Neuromechanics & Motor Control - Lecture 14

Voluntary control

Movement that is under conscious control of the brain

Rhythmic control

Movement with timing and spatial organization controlled autonomously

Reflexive control

Stereotyped responses to stimuli that may cause a threat

Acceleration phase

The initial phase of a goal-directed voluntary movement where speed increases

Movement phase

The middle phase of a voluntary movement where the motion is executed

Homing-in phase

The final phase of a voluntary movement involving deceleration as the target is approached

Inverse model

A model that transforms desired segment position changes into required motor commands considering body and environmental dynamics

Forward model

A model that predicts the outcome of a movement by transforming motor commands into a simulated result (efferent copy)

Feedback control

Correction of motor commands based on sensory information, limited by neural time delays

Feedforward control

Generation of motor commands based on learned task dynamics using internal models

Sensory discrepancy

The difference between actual and predicted sensory feedback, used to update internal models

Optimal control

A strategy for controlling a system that optimizes an objective or cost function

Possible criteria for optimal control

Time (as fast as possible), Error (minimal endpoint error), Jerk (smoothness of hand trajectory), Changes in Torque (smoothness of torque/motor commands), Effort (minimal energy), Robustness (minimal effect of (motor) noise/variability)

Optimal Feedback Control Law

Based on the feedback of the optimal state estimator.

Motor noise

Variability in muscle force that increases proportionally with motor command activation

Movement variability

Increases with motor command and effort due to motor noise

Cost function

A function that combines goal reward and motor cost to evaluate performance

State estimator

Combines forward model predictions and sensory feedback to estimate the current state

Kalman gain (K)

A component in sensory integration used to optimally combine prediction and sensory input (discussed in next lecture)

Minimum intervention principle

Corrections are made only for deviations relevant to the task, irrelevant variations are tolerated

When does Optimal feedback control occur

When there is feedback based on the state estimation (minimize the effect of sensory and muscle noise, feedback control in the presence of large time delays), Minimization of cost function results in distribution of activity across muscles (solves redundancy)

Motor learning

Acquisition of a new internal model to handle unknown environments or skills

Motor adaptation

Refinement of an existing internal model to accommodate changes in body or environment

Catch trial

A trial where a new environment is unexpectedly removed to assess adaptation

After effects

Residual movement patterns that reflect prior adaptation after returning to a normal environment

Visuomotor adaptation

Adjusting hand movements to match a visually altered cursor path, usually learned quickly

Force field adaptation

Learning to counteract external forces to maintain straight reaching paths

Types of feedback driving motor adaptation

Early phase: feedback driven, late phase: feedforward driven

Unpredictable impedance disturbances

Random force disturbances during movement that require feedback and feedforward adaptation to increase stiffness

Endpoint stiffness

The stiffness at the hand, adapted by the CNS to compensate for unstable force fields