Musical training is associated with various cognitive improvements and pervasive plasticity in human brains. Among its merits, musical training is thought to enhance the cognitive and neurobiological foundation of speech processing, particularly in challenging listening environments, such as in a noisy restaurant. However, the brain mechanisms supporting any such potential advantages related to speech perception are not well specified. 

A brain imaging study by Dr. DU Yi from the Institute of Psychology of the Chinese Academy of Sciences and her collaborator Dr. Zatorre Robert from the Montréal Neurological Institute and McGill University has revealed that musical training might improve speech perception in noisy environments via enhanced neural foundation in bottom-up auditory encoding, top-down speech motoric prediction, and cross-modal auditory-motor integration.  

This study entitled "Musical training sharpens and bonds ears and tongue to hear speech better" has just been published online in PNAS and highlighted in the front session of the journal called "In This Issue".  

The study analyzed brain activity measured by functional magnetic resonance imaging (fMRI) of 15 musicians and 15 non-musicians, all on average 21-22 years of age, as they identified speech syllables against changing levels of background noise. 

The musicians had started their musical education before age seven and accumulated at least 10 years of musical training. They had practiced consistently (≥3 times per week) over the past three years. The two groups showed no difference in their hearing level, education, auditory working memory or non-verbal IQ. 

While the two groups performed equally under a no-noise condition, the musicians outperformed non-musicians at all other signal-to-noise ratios (SNRs). This ability was associated with enhanced activation of the left inferior frontal and right auditory regions in the musicians’ brains. 

Additional analysis revealed that neural patterns related to phonemes, which compose syllables, were more distinct in the bilateral auditory regions and speech production-related frontal motor regions of the musicians’ brains, as compared with non-musicians (Figure). 

Moreover, musical training strengthened the functional connectivity of the auditory-motor network, which scaled with better behavioral performance. 

(A-I) Brain regions showing significant classification of phoneme-elicited neural responses at each SNR in musicians (left panels) and non-musicians (right panels). (J) Speech-relevant anatomical regions used in multivoxel pattern analysis. (Image by DU Yi)