in the presence of another talker (a different audiobook narrated by a distinct speaker). We used a multivariate temporal response function (mTRF) method where models were used to evaluate the extent to which acoustics of both the target and the masker story contributed to the predictive power of the recorded EEG (“prediction accuracy”). Receptive Language Index standard scores ranged from 89 to 135 on the CELF-5 (mean=112.25). Prediction accuracy (a z-transformed Pearson’s correlation between the predicted and measured EEG responses) was stronger in quiet than in noise, with a high degree of individual variability in each condition. We also found a significant effect of language proficiency on neural tracking in noise (r=0.482, p=0.015). This relationship was not found in neural tracking during the quiet condition. Children with stronger receptive language skills had higher levels of prediction accuracy as measured by neural tracking of continuous speech, specifically in noise. These early results may suggest that receptive language skills in school-aged children may not have a direct relationship with neural tracking in simpler, quiet conditions, but that stronger language skills become critical when the listening environments are complex, with competing speech signals.
Topic Areas: Language Development/Acquisition, Control, Selection, and Executive Processes