In a multitalker background, adults' auditory cortex mainly follows the attended speech stream rather than the global auditory scene. Indeed, converging evidences show that ongoing oscillations at auditory cortices synchronize with speaker’s voice temporal envelope rather than with the whole acoustic input. This coupling occurs mostly in the delta and theta band, and declines with increasing background noise.
Since speech-in-noise and attentional abilities are typically lower during childhood, this cortical tracking mechanism might be somewhat different.
In this talk, I will present neurophysiological evidences arguing for a different speech-in-noise cortical processing in children and adults. Recent magnetoencephalographic data showed that children exhibited significant coupling only in the delta band, with a noise-sensitive coupling that was significantly stronger for the whole acoustic input than for the attended speech stream. These results will be compared with those obtained from patients suffering from central auditory processing disorders.
This shed light on the neural bases of patients and children's difficulties understanding speech in noisy conditions and argues for a progressive development of speech-in-noise abilities in humans.