Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Brain and Behaviour > TB10 > Flashcards

TB10 Flashcards

You may prefer our related Brainscape-certified flashcards:
The Brain flashcards
1
Q

When does a neurone fire?

A

When the conditions are met which are determined by inputs.

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

What are neurones specialised for?

A

Receiving and sending signals.

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

What is rate coding?

A

Information about the stimulus or response is carried in the rate at which the cell fires.

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

What does the individual neurone represent?

A

The thing it detects.

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

How are individual neurones investigated?

A

Extracellular single unit recording.

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

How small is the electrode inserted into the brain?

A

40um

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

What is the difference between a raster plot and a pen-stimulus spike histogram?

A

A raster plot is dots whereas the histogram is spikes representing activation.

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

Name the three types of neurones in tuning curves. Explain them.

A
  1. sensory neurones - orientation of grating in that particular part of the receptive field.
  2. cognitive neurones - allocentic head turning of the animal, posysubiculum - formation of the hippocampus
  3. motor neurone - when looking at a hand, only activated if its moving.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Name one neurone that is tuned to more than one dimension.

A

V1 - orientation and retinal location

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

What are the three limits of the detector analogy?

A
  1. does the detector require a unique neurone for every object, concept and action?
  2. Ambiguity when looking at an individual neurone
  3. Grandmother cells - individual cells specialised to one thing.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Where are the Jennifer Anniston cells?

A

Medial Temporal Lobe

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

What did Wagdo et al find when investigating the Jennifer Anniston cells?

A

That those cells responded to 0.54% of possible stimuli and so these cells may be activated for the things.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Why is ambiguity a problem?

A

When looking at one cell, it may activate the same amount for three different orientations.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Why are broad tunings beneficial?

A

Allow for generalisations and communicate useful information.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What can the neurone firing rate do?

A
  1. signal a degree of match between the current position and the preferred value
  2. be interpreted as the probability of the particular stimulus being present
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What two cells is the nervous system made up of?

A

Nerve cells and neuroglial cells.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

How many neurones are there in the brain?

A

100bn

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What is synthesisia?

A

Individuals see the letters and numbers as different colours.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What is coarse coding?

A

Each neurone represents a range of possible input values when making comparisons between neurones.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Why can similarity between neurones be good?

A

Similar patterns can allow for generalisation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What are population codes?

A

These resolve ambiguity present in the firing of individual neurones.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Explain population codes with the use of colour.

A

There are three types of photoreceptor and these are overlapping, allowing us to see a range of colour. This code also limits the colour we can distinguish between as if we don’t have the cone and cannot make the wavelength, we cant see the colour.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

How is purple made by cones?

A

Needs high wavelength and low wavelength but nothing in between.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What are metamers?

A

These are mixtures that produce the same activity but cant be distinguished from pure spectral colours.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

Why is decoding useful?

A

It can be useful in the way in which information is presented, to summarise understanding and make the predictions about perception and action.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

What is sparsity?

A

The proportion of neurones that fire in a particular window. Or, the proportion of stimuli to which a given neurone responds.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

What is dense sparsity?

A

All neurones respond to some extent

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

What the disadvantages of sparse cells?

A

Would not be good for generalisation as there is no overlapping and may be inefficient as so many cells are required.

29
Q

What are the advantages of sparse cells?

A

They can distinguish between specific examples of a category, little interference as there is no overlapping, efficient as they are barely activated as there is so many and can be beneficial in rapid learning.

30
Q

What are the disadvantages of dense cells?

A

They have broad tuning so specifics may be difficult, prone to interference.

31
Q

What are the advantages of dense cells?

A

They have broad tuning, emphasises generalisation, captures similarity and complex learning.

32
Q

Where are shape representations analysed?

A

V4.

33
Q

What did Pasupathy and Conor find when looking at shapes?

A

They found that different neurones in the brain corded for different curvatures of the shapes and so when studying activation, decoding can occur and the specific shape can be found all from the neurone activation.

34
Q

What did Georgopoulous find?

A

Got a monkey to sit behind a chair and complete a centre-out and an out-centre task and found that no matter what, whenever the lever was pulled down towards himself, a specific neurone in the motor cortex was activated.

35
Q

What is a population vector?

A

When studying multiple neurones, the sum of them is created for the specific direction. The average of all the activation is the precise location as to where that neurone specifically activates.

36
Q

Give three other examples of population codes.

A
  1. Face cells in IT
  2. Place cells in hippocampus
  3. Head direction cells in the postsubiculum
37
Q

What are the premotor areas important for?

A

Planning information relates to behavioural goes, which is translated into the intention to perform specific movements in the primary motor cortex.

38
Q

What is the trade off when looking at neural computation models?

