Ick Flashcards

1
Q

What does the NS do

A
Senses 
Processes 
Reacts 
Memorises 
Adapts !
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

How much info can be transmitted by a neuron.

A

Not a lot because they are noisy and action potentials can fire at the wrong time.

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

Why are MRIs safe and where do they look at.

A

Does not use ionising radiation and X rays and CAT scans do.

All body tissues.

Can’t have when pregnant

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

Brief of how MRI works

A

Powerful magnets elicit a signal from the protons contained in the water molecules of the body.

Magnets have no affect on health.

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

What is acceleration and velocity

A

Acceleration is the rate of change in velocity.

An accelerating charged particle emits EM waves.

Velocity comprises magnitude and direction.

So a change in magnitude or direction means acceleration is happening.

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

An example of how particles accelerate.

A

A particle spinning on its axis is accelerating because the direction is changing.

And a change in direction of the axis is also acceleration.

Both these happen at once and causes release of EM waves.

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

Protons in the brain tissue and MRIs

A

The poles of the protons are all random and in different directions when there is no magnetic field.

By applying a powerful static external magnetic field all the protons will point in the same direction. And an EM field is made. The brain becomes slightly magnetic.

If you apply a smaller horizontal magnetic field the protons tip over and rotate horizontally in sync. The protons will precess and wobble on their axis and this makes a detectable magnetic field.

MRI measures the difference in the horizontal and vertical components of the wobble and makes an image.

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

Typical MRI sequence

A

Subject is placed in a strong vertical magnetic field and all their protons line up.
There is a super cooled and super conducting coil producing the static magnetic field which can be up to nine Tesla.

Then a horizontal radio frequency pulse is applied from the head coil which tips the protons on their sides. And EM waves are released.

The horizontal pulse is fired on and off very quickly. When it is off the the protons dephase and less EM waves are made. When it is on the protons are spinning in phase and lots of EM waves are made.

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

T2 for MRI

A

Dephasing is a time constant known as T2.

T2 weighting refers to when they type of image made shows dephasing.

There is a loss of horizontal magnetisation.

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

T1 MRI

A

If you keep the horizontal pulse off, the protons realign themselves with the vertical field and restoring the vertical magnetisation.

This is much slower than dephasing and is measured indirectly.

The time constant of recovery of vertical magnetisation is called T1 and it can make a T1 weighted image.

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

T1 and 2 graphs showing signal intensity.

A

T2-
As the protons dephase they reduce signal intensity.

T1-
As the protons realign with the vertical the signal intensity increases.

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

Producing an image from T1 and 2

A

If the protons are in a very dense tissue like bone they are going to dephase quicker than in less dense tissue such as CSF.

A computer assigns a brightness value to various T1 and 2 values and this makes an image.

T1- the black spaces are air with the least protons and the white spaces are dense structures like bone.

T2- black is bone. This is best for viewing CSF as it will be white.

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

Blood flow to active brain areas.

A

Blood flow increases to more active brain areas.

It takes a long time for vessel diameter to change.

When it stops being active the diameter returns to normal.

Top vessel layer is the arachnoid layer.

James tilt table.

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

FMRI basics

A

Gives you anatomical data about the structures that are there.

Deoxygenated blood is paramagnetic and oxy is not. This means deoxygenated blood has a magnetic field we can detect using MRI.

BOLD signal- blood oxygenation level dependant signal.

It indicates which area of the brain is active because more deoxygenated blood is being removed.

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

fMRI image making.

A

A computer can differentiate the differences in deoxy blood and produce an image.

The colours of the image are changeable. One could be that yellow means high flow and it will fade out to red as flow decreases.

SPM program calculates the probability that a small chunk of brain has a different BOLD signal than the one next to it. It compares the voxel (3D pixel of brain chunk) with the voxel next to it using a t test.

The probabilities are assigned a colour to make an image.

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

MRI with fMRI

A

MRI scan first to make a monochrome brain image.

Then fMRI and overlay the coloured image on top.

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

BOLD signal problems.

A

Some think it has nothing to do with nutrition of the brain since the increased blood flow far exceeds the requirements of the active neurons.

Some think the blood is a cooling system.

18
Q

fMRI evaluation

A

Patterns of activation can be different for different subjects performing the same task.
Left handed people are less lateralised and use both hemispheres more often. Comparing scans should only be done with people who use the same hand.

Comparisons are done using non linear registration. Morph all the scans of people together to make an average brain image. It will be fuzzy and less localised.

The image has low temporal but high spatial resolution. Good at locating active areas but bad at telling you when it happens as vessels take a long time to dilate.

19
Q

What to do instead of MRI if you are investigating rapid cognitive tasks.

A

You need to couple fMRI with electroencephalography which has high temporal resolution but low spatial resolution.

20
Q

How to study a new neuron

A

Describe morphology, shape and size of the dendritic trees and this links to the function.

Map connections of the neuron.

Describe the activity and how they are stimulated.

21
Q

what is the function of the LGN and what is its structure and layout
where does it receive info from

A

we dont know the function
there are six layers in the LGN and there is one LGN in each hemisphere.

each of the six layers receives info from only one eye, layer one gets info from the contralateral eye (opposite side of the body) and layer two gets info from the ipsilateral eye (same side). this alternates.

the first four layers get info from parvocellular ganglion cells and the last two layers recieve info from the magnocellular ganglion cells.

