Neural Measures of Pitch Processing in EEG Responses to Running Speech

Abstract

Linearized encoding models are increasingly employed to model cortical responses to running speech. Recent extensions to subcortical responses suggest clinical perspectives, potentially complementing auditory brainstem responses (ABRs) or frequency-following responses (FFRs) that are current clinical standards. However, while it is well-known that the auditory brainstem responds both to transient amplitude variations and the stimulus periodicity that gives rise to pitch, these features co-vary in running speech. Here, we discuss challenges in disentangling the features that drive the subcortical response to running speech. Cortical and subcortical electroencephalographic (EEG) responses to running speech from 19 normal-hearing listeners (12 female) were analyzed. Using forward regression models, we confirm that responses to the rectified broadband speech signal yield temporal response functions consistent with wave V of the ABR, as shown in previous work. Peak latency and amplitude of the speech-evoked brainstem response were correlated with standard click-evoked ABRs recorded at the vertex electrode (Cz). Similar responses could be obtained using the fundamental frequency (F0) of the speech signal as model predictor. However, simulations indicated that dissociating responses to temporal fine structure at the F0 from broadband amplitude variations is not possible given the high co-variance of the features and the poor signal-to-noise ratio (SNR) of subcortical EEG responses. In cortex, both simulations and data replicated previous findings indicating that envelope tracking on frontal electrodes can be dissociated from responses to slow variations in F0 (relative pitch). Yet, no association between subcortical F0-tracking and cortical responses to relative pitch could be detected. These results indicate that while subcortical speech responses are comparable to click-evoked ABRs, dissociating pitch-related processing in the auditory brainstem may be challenging with natural speech stimuli.

Predictors for estimating subcortical EEG responses to continuous speech

“…It must be noted that although the wave V peak was used as the primary metric of performance, the conventional click ABR consists of several other morphological features [36]. The wave V peak was selected here to both be consistent with prior work [6, 19, 16], and because it was the only consistent feature that was detected in all subjects. TRFs using ZIL had shorter wave V peak latencies (see Figures 1 and 3).…”

“…TRFs were estimated from EEG data recorded from 24 participants listening to continuous speech. Prior work indicates that the most prominent feature of subcortical TRFs is the wave V peak [6, 16]. This peak was used as the primary measure of performance in our study.…”

“…One method to estimate these subcortical responses is the temporal response function (TRF), a linear encoding model of time-locked neural responses to continuous stimuli [7]. TRFs have been widely used for estimating cortical responses to speech [8][9][10][11][12], but fewer studies have investigated subcortical TRFs [5,[13][14][15][16].…”

The neural response at the fundamental frequency of speech is modulated by word-level acoustic and linguistic information

“…algorithm for computing the fundamental waveform was based on empirical mode decomposition (Huang and Pan, 2006;Forte et al, 2017). However, Etard et al (2019) showed that direct band-pass filtering of the speech signal is considerably simpler, faster to compute and leads to the same result (Kulasingham et al, 2020;Bachmann et al, 2021;Van Canneyt et al, 2021a,b). Here, we also employed a band-pass filter to extract the fundamental waveform from the voice recordings.…”

The Neural Response at the Fundamental Frequency of Speech is Modulated by Word-level Acoustic and Linguistic Information

“…The original algorithm for computing the fundamental waveform was based on empirical mode decomposition (Huang and Pan (2006); Forte et al (2017)). However, Etard et al (2019) showed that direct band-pass filtering of the speech signal is considerably simpler, faster to compute and leads to the same result (Kulasingham et al (2020); Van Canneyt et al (2021b,a); Bachmann et al (2021)). Here, we also employed a band-pass filter to extract the fundamental waveform from the voice recordings.…”