A

The complexity of the model (bad) and its capacity to offer a mechanistic explanation (good)

39
Q

What is neural computation?

A

A specific pattern of activity in one of the populations of neurones which induces a pattern of activity in another.

40
Q

Explain how things are processed when shown an R.

A

It is seen in the retina and passed to V1 which looks at the orientation. This then passes that information into visual cortex which passes information to letter processing mechanisms so that the letter can be identified.

41
Q

Explain Selfridges Pandemonium model.

A

Feature demons which look at the characteristics and then pass information onto the cognitive demons which show levels of activation. R is highly activated, B is slightly activated and X isn’t activated at all. The the decision demon works out which cognitive demon is the most active.

42
Q

How does emotion link to the model?

A

Visual features – facial expressions – emotion.

There was a relationship between the face and the social characteristics linking to what the model perceived.

43
Q

How do neural networks project information?

A

The input is linked to only one output and there have to be enough layers of units.

44
Q

In the neural model of a neurone, what do the neurones mean?

A

Each neurone modelled to a unit, or node, has no physical structure except for incoming and out coming information. Each unit has a property called activation which is known as firing rate. Each connection has a weight which is the total strength of all inhibitory and excitatory connections.

45
Q

Provide a brief description of the typical model of a neurone (McCulloch and Pitts)?

A

Total firing rate = the function of the sum of the weight x the action potential of the previous neurone.

This is calculated for all of them and then added to find the net input for that node.

46
Q

What is supervised learning?

A

The network learns by having an input pattern and a desired output pattern. The weights are random. Weights are gradually and systematically changes to reduce discrepancy. After going through the data many times, the observed and desired output becomes closer together. This is repeated for many input patterns.

47
Q

How is discrepancy reduced in supervised learning?

A

The weights are gradually and systematically changed.

48
Q

What is deep learning?

A

It is a multilayer neural network involved in supervised learning.

49
Q

What is machine learning?

A

The primary objective is artificial intelligence.

50
Q

What is unsupervised learning?

A

Competitive learning is a type of unsupervised learning. Learns to identify the clusters in input payers and the output neurones compete to respond to each neurone pattern, known as lateral inhibition. Eventually each cluster is presented by a different neurone pattern whose inputs best match that output pattern. The winning inputs weights are then changes so that if the input is then seen again, it will win and respond to that pattern again. If some connections are strengthened then others have to be weakened. Without this, the same unit would win again and not become selective for different inputs.

51
Q

When does the arrow move closer to the cluster?

A

Every time that the input wins.

52
Q

What is lateral inhibition?

A

The output neurones compete to represent each input pattern.

53
Q

What is parallel distributed processing?

A

Information processing occurs in parallel across are distributed units.

54
Q

What is the difference between supervised and unsupervised learning?

A

Unsupervised learning don’t know what the correct answer is which is different to supervised learning.

55
Q

What do visual neurones respond to?

A

Retinal location of orientation, ocular dominance, direction of motion and colour.

56
Q

How do Hubel and Wiesel say that things are organised?

A

State they are organised in pinwheels.

57
Q

What are the two types of optical imaging?

A

Camera over the cortex and voltage stain dye.

58
Q

What happens when there is high blood flow in optical images?

A

Intrinsic optical imaging

59
Q

What was Ohkis method?

A

2 proton CA+ imaging

60
Q

What does retinotopy mean?

A

When the same cells are activated from the same electrode, they are programmed in the same part of the visual field.

61
Q

What is corticoretinotopy?

A

It is when there is more affect of the whole cortex, from a brain lesion, affecting those neurones

62
Q

Why is fMRI good?

A

It is non-invasive and looks at the whole brain

63
Q

How is sound mapped in the brain?

A

Tonotopy - central auditory cortex when low frequency and move towards the periphery when the frequency increases.

64
Q

How is the body mapped?

A

Somatotopy - the body parts are mapped adjacently and then there is a line going down the central sulcus which programmes whole body configuration

65
Q

Where are place cells located?

A

Ventral hippocampus

66
Q

Where do grid cells fire more?

A

Ventral hippocampus

67
Q

Explain the somatotopic cortisol magnification?

A

In the somatosensory cortex - the things that are more sensitive to sensory input have larger cortisol areas applied to that area and then for the somatomotor cortex - the parts of the body that have the most motor action, are mapped larger in the cortex.

68
Q

What are self organising maps?

A

Outputs compete for inputs via lateral inhibition. When they are activated, it causes the random weighting to be moved, systematically so that they are more likely to be activated in the future. The neighbouring neurones are also more highly activated as they have similar patterns.

69
Q

What happens to plasticity as age increases?

A

It decreases and becomes limited.

Brain and Behaviour (3 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions