Suprasegmental speech cues are automatically processed by the human brain: a mismatch negativity study.

This study investigates the electrical brain activity correlates of the automatic detection of suprasegmental and local speech cues by using a passive oddball paradigm, in which the standard Hungarian word 'banán' ('banana' in English) was contrasted with two deviants: a voiceless phoneme deviant ('panán'), and a stress deviant, where the stress was on the second syllable, instead of the obligatory first one. As a result, we obtained the mismatch negativity component (MMN) of event-related brain potentials in each condition. The stress deviant elicited two MMNs: one as a response to the lack of stress as compared to the standard stimulus, and another to the additional stress. Our results support that the MMN is as valuable in investigating processing characteristics of suprasegmental features as in that of phonemic features. MMN data may provide further insight into pre-attentive processes contributing to spoken word recognition.