This dissertation investigates the predictive and dynamic mechanisms underlying rhythm processing and groove. It accomplishes this by measuring the pleasurable urge to move to music (groove) and the beat perception abilities necessary for producing it. The first two studies do this by manipulating rhythmic complexity while recording changes in pupil size to assess cognitive effort, which is theorized to be necessary for reducing prediction error surrounding the beat. The second study additionally recorded the variability and intensity of participants’ foot taps as a measure of how precisely and confidently they could synchronize to the musical beat, respectively, to examine their relationship to groove. The third study explored how beat perception ability (which impacted the results of the first two studies) is shaped by experience and expertise. Namely, it examined whether inactive musicians retain the heightened beat perception abilities of their regularly practicing counterparts. In this work, I demonstrate that: 1) the pupil drift rate indexes groove ratings, 2) pickups elicit greater pupil dilations while 3) syncopations elicit greater groove ratings, 4) synchronizing to the beat increases groove ratings and pupil dilations, 5) tapping becomes softer and less precise with increasing complexity, 6) the inverted U-shaped curve between rhythmic complexity and groove is stronger for real music than synthetic drumbeats, and 7) greater beat perception is associated with more musical training, more precise tapping, and more prominent inverted U-shaped curves. In summary, this dissertation contributes a number of novel findings, extensions, and replications to the rich literature on beat perception and groove.