Slide 1Slide 2Slide 3Slide 4Slide 5Slide 6Slide 7Slide 8Slide 9Slide 10Slide 11Slide 12Slide 13Slide 14Slide 15Slide 16Slide 17Slide 18Slide 19Slide 20Slide 21Slide 22Slide 23Slide 24Slide 25Slide 26Slide 27Slide 28Problems in motor controlRosenbaum Ch 2Principals of motor organizationSquire et al Ch 28, Kandel et al Ch 33The degrees of freedom (dof) problem: multiple ways to perform an actionThe body has a large number of dof, or ways to move.Arm has 7 dof (3 for shoulder, 2 each for elbow and wrist)Point in space is specified by 6 dofHowever, humans typically don’t take advantage of all dof, eg move to touch the nose is typically straight(ish) and stereotyped.The degrees of freedom problem: some solutionsNot a problem for the body because actions specified by goal state rather than the trajectory, so having multiple ways of performing an action is an advantage.Greater degrees of freedom allows flexibility.Synergies: a computational solutionExamples – L and R arm linkage, head and hand linkageshoulder and wrist linkageLinkage reduces dof4 dof 3 dofEye-head synergyEye-hand synergyhandeye headeyeheadThe degrees of freedom problemMechanical constraints: http://www.andrew.cmu.edu/user/shc17/Robot/Robot_walkingVideos.htmhttp://userweb.cs.utexas.edu/~jcooper/walking/Other optimization constraints:Minimize energy, optimize posture at end of movement, minimum jerk (derivative of acceleration)Grasp may optimize position at end of movement or use spring-like propertiesof muscles to assist in completion of the action.Sequencing and TimingSpeech errors (spoonerisms: this lecture is dearly none) show that word segments are programmed partly inparallel. Co-articulation – speech (tulip), typing – finger movements initiated prior to completion of others fingers etc..Coordination of eye, head, hand body.Evidence suggests that motor system has knowledge of how longa movement will take.The Perceptuo-Motor Integration ProblemHow are actions guided by sensory information?Pre-programmed, ballistic phase – eg reach in the dark to a remembered target.Slower feedback phase – guidance at the end of the reach as the hand nears a visible target.Schematic Representation of Feedback and Feed-forward SystemsEg: pursuit, reaching, graspingEg: saccade, throwingEye velocity=image velocitysensoryretinal velocityMotor commanddelayLoad/fatigue/current positionwindLearnt motor commandballisticguidedConsequences of loss of feedback on reachingLarge fibre sensory neuropathy leadsto loss of proprioceptive feedbackfrom musclesVision compensatesfor lack of proprioceptionErrors in direction,distanceNo vision or proprioceptionNormal: proprioceptiononlyThe Learning ProblemCalibration: what is the necessary command to move the eye to a certain location?Need to experience the sensory (visual and proprioceptive) consequences ofone’s own movements in order to calibrate. Even eye movements are sensitive to the need to maintain calibration. Sensory feedback is needed to maintain calibration in addition to assisting incompletion of the action, as in reaching.What will the visual image be as a consequence of that eye movement?Developmental learning: walking, talkingBallistic component of an action must be learnt over trials/repetitionsAmount of practiceSpecificity of practiceShooting a basketball from the free throw line versus other distancesNeural PlasticityIf finger is amputated, neighboringcortical regions take over that partof the projection from the fingers.(Merznick)Motor rehab after stroke impliessignificant potential for corticalre-organization.Note: phantom limbs, painSomatotopic organization of primary motor cortexPrinciples of organization of motor systemCentral Pattern Generations: neural networks for control of rhythmic movementsCentral Pattern GeneratorsPathways for signals to musclesHierarchical Organization of the motor systemSupplementary motor ctxPre-motor cortexPosterior parietal ctxPrimary motor ctxProprioceptive signals (muscles, joints) Spinal feedback: 30-50 msec80 msecTarget selectionMuscle commandsCortico-spinal tractInitiation of movementSmoothness, timingLearning new skillsMonitor feedbackPlanning/ sequences“Efferent copy”??M1Selection of trajectoryActivity prior to movementeyesSomatosensory ctxVisual consequences of movementMotor ProgramsMotor Equivalence; evidence for motor programsRight handWrist immobilizedLeft handTeethToeMovements are programmed with respect to the trajectory in space, not the jointanglesKinematic data for hand paths c, d, and e shown in part B. All paths are roughly straight and all hand speed profiles have the same shape and scale in proportion to the distance covered. In contrast, the profiles for the elbow and shoulder angles for the three hand paths differ. The straight hand paths and common profiles for speed suggest that planning is done with reference to the hand because these parameters can be linearly scaled. Planning with reference to joints would require computing nonlinear combinations of joint angles.The average acceleration and velocity profiles are scaled linearly as a function of the extent of movement to the target. The single peaks indicate that the extent of movement is specified prior to actual movement as a scaled impulse of force accelerating the limb.Variation in reaction time with age and number of alternatives.Variation in reaction time with number of alternatives and reduction with practice.Supplementary Motor
View Full Document
Unlocking...