The beat, i.e. perceiving temporal periodicities in music, can be considered a cornerstone of musical rhythm perception and production. Even when music is not periodic, humans mentally organize the constituent acoustic rhythms according to internal periodic pulse-like beats. These periodic beats are used to coordinate body movement in time with the music, for example when we spontaneously bob the head or clap the hands in time with music.
Crucially, beat perception is considered a high-level perceptual process rather than a direct response entirely determined by the physical properties of the rhythmic sensory input. For example, the perceived beats are not necessarily marked explicitly in music. Many musical scenarios across cultures use highly complex rhythmic patterns in which the perceived beats are not physically cued by prominent acoustic onsets. Yet, people can spontaneously feel the beat with these complex rhythms and move to it. Moreover, a single musical rhythm can lead to different perceived sets of beats and, in turn, various rhythmic patterns can give rise to similar perceived beats. These examples illustrate the remarkable flexibility of the association between rhythmic input and internally perceived beat.
A promising approach to capturing the brain processes underlying beat perception in music is electroencephalography (EEG), which offers the advantage of measuring brain activity at the millisecond scale. In particular, the combination of EEG with frequency-tagging has proven successful at objectively relating the rhythmic input with brain processing of the rhythmic input.
Based on this methodological development, key outcomes of the research project so far included replication and extension of evidence that beat perception is related to selective activity of neural populations at beat frequencies. This neural activity cannot be trivially explained by bottom-up responses to rhythmic stimuli, as it is also observed when the beat frequencies are not physically prominent or even present at all in the acoustic stimulus, thus controlling for acoustic confounds. Moreover, this prominent neural activity at the beat frequencies cannot be fully explained by the processing of peripheral and subcortical relays of the ascending auditory pathway, as demonstrated using biologically-plausible models. This suggests that this neural activity might take place at the cortical level. Finally, neural activity elicited by an auditory rhythm is modulated by prior auditory rhythmic context. This activity persists longer when the rhythmic input gradually changes from regular to complex, as compared to rhythmic sequences gradually changing from complex to regular. Importantly, this effect of context also reflects into behavioral data of tapping the beat with the sequences, thus corroborating the view that this selective neural activity at the beat frequencies is functionally relevant. Together, these findings thus move us forward in our understanding of the neural underpinnings of musical rhythm: they indicate that beat perception is related to selective neural activity at the beat frequencies, and that this selective neural activity relies on high-level cortical processes that can be shaped at short-term level, especially when the rhythmic input is highly complex.