22
Q

what type of connection do the ganglion cells make with the LGN neurons

where is most of the input to the LGN from

A

a 1:1 connection

60% of the input is back propagation from the cortex which regulates the activity of the LGN.
there are a lot of local interneurons in the LGN.

23
Q

why cant we find the function of the LGN

A

it is too deep in the brain to use imaging methods and this makes electrophysiology difficult too.

24
Q

where does the visual info go after the LGN and how is this split

A

to the primary visual cortex at the back of the brain

it is split into the ventral stream for object recognition and the dorsal stream for localizing stimuli.

25
Q

what is the pathway of the dorsal and ventral streams and do they interact

A

ventral-
P ganglion cell, LGN parvo, V1, V2, V4, inferior temporal cortex

dorsal-
M ganglion cell, LGN magno, V1, V2, V3 and then it can either go to V4 or it can go to MT and the parietal lobe.

yes they do interact

26
Q

what is the orientation invariance property and scale invariance

A

this means we can recognise objects no matter what way up they are.
we can recognise an object no matter what size it is.

we can recognise something even though it may look very different to what we learned the first time

27
Q

what is the hierarchical model of recognition

A

first in the V1 cortex, cells respond to the detection of edges in various orientations.
then they respond to combinations of edges and contours
next they detect parts of the object for example a face.
then they detect the point of view eg the front.
then we are able to recognise the object in any position.
the neruons will then respond to what category the object is in (animals or vehicles)

28
Q

what happens when you make lesions in the inferior temporal cortex

why cant we use mathematical models to study the process

A

it decreases our ability to recognise objects

it is just a model so it doesnt mean that it will really work.

29
Q

what happens as you go down the ventral stream

A

the neuron responses become more complex and the receptive field size increases

30
Q

what are three key features of cortical structure

A

it has layering and each layer does something different and they have different amounts of cell bodies

they have columns which are parts of the brain that have very similar functions

different neurons send their projections into different layers of the cortex.

31
Q

how do the layers of the cortex communicate and how do they send info to the next area

A

inter neurons known as resident cells send information between the layers of the cortex.
whichever layer the impulse comes from when it leaves the cortex depends on where it is being sent to next, the output is specific for individual layers.

32
Q

describe the columnar organisation of the cortex

A

ocular dominance column- different parts of the brain within the same cortical area will receive input from only one eye.

orientation direction column- neighboring parts of the brain have neurons with different properties so respond to things in different orientations.

blobs- parts of the brain have completely different properties and are important for processing colour information.

33
Q

what would you see if you injected radioactive proline into one eye

A

it would enter the ganglion cells and diffuse along the axons and go to the LGN.
it will then go to the cortex, you will see stripes of fluorescence in the cortex meaning that these stripes receive the info from the injected eye only and the non fluorescent stripes must get info from the other eye that was not injected.

34
Q

what will happen if you inject radioactive glucose into the cortex and then stimulate only one eye with light

A

this again will show the stripes of radioactivity in the cortex because only one eye is being stimulated so and so only the downstream neurons from that eye will use the glucose.
this proves that there are ocular dominance columns in the cortex (also in the LGN)

35
Q

where are the orientation columns found and what are they

what is the hypercolumn

A

different parts of the brain respond to motion in different orientations, this happens for all possible directions.
this is found in both the cortex and the LGN.

the hypercolumn shows the three types of column together. PICTURE.

36
Q

what happened in the Hubel and Weisel experiment

what type of cells did they find

A

they looked at the cat V1 cortex.
the cat was looking at different images on the screen and they did recordings from the neurons and from this they could map the receptive field.
when they showed the stimulus in another orientation the specific neuron they were recording did not respond and when they stimulated the entire receptive field the neuron did not respond.

they had found simple cells, which respond to the edges of an object in a certain orientation, the closer the object gets to a particular orientation the more the neuron will fire.

37
Q

what has to happen for a simple cell to respond and where are they found

what is their receptive field like

A

the stimulus has to be in the centre of the receptive field and they are found in layers four and six of cortev VI.

they respond to the orientation of a bar so their field has a line in the centre. PICTURE

38
Q

what do complex cells respond to and where are they found

where do they collect their info from

A

they respond to a bar in a particular orientation but it can respond when the bar is anywhere in the receptive field and not just in the middle.
they are located in layers 2,3,5 of the cortex.

complex cells collect info from many simple cells with similar orientation.

39
Q

hypercomplex cells

A

these will only respond when the bar is in the field, but if the bit outside the field is bent then they will respond.

PICTURE

40
Q

what are the types of neurons like in the inferior temporal cortex and what do they respond to

and what is the most complex like

A

they are able to respond to more complex shapes, there are face sensitive neurons that respond to faces.
as you remove characteristic features of the face the neurons start to respond less and less.

the most complex neurons respond to a specific object, they are called grandmother/Jenifer anniston neurons.
this is because they will respond to jenifer aniston and nothing else.
these neurons can also fire if you wrote the name on a piece of paper.

41
Q

evaluation of the hierarchical model for stimulus recognition

A

they are poor in explaining the scale and orientation invarience.
they dont take into account the feedbacks from higher cortical areas, top down regulation.