The cerebellum and motor control Flashcards

1
Q

What are some useless facts about the cerebellum?

A

Contains 1/2 of all neurons but 10% of brain weight

Projects to almost all UMNs

Jerky, erratic and poorly coordinated movements + voluntary movement seems stereotyped with damage;

Intention tremor, dysarthria and ataxia = names for symptoms

You can be born without one and be okay-ish

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

How does the cerebellum influence motor control?

A

No direct projections to LMNs - modulates UMN activity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What are the main inputs to the cerebellum?

A

Cortical - motor cortex (a copy of a motor command), somatosensory + visual areas of parietal cortex

Spinal - proprioceptive information about limb position/movement (muscle spindles/golgi tendon organs)

Vestibular - rotational and accelerational head movement (semicircular canals/otoliths in inner ear)

From the inferior and middle cerebellar peduncles

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is the primary output of the cerebellum?

A

To the thalamus via purkinje cell axons, inhibitory

Made up of the sum of the parallel fibre inputs multiplied by the weight (strength of parallel-purkinje synaptic connection)

From the superior cerebellar peduncle

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is the function of the cerebellum?

A

Knows what the motor command is - integrates with feedback about actual body position - projects back to cortex to adjust further movements

Also motor learning (+basal ganglia and cortical circuits)

Also non-motor tasks - sensory prediction (image changes when you move your hands); active sensing (knowing what object is in your pocket without looking); emotioal and cogntiive processing )verbal memory)

Disorders - autism, dyslexia, schizophrenia? (verbal prediction failures = auditory hallucinations)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Describe the cytoarchitecture of the cerebellar cortex

A

Three layers

Mossy fibre - input (white matter under the cortex) from many cortical and spinal inputs,
synapse onto granule cells

Granular layer - granule cells outnumber mossy fibres 50:1, project axons to top molecular layer where they split into two parallel fibres - these synapse with Purkinje cells

Purkinje layer - middle, sole output, each cell synapses with 150,000 parallel fibres in the top molecular layer; also receive 1 climbing fibre input

Climbing fibres - axons of inferior olive, wraps itself around Purkinje cell dendritic tree and forms 1000+ synapses with same input signal

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Describe Purkinje cell output

A

Fire spontaneously = simple spikes at c.50 spikes/sec but with parallel fibre input can increase to >200/s

Complex spikes = unusual shape, low frequency, produced by climbing fibre input = reliable, when climbing fibre fires as does PC

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What did Brindley (1964) suggest about the function of the cerebellum?

A

Assumptions: cerebellar damage does not usually cause paralysis; other parts of the brain issue motor commands but badly

Possibly to learn motor skills so when a simple or incomplete message from the cortex arrives, its sufficient to produce a proper execution - frees up space in cortex too

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the core of the Marr-Albus theory of the cerebellum?

A

Error signals needed in order to process what you’ve done wrong =climbing fibre input

As climbing fibre input/error frequency is low frequency - doesnt really affect your movement = good

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is a synaptic weight and a decorrelation learning rule?

A

Synaptic weight - between parallel fibre and Purkinje cell - is a change according to the correlation between parallel-fibre signal and the error signal conveyed by he climbing fibre:

+ve correlation = reduce weight (as is contributing to the error); -ve correlation = increase weight

Learning stops when there is not longer a correlation between any parallel-fibre signal and climbing fibre signal = decorrelation learning rule

In this system correlation = causation

Version of the least mean squared rule (used in signal processing technology)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What evidence do we have for the decorrelation learning rule?

A

Stimulate parallel fibres at the top of the cortex - measure Purkinje cell output - then stimulate climbing fibre input

Purkinje cell activity found to reduce for a substantial time = long term depression (me when i realised this exam was not multiple choice)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is some evidence that decorrelation learning work?

A

Vestibulo-ocular reflex (VOR) - inactivate the flocculus in the cerebellum - failures in VOR

Semicircular canal output to brainstem vestibulo- occulomotor nuclei is modulated by floccular output which receives a copy of the eye movement command from the brainstem nuclei via mossy fibre inputs + error signals from retinal slip from climbing fibres

We have made robots that can stabilise camera images in a robot head using the equivalent maths

How well did you know this?
1
Not at all
2
3
4
5
Perfectly