Visuo-manual tracking: does intermittent control with aperiodic sampling explain linear power and non-linear remnant...

Area of Science:

• Neuroscience
• Human-Computer Interaction
• Control Systems

Background:

• Human control systems are traditionally modeled as linear sensory-to-motor output.
• A non-linear remnant in motor output is often attributed to sensorimotor noise.
• Recent findings suggest refractoriness plays a significant role in manual tracking.

Purpose of the Study:

• To compare the explanatory power of noise versus non-linear transformations for manual tracking remnants.
• To determine if non-linear transformations can represent serial motor decision-making in tracking.
• To investigate the mechanisms underlying the non-linear aspects of human manual tracking.

Main Methods:

• Twelve participants manually controlled first and second-order systems under periodic disturbance.
• Joystick power was analyzed using three models: continuous-linear-control (CC), continuous-linear-control with noise (CCN), and intermittent control with aperiodic sampling (IC).
• Intermittent control (IC) used prediction error thresholds to trigger aperiodic sampling.

Main Results:

• The intermittent control (IC) model explained significantly more total power (77-87%) compared to linear models (8-48%).
• IC thresholds and open-loop intervals aligned with experimental conditions and prior measurements.
• The continuous-linear-control with noise (CCN) model required unrealistic noise spectrum changes.

Conclusions:

• Manual tracking is best explained by open-loop predictive control with aperiodic sampling.
• Aperiodic sampling is linked to serial decision-making in frontal, striatal, and parietal networks.
• These brain networks are likely integral to visuo-manual tracking processes.