The linguistic genius of babies | Patricia Kuhl

Adults become 'culture-bound' listeners because stable phonetic representations formed early in development reduce ongoing updating of speech statistics, so adults stop tuning to new phonetic patterns in the input.

Between about six to eight months infants rapidly specialize to native-language sounds because they absorb the statistical distributions of the speech they hear, which sharpens discrimination for frequent contrasts and reduces sensitivity to absent ones.

Bilinguals must maintain and switch between two phonetic/statistical models because learning tracks language-specific distributions, so they hold separate categories and flip them depending on who they're speaking to.

Exposure to a new language during the critical period can produce native-like improvement in discriminating its sounds because repeated, targeted input supplies the statistical patterns infants use to form phonetic categories.

Infants can initially discriminate sounds from all languages because their auditory system starts unspecialized and broadly samples phonetic contrasts, letting them detect distinctions adults no longer notice.

Children's ability to acquire a second language declines after early childhood because there is a sensitive developmental window when language-related neural and representational systems are highly plastic, and those systems stabilize afterward, reducing learning speed and capacity.

Infants learn phonetic statistics only from live social interaction because social presence triggers attention and engagement that gates their sampling and encoding of speech, while passive audio or video fails to evoke the same learning state.

Magnetoencephalography (MEG) can reveal infant language processing in real time because arrays of SQUID sensors detect millisecond magnetic fields from neural activity, showing auditory cortex activation followed by coordinated engagement of surrounding areas involved in cross-region coherence.