Feeling or listening to musical rhythm: behavioural and neural responses to tactile and acoustic rhythm

Lenoir, Cédric [UCL] et al.

The propensity of humans to synchronize body movements along with musical rhythm has been attributed to a close coupling of the auditory and motor systems. The ability of the auditory system to integrate information at different time scales which is critical for speech is thought to be key in audiomotor entrainment. How other sensory systems integrate temporal structures from sub-second to supra-second scales remains largely unknown. Moreover, discrepant results have been reported so far on the performance of sensorimotor synchronization to rhythm across sensory modalities (such as visual or tactile). In addition, the characterisation of the neural activity that underpins the perception of rhythm required to ensure an accurate movement coordination in different sensory modalities is lacking. Here, we investigated the processing of rhythm conveyed with acoustic or vibrotactile inputs, given the similar physical characteristics of these inputs and their often concomitant production in musical contexts. We recorded the electroencephalogram (EEG) in 45 healthy volunteers when exposed – without moving – to acoustic or vibrotactile rhythm. In a separate session, we measured their ability to overt synchronization to the same stimuli using a finger tapping task. The rhythmic input consisted of a repeated pattern known to elicit perception of periodic pulse-like beats when presented in the auditory modality. Moreover, previous studies have shown that neural responses to this rhythm is enhanced at frequencies corresponding to the perceived beats, although the beat frequencies are not prominent in the physical structure of the input. This enhanced neural representation of the beat frequencies is thought to be related to stability of internal representation of the beats perceived by the participants. At the behavioural level, the stability and precision of synchronization was significantly higher in the auditory modality, even though the perceived beat was not significantly different between modalities. Similarly, at the brain level, we observed a significant enhancement of frequencies related to the perceived beat in the auditory modality but not in the somatosensory modality. Importantly, the main difference between the two modalities was found close and within the supra-second range, namely with larger amplitude at the slowest frequencies elicited by the rhythm in the auditory condition compared to touch. These results suggest that slow fluctuations in brain responses to rhythm may be a property enabling the auditory system to integrate incoming signal at different time scales, which is critical for rhythm perception and synchronization. In contrast, in other sensory modalities such as touch, brain responses would closely track each individual event in the stimulus without generating slow fluctuations. This lower propensity for temporal integration could explain general lower stability in motor entrainment to tactile rhythm as compared to auditory rhythm.