Lecture 4: Motor Control

1
Lecture 4Motor Control

• Prof.dr. Jaap Murre
• University of Maastricht
• University of Amsterdam
• jaap_at_murre.com
• http//neuromod.org

2
Overview

• The anatomy of the motor system
• Population coding
• The role of spinal cord, cerebellum, and basal
  ganglia
• This is not in the book, but can be on exam anyway

3
Muscles are activated by alpha motor neurons
4
The stretch reflex reveals some elementary
processing in the spinal cord
5
Cortical anatomy of the motor system lateral view
6
Medial view
7
Schematic overview of the motor system
8
Basic questions regarding motor control can
nowadays be answered

• How are motor movements represented in the brain?
• How are they used in the production of movement?
• Which brain areas are involved?

9
How to be precise with noisy components
Area 5 neuron during repeated reaching movements
each individual trial gives a rather imprecise
signal
10
Population coding

• Population coding allows precise representations
  on the basis of (very) noisy or even damaged
  components
• Population coding is based on the statistics of
  averages
• They rely on coarse-coded neural representations

11
Coarse coding

• If a neurons representation responds to many
  inputs, this is called coarse coding
• The advantage is that more accurate
  representations can be formed by suitable
  combination of the coarse representations

12
Why coarse coding works

• If we move along a straight line, each time we
  cross a receptive field boundary one neurons
  changes its activation
• the representation changes.

13
In primates abundant evidence exists for coarse
coding
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
Georgopoulos shows that movement is coded in
population vectors
18
Population vectors give accurate movement
direction signals
19
Motor cortex sets up the signal, but execution is
dependent upon other areas
20
Original plans in motor cortex are sometimes
revised on the go
21
Activation of motor areas is a cascade rather
than a sequence
22
Simple movement activations motor cortex and
somatosensory cortex
23
More complicated sequences involve other areas
SMA supplementary motor area (part of area 6)
24
Imagined movements remain limited to the
supplementary motor area (SMA)
25
Internally and externally generated movements
PMC premotor cortex (also part of area 6)
26
Skilled (Old) versus new motor movements
27
Response competition
28
Elastic constraints in motor development

• The problem of grasping is overdetermined given
  an end-location, many possible joint positions
  solve the problem
• In order to make the problem soluble elastic
  constraints are necessary (cf. Mike Jordan)
• Muscles (as springs) are one source of such
  constraints

29
Spinal cord
30
Coarse maps of limb movements in the frog

• Spinal cord of frogs does significant motor
  processing
• Frog can still clean itself after severing of
  cord (dogs can also still scratch themselves)
• The data suggest that even at a spinal level
  coarse coding is used
• It is likely that similar types of coding are
  used in mammals

31
Cats with severed spinal cord could still walk on
a treadmill
32
Method followed by Emilio Bizzi
Based on the idea of muscles as springs by
Feldman
33
Limb movements in frog spinal cord are coded with
respect to their end-positions
34
The interactions of force fields can be described
by vector calculus
Fields A and B combined predict field ltABgt (see
C). When A and B are stimulated the resulting
field (see D) corresponds to the theoretical
field ltABgt
35
Cerebellum
36
Glickstein it is not completely clear what the
cerebellum does

• Bimanual control
• Motor learning?
• vestibuloocular reflex
• nictitating membrane response
• Coordination and integration of movements

37
Global anatomy of cerebellum
38
More detailed anatomy of cerebellum
39
Louis Bolk midline cerebellar vernis controls
bilaterally synchronized movements cerebellar
hemispheres control unilateral movements
40
(No Transcript)
41
David Marr (1969) cerebellum is excellent for
simple associative learning (conditioning)
42
Hebb-Marr networks

• Marrs views can be combined with those of Hebb
  to yield associative networks
• These networks can store input-output patterns
  (hetero-associative learning)
• The exhibit
• pattern completion or content-addressable memory
• fault tolerance

43
Willshaw networks

• Highly abstracted, early neural network from 1969
• Activations are 0 or 1
• A weight either has the value 0 or 1
• A weight is set to 1 if input and output are 1
• At retrieval the net input is divided by the
  total number of active nodes in the input pattern

44
Example of a simple heteroassociative memory of
the Willshaw type
1 0 0 1 1 0
0 0 1 0 1 1
1 1 0 1 0 0
0 0 0 1 1 1
1 1 1 1 1 1 1 1 1
1 0 1 0 1 0
0 0 1 0 1 1
1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1
45
Example of pattern retrieval
(1 0 0 1 1 0)
1 1 1 1 1 1 1 1 1
0 0 1 0 1 1
1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1
3 2 2 3 3 2
Sum 3
1 0 0 1 1 0
Div by 3
46
Example of successful pattern completion using a
subpattern
(1 0 0 1 1 0)
1 1 1 1 1 1 1 1 1
0 0 1 0 0 1
1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1
1
2 1 1 2 2 1
Sum 2
1 0 0 1 1 0
Div by 2
47
Example graceful degradation small lesions have
small effects
(1 0 0 1 1 0)
1 1 1 1 1 1 1 1 1
0 0 1 0 1 1
1 1 1 1 1
1 1 1 1 1 1 1 1
3 2 1 2 3 1
Sum 3
1 0 0 0 1 0
Div by 3
48
Correspondence views on brain structures
49
Computational views on cerebellum
50
Summary

• Like vision, motor behavior has a lot of special
  purpose circuitry
• We can understand many aspects of this circuitry
  in terms of why this representation makes sense
• For example, coarse grained coding has the
  advantage of precise control despite noisy
  components

51
Summary (continued)

• Motor behavior is not simply stringing together
  some basic movements
• Motor planning and execution are very much
  cognitive functions