Presentation

Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks

The role of prosodic and statistical cues for speech segmentation during the first year of life

Poster E98 in Poster Session E, Thursday, October 26, 10:15 am - 12:00 pm CEST, Espace Vieux-Port

clement francois^1,2, Christelle Zielinski^1,2, Laura Bosch^3,4,5, Antoni Rodriguez-Fornells^3,6,7; ¹CNRS, Laboratoire Parole et Langage, ²ILCB, Aix-Marseille University, ³Department of Cognition, Development and Educational Psychology, University of Barcelona, ⁴Institut de Recerca Sant Joan de Déu, Barcelona, ⁵Institute of Neurosciences (UBNeuro), University of Barcelona, ⁶Cognition and Brain Plasticity Group [Bellvitge Biomedical Research Institute-] IDIBELL, ⁷Catalan Institution for Research and Advanced Studies, ICREA, Barcelona

One of the earliest problems infants face in language learning is breaking the continuous speech input into words, which is a necessary step in language acquisition. Statistical and prosodic cues are crucial information in the speech signal that contributes to solving this speech segmentation problem. Understanding how prosodic processing may constrain and interact with statistical learning during early infancy for the development of speech segmentation remains unclear. Importantly, previous electrophysiological studies in adults suggest that brain oscillations are sensitive to different hierarchical complexity levels of the input, making them a plausible neural substrate for speech parsing. Specifically, successful speech segmentation might be associated with a peak at both syllable and word frequencies during exposure to statistically structured streams. In this poster, we show the results of a longitudinal study in which electrophysiological data were collected in the same infants at birth, at 6, and at 10 months during statistical learning of flat contour and melodically enriched speech streams. Frequency-tagging analyses were performed on the EEG data and show (i) enhanced brain responses to melodically enriched compared to flat speech streams across the three ages and (ii) increased word frequency power through development. These results illustrate the developmental brain dynamics associated with the gradual weighting of prosodic and statistical cues for speech segmentation during the first year of life.

Topic Areas: Language Development/Acquisition, Speech Perception