Coördination Flashcards
Computational approach
Sensation –> Information processing –> Perception –> Decision –> Motor programming –> Action
Bayesian Information processing
Uncertainty in sensory signals. Important to reduce uncertainty. this can be done by likelihood and prior information. Combine those 2 and you get the final perception (posterior)
Bayesian state estimation
Likelihood: Probability of observation ‘o’ if the value is ‘x’
Prior: Probability that the value is ‘x’ based on prior experience
Posterior: combination of both.
Forward model
CNS calculated via the already known motor commands what the effect is to predict the outcome
Computation approaches or information processing approaches
Stimuli –> Information processing –> Perception.
Indirect perception. Processes needed to interpret stimuli
Ecological psychology
No processing/computation is required. Because information is already present in the structure of the light that surrounds you. The optic array. –> Direct perception
Open loop control
Computational approach. No sensory feedback infolved.
Desired movement –> CNS –> Controlled system (muscles) –> Actual movement.
Forward dynamic model
Predicting what the movement will be.
Forward sensory model:
Predictig what the sensory feedback will be when i perform this movement
Internal models in CNS
Representations of motor apparatus (stiffness damping, segment length) & context (know the size of the ball and racket during tennis)
Essential role in sensorimotor loop
Inverse model
You want to achieve something in the future –> the CNS is already calculating what the motor commands are that you need to achieve it
Adequacy
Depends on:
- Quality of inverse model
- quality of current state estimation
- influence of external disturbances
Error correction (in Forward models)
Feeback:
- Reponse to external pertubations
- Involves time delay
- Risk of oscillations because the CNS bases the corrections on a context a few seconds ago
Feedforward:
- Very fast.
- Cannot correct voor pertubations: Because you need sensory information to do this.
Motor Learning
Changes in internal models. Inverse models (in cerebellar cortex) Improves based on feedback, during learning. The model is adapting during the movements you make.
TTC
How long does it take before point of observation and object will collide. distance divided by speed