Neural Entrainment in the Theta Band Predicts Groove Perception in Popular Music

When humans listen to popular music, they often feel ‘the groove’, which is a pleasurable urge to move along with music. Previous studies suggest that groove arises from the brain’s entrainment to rhythmical patterns of music in the delta-theta and beta bands, which enables the brain to predict temporal structures. However, this notion has only been tested using simplified or artificial acoustic stimuli and melodies, rather than naturalistic popular music. In this study, we combined electroencephalography (EEG) measurements with two measures of neural entrainment to test whether neural entrainment in the delta, theta, and beta bands predicts individuals’ groove ratings of musical and clapping stimuli, which served as a control. We presented nine pop music songs and twelve MIDI-based clapping rhythms, which participants rated for their groove perception on a 5-point scale. Our results show that music songs received very variable groove ratings while ratings of clapping stimuli showed less variability. Univariate analyses demonstrated that both music and clapping stimuli consistently entrained neural oscillations in the delta–theta bands, but not beta bands, as measured by intertrial phase coherence (ITPC) and stimulus-brain coherence (SBC). ITPC and SBC in the theta band weakly, but significantly correlated with groove ratings in fronto-central clusters. Decoding of groove ratings from multivariate spatiotemporal-spectral EEG patterns of entrainment with curve-fitting and machine-learning models predicted groove perception with high accuracy, up to 0.8 Pearson correlation, where combined patterns of inter-trial phase coherence appeared more informative than patterns of stimulus-brain coherence. Our results support the theory that the perception of groove is rooted in the brain’s ability to entrain to ongoing dynamic temporal structures of music in the theta band. They further show that multivariate decoding models can be leveraged to predict subjective groove online from neural entrainment with high accuracy and temporal resolution.