A neurological motivated computational model for beat induction

Abstract

Current models for beat induction give an accurate description of the phenomenon. These models however have their emphasis on recreating behavioural aspects; lacking an explicit neural correlate. Finding this neural correlate is one of the current challenges in the field of music cognition. In this research proposal the Fitzhugh-Nagumo model is taken as a starting point for a model for beat induction. This model is able to synchronize phase at equal tempi of the internal oscillator and external input, however it lacks the potential to synchronize to other tempi. In two experiments we will measure whether the influence of perturbations in the external input signal is dependent of the metric position of that perturbation. The literature suggests that perturbations close to a beat have higher influence that those further away. Also a slight asymmetry is expected for “too early” and “too late” onsets. One experiment will focus on the influence on the conscious perception, the other on synchronization of finger movement. Using these experiments, and using existing literature, a subdivision is made between the unconscious synchronization of body movement, phase synchronization and the cerebellum on one hand and conscious perception of asynchrony, period (tempo) adjustment and the prefrontal cortex on the other hand. Relations within these groups suggest a model in which phase synchronization take place subconsciously in the cerebellum, under control of other brain areas to adjust the tempo of the internal timekeeper. The result of this project will be a model for beat induction that describes a neural process.