Cracking the pitch code of music-motor synchronization using data-driven methods

The study of auditory-motor synchronization with music has been so far mostly concerned with timing. For instance, research has established that people are able to spontaneously coordinate with musical beat when walking or running. Yet, music is more than a metronome, and the relation between the spectral dimension of music, i.e. parameters such as its pitch, timber, or harmonic structure, and simultaneous motion remains nearly unknown. Here, we introduce a novel data-driven paradigm in which participants were asked to walk on a treadmill while listening to a large variety of musical tones systematically varied in pitch. Using analysis techniques inspired by psychophysical reverse correlation, we show that participants' gait patterns while walking to music spontaneously encode pitch height: despite being instructed to simply synchronize in time, participants steps were both longer and heavier on tones with lower pitches. These findings reveal that, similarly to time perception, pitch cognition is not purely 'disembodied' and suggest that listeners' spontaneous motor reactions to pitch might ground their ability to mentally represent music.