Brainscape's Knowledge GenomeTM

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


See full index

Neuroscience > Elements of Neural computing > Flashcards

Elements of Neural computing Flashcards

1
Q

Which two properties allow individual neurons to encode information?

A

Rate coding and modality

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

what is rate coding

A

Property of neurons where firing frequency signals stimulus intensity and how it changes with time in afferent neurons, and motor neuron firing rate encodes the timing and force of contraction of a discrete population of muscle fibers

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

what is sparse coding and what is its advantage?

A

The accurate encoding of a given feature specified by a few neurons. It is energy efficient

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

What is population coding? What is thought to compensate for?

A

Features specified by activity in an ensemble of cells. Thought to compensate for the fact that generally neurons are very noisy.

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

What is temporal coding

A

Refers to a variety of situations in which neurons fire at very precise times which allows neural systems to time events much more precisely than is possible with rate coding. Exact spike timing or high frequency firing fluctuations carry timing information.

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

How does a slowly adapting receptor respond to a stimulus, and what is the consequence for the sensory neuron?

A

Responds to a protracted stimulus for as long as that stimulus lasts causing its sensory neuron to fire repetitively with a frequency that relates to the magnitude of the stimulus. These neurons exhibit static/ tonic responses to a constant stimulus

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

How do rapidly adapting receptors respond to a constant stimulus?

A

Respond only briefly to a constant stimulus because they soon become insensitive or adapt to it. They respond best to changes on intensity. Their afferents show dynamic (phasic) responses

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

How is a stimulus duration encoded?

A

The beginning and end of a stimulus will be signaled by changes in the rate of firing of slowly adapting afferents, and by transient bursts of firing from rapidly adapting afferents

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

What is the relationship between stimulus intensity and response for static sensory neurons?

A

It can be linear (skin thermoreceptor afferents). Commonly the firing rate rises with the logarithm of the intensity (skin mechanoreceptors and all photoreceptors)

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

What is the advantage and disadvantage of firing frequency increasing with the logarithm of the intensity of the stimulus?

A

Wide range of stimulus intensities can be accommodated within the dynamic range of neurons (maximum firing frequency of the neuron). Disadvantage: for high intensities the ability to discriminate intensity differences is reduced

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

Action potentials are binary digital signals (all-or none) why is this less prone to error than analog signalling?

A

It is less prone to corruption of the signal by noise because only two states need to be discriminated

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

How is error protection facilitated with rate coding?

A

The spurious absence or inclusion of occasional action potentials will not change the mean frequency of a train of action potentials much, unless the train is short

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

What is graceful degradation?

A

Describes when systems may have a sizeable number of rogue cells but does not fail catastrophically- as firing errors by a few neurons are swamped by proper firing of the majority- all that happens is that the information conveyed will be less precise.

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

What is an interspike interval

A

The time between two successive APs

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

What is the disadvantage of rate coding?

A

A minimum of two APs is necessary. However, for accuracy sufficient time must elapse to sample a reasonable amount of APs, therefore for short integration times, accuracy of stimulus intensity is sacrificed. Precise timing requires temporal coding in ensembles of neurons

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

How is the disadvantage of rate coding partly solved?

A

By population coding- briefly samples simultaneous output of many neurons carrying the same information.

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

What is the qualitative nature of a sensation termed?

A

Modality

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

What are the two hypotheses that account for how stimulus quality (modality) is encoded?

A

The sparse coding (labeled line) hypothesis and the population coding (ensemble, across-fiber) hypothesis. Most sensory systems lie somewhere along a spectrum between these two end-members.

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

What is the sparse coding (labeled line) hypothesis?

A

That a single class of sensory receptor and its afferent are necessary and sufficient to account for each type of sensation. The correspondence between receptor class and the nature of the sensation occurs because a sensory receptor responds only to specific type of stimulus.

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

What is population coding hypothesis?

A

that firing of several types of afferent is required to produce a given sensation. It is required to account for compound sensations., which must involve simultaneous activation of several receptor types by a single stimulus–>rich variety of higher order sensory experience

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

Neurons that fire over a wide range are said to be…

A

Broadly tuned

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

What is stimulus quality determined by?

A

the sense organ

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

How is the spatial location of a stimulus on a sensory surface given?

A

Which particular subset of neurons respond

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

What is a receptive field?

