Self versus environment motion in postural control.

Area of Science:

• Neuroscience
• Biomechanics
• Sensory integration

Background:

• Human balance relies on integrating visual and non-visual motion cues.
• Visual cues for motion can be ambiguous, arising from self- or environmental movement.
• The nervous system must resolve this ambiguity using noisy sensory input to maintain postural stability.

Purpose of the Study:

• To develop a computational model of how the nervous system integrates sensory information for postural control.
• To formalize the process of resolving visual cue ambiguity using physical motion cues.
• To explain the nonlinear influence of visual motion on human postural behavior.

Main Methods:

• Developed a Bayesian model to simulate sensory cue integration for body movement estimation.
• Analytically derived the optimal strategy for weighting visual motion cues.
• Compared model predictions to experimental data on human postural responses.

Main Results:

• The model demonstrates that the nervous system estimates body movement by combining visual and physical cues.
• An optimal strategy involves weighting perceived visual velocities by a power law under specific conditions.
• This power-law weighting explains nonlinearities in human postural responses to visual motion.

Conclusions:

• The developed Bayesian model provides a framework for understanding sensory integration in postural control.
• The findings suggest a specific, optimal strategy the nervous system employs to resolve visual ambiguity.
• The model successfully accounts for observed human postural behavior, validating its approach.