Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Neural mechanisms of sensorimotor integration in speech perception
There is a Poster PDF for this presentation, but you must be a current member or registered to attend SNL 2023 to view it. Please go to your Account Home page to register.
Poster C81 in Poster Session C, Wednesday, October 25, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
This poster is part of the Sandbox Series.
Olivia Bizimungu1, David Ostry1, Pascal Perrier2, Lucie Ménard3, Sylvain Baillet1; 1McGill University, 2Université Grenoble Alpes, 3Université du Québec à Montréal
Speech perception is an active and flexible process which can be influenced by the listening context and inputs beyond audition. The sensorimotor system is crucial for controlling speech articulators during production, but may also be involved in perception of auditory speech. Engaging the articulators can influence the perception of sound, and their respective brain areas are activated in response to speech stimuli; even in the absence of production requirements. Mounting evidence endorses an active role of the sensorimotor system in guiding speech perception; however, little is known about the neurophysiological mechanisms underlying this integration. In this study, we disrupt articulatory sensorimotor feedback by applying a simple mechanical perturbation to the lip shape configuration; and assess the effective influence on both perception and neural dynamics. So far, 12 French-speaking participants classified vowels along an acoustic continuum between phonemes /u/ to /œ/ both normally (baseline), and with a small plastic tube held between the lips (liptube). We measured their ongoing brain activity from 64-channel EEG and explore the temporal, spatial, and spectral dynamics of sensorimotor integration during active speech perception. Preliminary results demonstrate variable effects on perception, with no significant differences between conditions in either the perceived boundary between phonemes, nor in the slope of psychometric curves. Nonetheless, average measures of both boundary and slope were lower in the liptube condition compared to baseline. Neural responses to vowel stimuli differed between conditions: support vector machines were able to decode condition from single trial data with up to 88% accuracy, averaged across participants. This differentiability was visible at the source level: the N100 stimulus-evoked component showed pronounced differences within the left posterior temporal lobe. Brain regions also showed differences in oscillatory activity: the beta rhythm (13 – 29 Hz), which is associated with sensorimotor processing, was most modulated throughout temporal auditory regions, and ventral sensorimotor cortices. Altogether, these findings suggest differences in speech stimulus processing and ongoing neural dynamics due to altered somatosensory feedback between conditions, which could have a downstream influence on vowel categorization. Although final results are still pending, the current findings suggest an influence of sensorimotor perturbation on the perception and associated neural processing of speech stimuli; encouraging further work into elucidating these mechanisms. Behaviourally, we observe a disruption of phonemic categories due to the articulatory perturbation. This perturbation was also reflected in neural activity, with altered brain responses to speech stimuli in regions associated with auditory speech processing. Finally ventral sensorimotor regions, which house the mouth and lip area representations, showed changes in beta-activity, which is consistent with theories of predictive motor-to-auditory signalling occurring via mu rhythm (8 – 30 Hz) dynamics. These findings highlight the importance of sensorimotor brain systems in the active perception of speech; suggesting that articulatory representations activated during listening may guide phonemic processing.
Topic Areas: Signed Language and Gesture, Multilingualism