A

The receptive field of neuron is the region of a sensory surface which when stimulated causes a change in the firing rate of a neuron. Primary efferents have small RFs, size is governed by the distribution of the cluster of sensory receptors which supply the afferent. RFs of neighbouring neurons responding to the same stimulus tend to overlap

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

What are the receptive fields of proximal neurons?

A

More proximal neurons in a sensory pathway have receptive fields that are composites of the RFs of more distal neurons. They have larger RFs because of convergence (several afferents may synapse on a single more proximal neuron.

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

When is high and low convergence required for RF of proximal neurons?

A

Low convergence seen where high spatial resolution (the ability to sense stimuli that are close together as independent) is important. High convergence is necessary when it is required to integrate weak signals from a number of receptors to achieve a high sensitivity

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

Why is it that the more proximal the neuron the more complex its RF?

A

Proximal neurons get input from a wider range of sources than distal neurons and due to lateral (surround) inhibition

28
Q

what is lateral inhibition?

A

RF of a neuron has two zones , an excitatory central area (stimulation causes an increase in firing) and a surround (stimulation reduces firing, brought about by inhibitory interactions), from which opposite and antagonistic effects are produced. Cell’s behaving in this way are called on-center cells.

29
Q

What is the purpose of lateral inhibition?

A

Sharpens spatial resolution and enhances contrast at boundaries between stimuli. In skin mechanoreceptors this improves two-point discrmination, light-dark contrast at edges is enhanced in the retina and tone discrimination sharpened by central auditory neurons

30
Q

What is topographic mapping

A

In most sensory pathways primary afferents are wired to specific subsets of more central neurons in a strictly ordered fashion so that nearest neighbourhood relations are conserved. This means that information about stimulus location is not lost in more proximal parts of the pathway. RFs are aligned to produce an ordered map (neural representations of a sensory surface or feature of a sensation) across brain structures.

31
Q

What are the three types of topographic maps recognised (determined by extent of connections between neurons)

A

-Discrete maps: anatomically accurate and complete representations of a sensory surface.
-Patchy maps: consist of several domains within each of which the body is accurately represented, but adjacent domains map regions that are not anatomically close / disorientated
-Diffuse maps: no underlying topgraphy

32
Q

Why do discrete maps arise in the brain?

A

Because neurons are connected mostly to their neighbours, allowing local interactions between cells.

33
Q

What type of maps are cerebellar motor maps?

A

Patchy maps- they exhibit fractured somatotopy

34
Q

Why do patchy maps arise in the brain

A

because while some groups of neurons are locally connected, others are wired to distant neurons allowing global interactions to take place.

35
Q

What is divergence?

A

Few cells connecting with many downstream. It serves to disseminate information to a wide variety of targets

36
Q

2 examples of divergence?

A

-Primary afferents which relay with many interneurons so that other inputs and motor output in cord and brainstem can be modified
-Small numbers of preganglionic autonomic neurons supply up to 100 fold greater numbers pof post ganglionic neurons

37
Q

2 examples of convergence?

A

-Retina- 100 million photoreceptors but only one million output neurons
-Spinal cord: motor neurons outnumbered by primary afferents about 10-fold

38
Q

What is convergence?

A

The funnelling of connections from many cells to a few cells. It is the means by which target cells are able to integrate information from several sources.

39
Q

What are feedforward circuits?

A

Input neurons establish connection (either excitatory or inhibitory) with cells that are closer to the output (higher order) neurons than themselves

40
Q

What happens in feedforward inhibition?

A

Lower order cells excite inhibitory interneurons which project forward to dampen the activity of neighboring higher order cells.

41
Q

What is the overall effects of feedforward inhibition

A

This results in only the strongest signals being propagated further. In the form of GABAergic interneurons it is responsible for generating the surround inhibition in sensory pathways, and may contribute to selective attention

42
Q

What is selective attention

A

The facility to attend to one stimulus in preference to others.

43
Q

What is reciprocal inhibition?

A

A special case of feedforward inhibition where a response is enhanced by attenuating an opposing action. this operates in spinal cord reflexes that time the activities of limb flexors and extensors

44
Q

What are feedback circuits?

A

Higher order cells establish connections to lower order cells. The connections may be excitatory but are more usually made via inhibitory interneurons to cause feedback inhibition

45
Q

Why is feedback inhibition important?

A

Allows motor systems to correct errors during the execution of a movement.

46
Q

A neuron may feedback on itself by making recurrent connections, what are two important examples of this?

