A Dynamic Systems Approach to Modeling Human-Machine Rhythm Interaction

Area of Science:

• Neuroscience
• Computational modeling
• Human-computer interaction

Background:

• Rhythm perception and meter anticipation are fundamental human behaviors, present from infancy.
• Existing time series prediction models often lack biological realism, failing to capture the imprecision of human internal clocks.
• Neuroscientific evidence highlights the need for biologically plausible models of rhythm perception.

Purpose of the Study:

• To develop a biologically realistic model of meter anticipation.
• To simulate human-like rhythmic behavior in dynamic systems.
• To improve synchronization in human-machine and interhuman interactions.

Main Methods:

• Proposed a neuron oscillator-based dynamic system.
• Incorporated two tunable parameters for local and global adjustments.
• Conducted experiments in human-machine interaction scenarios.

Main Results:

• The model successfully emulates human-like rhythmic behavior and reactions.
• Demonstrated human-like responses in common human-machine interaction scenarios.
• Replicated real-world rhythmic behavior in both human-machine and interhuman interactions.

Conclusions:

• The proposed neuron oscillator model offers a biologically plausible approach to meter anticipation.
• The model advances the development of natural and synchronized human-machine rhythm interaction.
• This work bridges computational modeling with neuroscientific insights for understanding rhythm perception.