Music training enhances the automatic neural processing of foreign speech sounds

International audienceGrowing evidence shows that music and language experience affect the neural processing of speech sounds throughout the auditory system. Recent work mainly focused on the benefits induced by musical practice on the processing of native language or tonal foreign language, which rely on pitch processing. The aim of the present study was to take this research a step further by investigating the effect of music training on processing English sounds by foreign listeners. We recorded subcortical electrophysiological responses to an English syllable in three groups of participants: native speakers, non-native nonmusicians, and non-native musicians. Native speakers had enhanced neural processing of the formant frequencies of speech, compared to non-native nonmusicians, suggesting that automatic encoding of these relevant speech cues are sensitive to language experience. Most strikingly, in non-native musicians, neural responses to the formant frequencies did not differ from those of native speakers, suggesting that musical training may compensate for the lack of language experience by strengthening the neural encoding of important acoustic information. Language and music experience seem to induce a selective sensory gain along acoustic dimensions that are functionally-relevant-here, formant frequencies that are crucial for phoneme discrimination. Music and language are universals of human culture, and both require the perception, manipulation, and production of complex sound sequences. These sequences are hierarchically organized (syllables, words, sentences in speech and notes, beats and phrases in music) and their decoding requires an efficient representation of rapidly evolving sound cues, selection of relevant information, construction of temporary structures taking into account syntactic rules, and many other cognitive functions. It is thus not surprising that music and speech processing share common neural resources 1-4 , although some resources may be distinct 5. The acoustic and structural similarities as well as the shared neural networks between speech and music suggest that cognitive and perceptual abilities transfer from one domain to the other via the reorganization of common neural circuits 2. This hypothesis has been verified by showing that musical practice not only improves music sound processing 6-9 , but also enhances several levels of speech processing, including the perception of prosody 10 , consonant contrasts 11 , speech segmentation 12 and syntactic processing 13. Interestingly, these findings extend to the subcortical level, showing an enhancement of the neural representations of the pitch, timbre, and timing of speech sounds by musical practice 14. Subcortical responses to speech are more robust to noise in musicians than non-musicians, and this neural advantage correlates with better abilities to perceive speech in noisy background 15. Overall, these studies suggest that the perceptual advantages induced by intensive music training rely on an enhancement of the neural coding of sounds, in both cortical and subcortical structures and extending to speech sounds. Interestingly, musical experience has also been associated with better perception and production of sounds in foreign languages 16-18. At the cortical level, the slight pitch variations of both musical (i.e. harmonic sounds) and non-native speech syllables (i.e. Mandarin tones) evoke larger mismatch negativity (MMN) responses in non-native musicians as compared to non-native nonmusicians 17,19. At the subcortical level, Wong and colleagues (2007) have shown that American musicians have more faithful neural representation of the rapid variations of the pitch of Mandarin tone contours as compared to American non-musicians 20. Moreover, this advantage correlates with the amount of musical experience