Using a sensorimotor synchronization task through space and time, we have recently shown that timing modes are used differently in function of the time constraints set upon motor planning. Predictive timing is used for slow execution whereas emergent timing seems to dominate for sequences performed at fast tempi. In the present study, we applied the circle-tapping task paradigm in patients with schizophrenia to test, which timing mode may cause the motor fluency deficits previously reported in schizophrenia. Eighteen patients and their controls were instructed to tap a sequence of 6 visual targets following the rhythm of a regular metronome. This synchronization task was performed at 10 distinct tempi (inter response interval–IRI of 1100 ms to 300 ms, for the slowest to the fastest rhythms). Results showed that the mean IRIerror was similar in patients and in controls indicating that all participants performed the task correctly and were able to adapt their rhythmic tapping to the imposed tempo. Patients presented nevertheless longer contact times suggesting a less fluent execution of actions than the controls. Finally, patients tapped systematically after the beep, results that were associated to the patients’ difficulty to correct for minimal timing errors during predictive motor planning. Overall, these findings confirm the distortion of predictive timing in schizophrenia. Preliminary data using rhythmic music during adapted physical activity will be presented to show how rhythm in the environment may be used to improve predictive timing for motor planning in pathological populations.