Presentation

Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Poster Slams

Speech Motor Adaptation During Synchronous and Metronome-timed Speech

Poster E13 in Poster Session E, Saturday, October 8, 3:15 - 5:00 pm EDT, Millennium Hall

Abigail R. Bradshaw^1,2, Daniel R. Lametti³, Douglas M. Shiller⁴, Kyle Jasmin⁵, Carolyn McGettigan¹; ¹University College London, ²University of Cambridge, ³Acadia University, ⁴l’Université de Montréal, ⁵Royal Holloway, University of London

Research with the altered auditory feedback paradigm has provided evidence for the critical role played by self-generated speech auditory feedback in speech motor control. Specifically, speakers exposed to a predictable sustained perturbation of real-time auditory feedback (e.g. a change in the first or second formant) gradually start to adapt to this perturbation; for example, by shifting their produced formant frequencies in an opposite direction to the perturbation. Less is known however about the impact of other voices on such speech motor adaptation, and whether adaptation is robust across different speaking contexts. In particular, despite speech typically being a social act, few previous studies have examined speech adaptation in contexts involving a social element. In this study, we tested the effect of synchronous speech (the act of speaking in synchrony with another voice) on the adaptation response. Using the sentence-level adaptation paradigm developed by Lametti, Smith, Watkins, and Shiller, (2018, Current Biology), in Experiment 1 we measured participant’s adaptation to a joint F1-F2 perturbation during production of sentences, either while speaking alone (solo reading group, n = 15) or while synchronising their speech with another (pre-recorded) voice (“the accompanist”; synchronous speech group, n = 15). We found that both groups exhibited a significant adaptation response, with no significant difference between the magnitude of the average group responses. There was however a significant difference in the level of between-participant variability in adaptation within the two groups, with more variable adaptation responses in the synchronous speech group. It was further found that participants in the synchronous speech group showed evidence of convergent changes in the F2 and F0 of their speech productions towards those of the accompanist voice prior to introduction of the feedback perturbation. An exploratory analysis suggested that individual variability in adaptation within the synchronous speech group may be partly explained by variability in the extent to which the formant changes required for convergence to the accompanist voice matched those required for adaptation; the greater the match, the greater the adaptation observed. However, Experiment 2 found a similar profile of adaptation responses for a metronome-timed speech condition (n = 15), in which no convergence is possible. This suggests that the act of synchronising speech with an external rhythm in the absence of another voice can also cause increased between-participant variability in adaptation, without affecting the group-level adaptation response. These findings demonstrate that speech motor adaptation can be affected by concurrent performance of a speech task involving exposure to another voice and/or the coordination of speech timing with external stimuli. This suggests that such adaptation is not always an automatic response that remains impervious to the effects of speaking context and style. Further work is planned that will aim to better isolate the potential contribution of vocal convergence mechanisms to the increased individual variability in adaptation observed during synchronous speech. This will involve altering the formant frequencies of the accompanist voice in order to explicitly manipulate the agreement between the direction of formant change required for convergence versus adaptation.

Topic Areas: Speech Motor Control, Language Production