Authors: Irene Arrieta¹
¹Universite Paris Cite, INCC
Background: Language acquisition occurs primarily in face-to-face interactions, yet research often relies on auditory-only paradigms. This study investigated how 7-9 month-old infants process audio-visual speech under reduced intelligibility. The researchers hypothesized that degraded speech would increase visual attention to the speaker’s mouth, increase hemodynamic activity in the prefrontal cortex due to processing effort, and decrease neural-speech synchrony (cortical tracking).
Method: Forty-seven infants viewed videos of infant-directed speech with either intact or degraded (8-band vocoded) audio. Gaze behavior and neural activity were recorded simultaneously during a 12-minute session. Eye tracking measured attention to the mouth versus the eyes, functional near-infrared spectroscopy (fNIRS) indexed hemodynamic changes in fronto-temporal regions, and electroencephalography (EEG) quantified cortical tracking of speech.
Results: Behavioral findings indicated that infants significantly increased looking time toward the speaker’s mouth when speech was degraded. This suggests an early compensatory strategy similar to that seen in adults, where visual cues from the mouth are used to supplement degraded auditory information. Neural analysis of fNIRS and EEG data is currently ongoing to test the remaining hypotheses.
Conclusion: Infants demonstrate a clear behavioral shift toward the mouth in response to reduced speech intelligibility. This research aims to establish objective measures of speech processing in young listeners, which may provide critical insights for supporting infants with hearing loss in ecologically valid listening conditions.
