Leading and Following: Noise Differently Affects Semantic and Acoustic Processing during Naturalistic Speech Comprehension

Abstract

While our daily verbal communication is challenging with unavoidable environmental noise, the neural mechanisms underlying speech-in-noise comprehension remain to be elucidated. The present study investigated the neural tracking of acoustic and semantic speech information during noisy naturalistic speech comprehension. Participants listened to narrative audios mixed with spectrally matched stationary noise at three signal-to-ratio (SNR) levels and 60-channel electroencephalography signals were recorded. A temporal response function (TRF) method was employed to derive event-related-like responses to the continuous speech stream at both the acoustic and the semantic levels. Whereas the amplitude envelope of the naturalistic speech was taken as the acoustic feature, word entropy and word surprisal were extracted via the natural language processing method as two semantic features. Theta-band frontocentral TRF responses to the acoustic feature were observed at around 400 ms following speech fluctuation onset over all three SNR levels, and the response latency was more delayed with increasing SNR. Delta-band frontal TRF responses to the semantic feature of word entropy were observed at around 200 to 600 ms leading to speech fluctuation onset over all three SNR levels, and the response latency became more leading with increasing SNR. While the following responses to speech acoustics were consistent with previous studies, the leading responses to speech semantics suggest our brain could be taking an active approach to deal with noisy speech comprehension.

Compensatory Mechanisms for Preserving Speech-in-Noise Comprehension Involve Prefrontal Cortex in Older Adults

“…Despite numerous reports of its activation in the older adults, it is unclear how the prefrontal cortex was involved to offset declines in the language-specific areas (7). One approach to resolve this question is by modeling the neural responses to the speech features (from acoustics to semantics) in natural speech, which could be quantitatively estimated by recently advanced natural language processing algorithms (8)(9)(10)(11)(12). Yet, the inherence complexity of speech dynamics and the interdependency of multi-level speech features make it challenging to directly model such a brain-to-stimuli correspondence (13,14), calling for new computational methods to explore how the noisy speech is processed in the prefrontal cortex.…”