Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Action/Perception mechanisms in early speech perception: an EEG study
Poster C102 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.
Irene Lorenzini1, Thierry Nazzi2, Yasmine Baqqali3, Laurianne Cabrera4; 1Integrative Neuroscience and Cognition Center (INCC), Université Paris Cité & CNRS
Speech perception does not only recruit the auditory system but, in the (adult) human brain, entails parallel analysis by the sensorimotor system (Skipper et al., 2017). Recent event-related potential (ERP) results, measured using electroencephalography (EEG), have shown analogous sensorimotor processing in young infants (Choi et al., 2021). Moreover, developmental improvements in the perception of specific speech sounds (e.g., p, t) have been detected following the onset of production of those same sounds (e.g. DePaolis et al., 2011). These results led to the hypothesis that babbling, the onset of speech production, is a fundamental step in the ontogenetic setting of action/perception loops for speech perception, and may lead to major changes in perception skills. However, such results have only been observed behaviorally, based on the analysis of attentional patterns (infants’ looking times). Thus (despite indirect neurofunctional evidence, Imada et al., 2006), we do not know whether the perceptual improvements that are linked to production onset represent the signs of an actual neurofunctional reorganization of speech perception. Can we identify neurofunctional signatures following the establishment of the action/perception loop in speech perception? In an ongoing ERP study, we aim to compare electrophysiological signatures of phoneme processing before and after speech sounds are practiced in babbling. Ten-month-old infants undertake a multi-feature oddball paradigm recording with 11 active electrodes (F7, F3, Fz, F4, F8, C3, Cz, C4, P3, Pz, P4), presenting deviant stimulus that either fall into participants’ production patterns (‘deviant 1’) or fall outside from participants’ production patterns (‘deviant 2’). These stimuli (pa, frequent stimulus; ta, type 1 deviant, and ka, type 2) have comparable acoustic properties and were selected based on a preliminary parental survey targeting speech sound production in this age range. As to this survey, the sound p was produced by 75% of total participants (n=15); the sound t by approximately 50% and the sound k by only one participant out of 15. Capitalizing on this pattern, we aim to recruit two groups of participants (n=25 each): group 1, made up by infants producing both the frequent stimulus p and one out of two deviants; group 2, made up by infants producing the frequent stimulus only. Group 1 allows assessing possible differences in the electrophysiological signature to produced (t) vs non-yet-produced (k) speech sounds; Group 2 represents a control group. Speech production skills are home-recorded from each participant using the LENA® system and analyzed based on human perceptual judgment. Preliminary results from 14 participants (Group 1=10, Group 2=4; MA=10m18d) were analyzed in a repeated measures ANOVA assessing the effect of Region (Central, Frontal); type of Deviant (1,2) and production Group (1,2). This preliminary analysis did not show any significant effect of type of Deviant or production Group, but stronger responses in the frontal region for all participants (p=0.0001, F=28.594, ηp2 0.704). Participant recruitment and testing are ongoing, in order to reach full group sizes. Moreover, importantly, perceptual analysis of home-recorded speech productions is being completed in parallel and is susceptible to change the participants’ distribution in the two groups.
Topic Areas: Language Development/Acquisition, Speech Perception