Limited Pre-Speech Auditory Modulation in Individuals Who Stutter: Data and Hypotheses

Abstract

Purpose We review and interpret our recent series of studies investigating motor-to-auditory influences during speech movement planning in fluent speakers and speakers who stutter. In those studies, we recorded auditory evoked potentials in response to probe tones presented immediately prior to speaking or at the equivalent time in no-speaking control conditions. As a measure of pre-speech auditory modulation (PSAM), we calculated changes in auditory evoked potential amplitude in the speaking conditions relative to the no-speaking conditions. Whereas adults who do not stutter consistently showed PSAM, this phenomenon was greatly reduced or absent in adults who stutter. The same between-group difference was observed in conditions where participants expected to hear their prerecorded speech played back without actively producing it, suggesting that the speakers who stutter use inefficient forward modeling processes rather than inefficient motor command generation processes. Compared with fluent participants, adults who stutter showed both less PSAM and less auditory-motor adaptation when producing speech while exposed to formant-shifted auditory feedback. Across individual participants, however, PSAM and auditory-motor adaptation did not correlate in the typically fluent group, and they were negatively correlated in the stuttering group. Interestingly, speaking with a consistent 100-ms delay added to the auditory feedback signal-normalized PSAM in speakers who stutter, and there no longer was a between-group difference in this condition. Conclusions Combining our own data with human and animal neurophysiological evidence from other laboratories, we interpret the overall findings as suggesting that (a) speech movement planning modulates auditory processing in a manner that may optimize its tuning characteristics for monitoring feedback during speech production and, (b) in conditions with typical auditory feedback, adults who stutter do not appropriately modulate the auditory system prior to speech onset. Lack of modulation of speakers who stutter may lead to maladaptive feedback-driven movement corrections that manifest themselves as repetitive movements or postural fixations.

Figure 1.

Schematic of the experimental paradigm. The experiment included three conditions: (A) speaking, (B) silent reading, and (C) seeing. (D) An auditory stimulus (1 kHz, 50-ms duration, 75 dB SPL) was presented 400 ms after initial appearance of the stimulus word (in the speaking and silent reading conditions) or symbols (in the seeing condition) in one third of the trials of each condition, whereas no auditory stimulus was presented in the remaining trials. Adapted from Daliri and Max (2015b): Brain and Language, Vol. 143, A. Daliri and L. Max, “Modulation of Auditory Processing During Speech Movement Planning is Limited in Adults Who Stutter,” pp. 59–68, Copyright © 2015, with permission from Elsevier.

Figure 2.

Grand-averaged auditory evoked potentials (across participants and across electrodes in three regions of interest marked on the scalp maps) for (a) adults who do not stutter and (b) adults who stutter. The two groups differ in modulation of the N1 component (gray box) in the speaking condition relative to the silent reading and seeing control conditions. (c) Group-averaged amplitudes (with standard errors of the mean) of the N1 component in each condition illustrate the Group × Condition interaction. N1 amplitude of the nonstuttering group was smaller in the speaking condition relative to both control conditions; N1 amplitude of the stuttering group was similar across all three conditions. (d) A boxplot illustrates the consistency of N1 modulation (N1 modulation = |N1_Reading| − |N1_Speaking|; thus, positive values indicate an amplitude reduction in the speaking condition) among speakers who do not stutter as compared with the absence of this phenomenon among speakers who stutter. Adapted from Daliri and Max (2015b) Brain and Language, Vol. 143, A. Daliri and L. Max, “Modulation of Auditory Processing During Speech Movement Planning is Limited in Adults Who Stutter,” pp. 59–68, Copyright © 2015, with permission from Elsevier.

Figure 3.

Grand-averaged auditory evoked potentials (averaged over the colored electrodes in the scalp maps) for (a) adults who do not stutter and (b) adults who stutter. We found that N1 amplitude of the participants who do not stutter in both the speaking and listening conditions was smaller than that in the silent reading condition; however, (c) N1 amplitude of the participants who stutter was similar in all three conditions (error bars indicate standard errors of the mean). (d) The boxplot illustrates participant distribution in terms of N1 modulation in the speaking and listening conditions for the two groups (i.e., modulation calculated such that positive values indicate an amplitude reduction relative to the silent reading condition). Adapted from Daliri and Max (2015a), Brain and Language, Vol. 150, A. Daliri and L. Max, “Electrophysiological Evidence for a General Auditory Prediction Deficit in Adults Who Stutter,” pp. 33–44, Copyright © 2015, with permission from Elsevier.

Figure 4.

Nonstuttering speakers' grand-averaged auditory evoked potentials in response to (a) nonspeech stimuli (1-kHz tone) and (b) speech stimuli (/da/ syllable) that were presented either prior to vocalization in a speaking condition (pink) or at the same time point in a silent reading condition (blue). We examined changes in amplitude of the N1 and P2 components (highlighted by the gray-shaded areas in a and b) in the speaking condition versus the reading condition to determine auditory modulation for each type of stimulus. N1 modulation occurred to a similar extent in responses to speech and nonspeech auditory stimuli; (c) N1 amplitudes in the speaking and reading conditions; error bars indicate standard errors of the mean and (d) Boxplot of auditory modulation calculated as |N1_Reading| − |N1_Speaking|. P2 modulation occurred only in responses elicited by speech stimuli: (e) P2 amplitudes and (f) P2 modulation. All data averaged across the colored electrodes in three regions of interest. Adapted from Daliri and Max (2016), Creative Commons Attribution License (CC BY), Copyright © 2016 Daliri and Max.

Figure 5.

Relation between pre-speech modulation of the auditory evoked potential N1 component and speech adaptation to altered auditory feedback. Participants completed an auditory–motor adaptation paradigm that allowed us to quantify each participant's adaptation to formant-shifted auditory feedback. The same participants also completed our standard pre-speech auditory modulation (PSAM) paradigm that allowed us to quantify each participant's extent of N1 and P2 modulation prior to speech onset. N1 PSAM was not statistically significantly correlated with the amount of adaptation for speakers who do not stutter (r = .05, p = .87), and it was negatively correlated with the amount of adaptation for speakers who stutter (r = −.65, p = .02). Neither group showed a significant correlation between adaptation and P2 PSAM (p > .349). Adapted from Daliri and Max (2018): Cortex, Vol. 99, A. Daliri and L. Max, “Stuttering Adults’ Lack of Pre-Speech Auditory Modulation Normalizes When Speaking With Delayed Auditory Feedback,” pp. 55–68, Copyright © 2018, with permission from Elsevier.

Figure 6.

Grand-averaged auditory evoked potentials of (a) nonstuttering and (b) stuttering groups in response to probe tones presented during speech planning in a condition with nonaltered auditory feedback (NAF), during speech planning in a condition with delayed auditory feedback (DAF), and at the equivalent time points in a condition with only silent reading. Group-averaged N1 amplitudes in each of the conditions are shown in (c) (error bars correspond to standard errors of the mean). (d) Pre-speech modulation of the N1 component (|N1_Reading| − |N1_Speaking|) for the nonstuttering group decreased in the DAF condition relative to the NAF condition, whereas for the stuttering group, it increased for the same comparison. As a result, the between-groups difference in pre-speech modulation during speaking with NAF disappeared when speaking with DAF. Adapted from Daliri and Max (2018): Cortex, Vol. 99, A. Daliri and L. Max, “Stuttering Adults’ Lack of Pre-Speech Auditory Modulation Normalizes When Speaking With Delayed Auditory Feedback,” pp. 55–68, Copyright © 2018, with permission from Elsevier.