A

Recurrent excitation by axon collaterals in the hippocampus, recurrent inhibition of motor neurons in the spinal cord by Renshaw cells

47
Q

What are central pattern Generators?

A

Neural networks that produce cyclical patterns of activity autonomously

48
Q

What do central pattern generators mediate, for example?

A

The inspiratory-expiratory cycle of ventilation.
Limb movements during locomotion that involves alternate activation of flexors and extensors

49
Q

The basic operation of central pattern generators are modified or overidden by….

A

extraneuous pathways

50
Q

What is temporal coding used for?

A

Exact timing of events and as a way in which the firing of many neurons can be synchronised.

51
Q

Precise timing of events uses correlation detection, what is this?

A

Correlated inputs, those occurring within a short time of each other, increase the probability that a neuron will fire . Uncorrelated inputs will reduce spike probability.

52
Q

For correlation detection, how close the inputs must be is dependant on what?

A

The extent of temporal summation of the inputs and this depends on the time constant of the neuron. eg. neurons with fine dendrites are being bombarded with numerous ipsps, which consequently have short time constraints, will correlate inputs only if they are very close together.

53
Q

2 practical uses of correlation detection?

A

Pyramidal cells in the cortex act as coincidence detectors and modulate their behaviour to inputs arriving just 5 ms apart.
Coincidence detection underlies the way in which the brain stem allows tiny differences in arrival time of a sound in the two ears to be used by the auditory system to help localise the sound source

54
Q

In which regions of the brain do neurons exhibit regular oscillations in membrane potential

A

Cerebral cortex and thalamus

55
Q

What can oscillations in membrane potential cause?

A

They can be large enough to cause rhythmic fluctuations in firing rate which drive changes in synaptic activity.

56
Q

How do brain oscillations arise?

A

They arise from intrinsic pacemaker properties of individual neurons or by feedback through GABAergic interneurons known as Hub neurons (because of wide-spread connectivity)

57
Q

How are neural oscillations generated for sleep-wake cycles?

A

Those associated with wakefulness (Alpha) are established by reciprocal connections between the thalamus and the cortex. With the onset of sleep, changes is thalamic neurons result in altered connectivity with cortical neurons which consequently switch into oscillations of a different frequency.

58
Q

Where a theta brain rhythms seen in rat brains?

A

Predominantly in the hippocampus, where they are important in spatial learning. Hippocampal neurons require cholinergic excitatory input from the medial septum. Even though theta pacemaker lies within hippocampal neurons

59
Q

What is the critical function of neural oscillations?

A

to modulate the output of a neuron depending on the relative timing or synchrony between presynaptic and post synaptic neuron. A neuron is more likely to fire if an input coincides with the depolarizing phase of its oscillation than the hyperpolarizing phase. This mechanism is important in synchronizing neuron output and is thought to be involved in perception, attention and memory.

60
Q

How does the brain produce a unified percept from all disparate bits of information it has about the object (Binding problem)?

A

One possible solution is that all the segregated bits of information pertaining to a single stimulus are bound by synchronous firing of the neurons involved. This might be achieved by brain oscillations, perhaps of thalamic neurons which between them are reciprocally connected to all cortical areas.

61
Q

What is Hebb’s rule?

A

All synapses between two synapses become stronger if both of the neurons are activated at the same time. Synapses which show this type of plasticity are said to be Hebbian

62
Q

How do Hebbian neurons mediate associative learning?

A

They act as coincidence detectors that associate firing of presynpatic and postsynaptic cell.

63
Q

What is spike-timing-dependant plasticity?

A

Alterations to synaptic weighting that are fashioned by precise timing of neural activity.

64
Q

what are long lasting varieties of spike-timing-dependant plasticity?

A

Long-term potentiation (increase in synaptic weighting) and long-term depression (decrease in synaptic weighting)

65
Q

Where does LTP and LTD occur?

A

Both occur in the hippocampus. LTP is also seen in the neocortex, amygdala and other sites in the nervous system. While LTD also occurs in the cerebellum and spinal cord.

66
Q

What are the critical conditions required for LTPs and LTDs and what supplies them?

A

Require that activity in the pre and post synaptic cell occurs within a narrow time window of 10-20 ms. If the presynaptic spike precedes the post-synaptic spike then the result is LTP, if the presynaptic spike follows post synaptic spike then LTD ensues. Synchrony needed to trigger STDPs are supplied by gamma and theta frequency oscillations and the phase relations between them.

Neuroscience (15 decks)

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