Ecologically valid methods of auditory feedback (Tomassi et al., 2021)

Purpose: Auditory feedback is thought to contribute to the online control of speech production. Yet, the standard method of estimating auditory feedback control (i.e., reflexive responses to auditory–motor perturbations), although sound, requires specialized instrumentation, meticulous calibration, unnatural tasks, and specific acoustic environments. The purpose of this study was to explore more ecologically valid features of speech production to determine their relationships with auditory feedback mechanisms.

Method: Two previously proposed measures of within-utterance variability (centering and baseline variability) were compared with reflexive response magnitudes in 30 adults with typical speech. These three measures were estimated for both the laryngeal and articulatory subsystems of speech.

Results: Regardless of the speech subsystem, neither centering nor baseline variability was shown to be related to reflexive response magnitudes. Likewise, no relationships were found between centering and baseline variability.

Conclusions: Despite previous suggestions that centering and baseline variability may be related to auditory feedback mechanisms, this study did not support these assertions. However, the detection of such relationships may have required a larger degree of variability in responses, relative to that found in those with typical speech. Future research on these relationships is warranted in populations with more heterogeneous responses, such as children or clinical populations.

Supplemental Material S1. Reflexive responses to fundamental frequency (f_o) perturbations for all participants. Responses are normalized to average of the baseline region and negated for ease of viewing such that negative responses are compensatory. The shaded regions describe the 95% confidence intervals.

Supplemental Material S2. Reflexive responses to first formant perturbations for all participants. Responses are normalized to average of the baseline region. The shaded regions describe the 95% confidence intervals.

Supplemental Material S3. Fundamental frequency (f_o) centering in all participants. Inward movement represents “centering” whereas outward movement describes trials with no centering. The absolute value of the dashed lines represents the average initial distance to the median across all trials and the absolute value of the solid lines represents the average middle distance to the median across all trials. Arrows represent peripheral trials, those for which initial variability from the median was the greatest.

Supplemental Material S4. Vowel formant centering using the first (F₁) and second (F₂) formants in all participants. Inward movement represents “centering” whereas outward movement describes trials with no centering. Dashed lines represent the average initial distance to the median across all trials and solid lines represent the average middle distance to the median across all trials. Arrows represent peripheral trials, those for which the initial variability from the median was the greatest.

Tomassi, N. E., Weerathunge, H. R., Cushman, M. G., Bohland, J. W., & Stepp, C. E. (2021). Assessing ecologically valid methods of auditory feedback measurement in individuals with typical speech. Journal of Speech, Language, and Hearing Research. Advance online publication. https://doi.org/10.1044/2021_JSLHR-21-00377