Related Experiment Videos
Pseudo-inverse control in biological systems: a learning mechanism for fixation stability
1Department of Psychology, University of Sheffield, Western Bank, Sheffield, UK
Summary
This study shows how the brain learns motor control, similar to robots. A learning rule helps eye muscles coordinate, mimicking pseudo-inverse control and explaining the size principle in motor units.
Area of Science:
- Robotics and Neuroscience
- Motor Control Systems
Background:
- Redundancy is a challenge in both biological and robotic motor control.
- Pseudo-inverse control is a robotic solution, potentially used by the oculomotor system for eye movements.
Purpose of the Study:
- Investigate learning mechanisms for pseudo-inverse control in biological systems.
- Model integrator calibration for horizontal eye movements to understand ocular motor unit coordination.
Main Methods:
- Utilized a gradient-descent learning rule to adjust ocular motoneuron (OMN) input weights.
- Employed a retinal-slip estimate as the error signal for learning.
- Introduced a noise term to OMN firing rates to explore motor unit behavior.
Main Results:
- The learning rule suppressed noisy motor units, linking firing-rate threshold to motor-unit strength.
- This process led to the emergence of the size principle, where stronger units have higher thresholds.
- The trained system demonstrated an approximation of pseudo-inverse control within the central oculomotor range (+/-35 degrees).
Conclusions:
- A distributed learning mechanism can implement pseudo-inverse control in biological motor systems.
- The size principle naturally arises from a learning rule that minimizes noise-related errors.
- This research connects robotic control principles with biological motor control mechanisms in the oculomotor system.