Rapid Enhancement of Subcortical Neural Responses to Sine-Wave Speech

Abstract

The efferent auditory nervous system may be a potent force in shaping how the brain responds to behaviorally significant sounds. Previous human experiments using the frequency following response (FFR) have shown efferent-induced modulation of subcortical auditory function online and over short- and long-term time scales; however, a contemporary understanding of FFR generation presents new questions about whether previous effects were constrained solely to the auditory subcortex. The present experiment used sine-wave speech (SWS), an acoustically-sparse stimulus in which dynamic pure tones represent speech formant contours, to evoke FFRSWS. Due to the higher stimulus frequencies used in SWS, this approach biased neural responses toward brainstem generators and allowed for three stimuli (/bɔ/, /bu/, and /bo/) to be used to evoke FFRSWSbefore and after listeners in a training group were made aware that they were hearing a degraded speech stimulus. All SWS stimuli were rapidly perceived as speech when presented with a SWS carrier phrase, and average token identification reached ceiling performance during a perceptual training phase. Compared to a control group which remained naïve throughout the experiment, training group FFRSWS amplitudes were enhanced post-training for each stimulus. Further, linear support vector machine classification of training group FFRSWS significantly improved post-training compared to the control group, indicating that training-induced neural enhancements were sufficient to bolster machine learning classification accuracy. These results suggest that the efferent auditory system may rapidly modulate auditory brainstem representation of sounds depending on their context and perception as non-speech or speech.

Brainstem speech encoding is dynamically shaped online by fluctuations in corticalαstate

“…To further assess the behavioral relevance of α -FFR modulations to SIN listening, we investigated vowel decoding of FFRs at low vs. high α states. Similar ML approaches have been applied to speech-evoked FFRs to decode stimulus classes, e.g., Mandarin lexical tones (Llanos, Xie and Chandrasekaran, 2017; Xie et al, 2019) and speech tokens (Xie, Girshick, Dollár, Tu and He, 2017; Cheng, Xu, Gold and Smith, 2021). In these studies, decoding performance in correctly classifying FFRs is used as an objective measure of speech discrimination.…”

Perceptual warping exposes categorical representations for speech in human brainstem responses

“…Tk4) (Carter et al, 2022). Previous work has also demonstrated that changes in perception can drive enhancements of the FFR (Cheng et al, 2021), suggesting speech processing is influenced by predictions of the percept. Our response-to-response correlations and neural decoding results support perceptual encoding in the FFR.…”

“…Such findings could be explained by long term, experience-dependent plasticity (Krishnan et al, 2012). This evidence is further bolstered by findings of categorization-training studies that show once individuals learn to identify novel speech stimuli their FFR are enhanced relative to more novice listening states (Cheng et al, 2021;Reetzke et al, 2018).…”

Translational Applications of Machine Learning in Auditory Electrophysiology

“…The abovementioned rationale motivated a recent study by Cheng and colleagues 9 in which subcortical AEPs were evoked by three acoustically sparse speech stimuli known as sine wave speech (SWS). SWS replaces the dynamic formant structure of a speech signal with two or three time-varying sinusoids, and all other speech content is discarded ( Fig.…”

“…8 Do we want to know whether auditory training has positively impacted the function of the auditory nervous system? 9 Do we want to use OAEs and ABRs to explore hidden structures in our datasets that allow us to identify which patients are at risk for noise-induced hearing loss, synaptopathy, and/or tinnitus? 10 Do we want hearing aids or cochlear implants to learn which acoustic signals are important to the user, how to function in different environments, and to be "cognitively steered" via a brain-computer interface?…”