4. fMRI: Functional Magnetic Resonance Imaging

Question 1

1. Functional Magnetic Resonance Imaging (fMRI)

List 3 key aspects of fMRI and state the process used by Positron emission tomography (PET)

Answer 1

1. Measures regional cerebral blood flow (rCBF)
2. Cognitive functions are region specific, if a task involves a certain cognitive function, the areas involved will become more active, need more oxygen and more blood.
3. Measures regional levels of blood oxygen by detecting magnetic changes in red blood cells when they become de-oxygenated
4. (PET: uses a radioactive tracer bound to glucose or oxygen and follows blood to active areas – use of ionizing radiation disadvantage)

Question 2

2.

Functional Magnetic Resonance Imaging:

Break down the meaning.

Answer 2

-Functional: it’s about brain function, rather than structure

-Magnetic: magnetic field required

-Resonance: interaction between magnetic field and radio frequency

Question 3

3.

Describe how MRI works

Answer 3

MRI: create images of soft tissue in the body, which x-rays pass through undistorted

• Tissue = water-based, different amount of water for different tissues
• Enables 3D image of layout of these tissues

Question 4

4.

fMRI makes use of some characteristics of the molecules in the brain. List the way it works starting with the water molecules - H20.

Answer 4

1. Protons spin, and have a magnetic charge
2. Use strong magnetic field to line up protons
3. Send a radio pulse through the lined up protons, and see how they resonate
4. Different protons (different tissues) resonate differently (magnetic susceptibility), allowing composition of a tissue image
5. fMRI: oxygenated blood resonates differently to de-oxygenated blood, allowing composition of an (indirect) image of brain activity

Question 5

5.

What is the difference between MRI and fMRI?

Answer 5

MRI gives a structural picture of the brain while fMRI gives the location of blood flow within the brain.

MRI is more static while fMRI is more dynamic.

Question 6

6.

There are two types of fMRI scans:

Describe structural.

Answer 6

Structural (anatomical) MRI:

►Static image of brain structure

►High spatial resolution (1x1x2 mm)

►Used to overlap functional images onto

►T1 contrast (measures a different magnetic property to functional scans)

Question 7

7.

There are two types of fMRI scans:

Describe functional.

Answer 7

Functional MRI:

►dynamic image of brain blood flow: hemodynamic response

►lower spatial resolution than structural MRI, but much higher spatial resolution than other imaging methods: 3x3x3 mm, although higher resolutions with stronger machines

►temporal resolution is inherently limited by the slow blood flow

►Both spatial and temporal resolution depend on strength of the magnet: Tesla

►T2 contrast (measures a different magnetic property to structural scans)

Question 8

8:

Brain consumes ……. of body’s oxygen uptake, it does not store oxygen and only a little glucose.

Needs to be supplied from …….. blood supply

Oxygen rich blood looses …………… to tissue

More ………. tissue uses more ………… than less active tissue

Answer 8

Brain consumes 20% of body’s oxygen uptake, it does not store oxygen and only little glucose

Needs to be supplied from local blood supply

Oxygen rich blood loses oxygen to tissue

More active tissue uses more oxygen than less active tissue

Question 9

9.

What is the physics of MRI - List 4 key steps

Answer 9

1 Strong magnetic field created

2 Protons in water molecules have weak magnetic fields, initially randomly oriented, but some align with the external field

3 Radio pulse knocks orientation by 90 degrees, which leads to a change in magnetic field

4 Protons ‘relax’, and procedure can be repeated for different slices of brain

Question 10

10.

How long does it take to take a whole brain MRI image?

Answer 10

• Whole brain image in appr. 2 seconds (3 mm slices)
• T1: relaxation time -> T1-images – structural scans
• T2: local interactions (mainly deoxyhaemoglobin)

-> T2* image - fMRI

Very noisy environment. Sound is knocking the protons off their orientation.

Question 11

11.

Draw an image of the MRI setup.

Answer 11

Question 12

12

How is the T1 calculated?

Answer 12

The T1 is related to the relaxation period of the protons - it can deduce the water content and the type of tissue it was released from.

Question 13

13

How is the T2 calculated?

Answer 13

The T2 is linked more directly to interactions with the deoxyhaemoglobin in the blood.

Question 14

14. fMRI

Blood Oxygenation Level Dependent (BOLD) contrast:

Answer 14

Simple view BOLD finds the ratio of difference between oxygenated and de-oxygenated

• Compares level of oxygenated with de-oxygenated blood magnetic properties: hemoglobin is diamagnetic (only magnetic when exposed to external magnetic field) when oxygenated and paramagnetic (normally magnetic) when de-oxygenated
• Hemoglobin molecules resonate differently in these different magnetic states
• Hemodynamic response function
• An indirect measure of brain activity

Question 15

15.

BOLD contrast – a problem:

Answer 15

• You want to measure the level of de-oxygenated blood, since that would be the most direct evidence of increased brain activity
• However, BOLD cannot measure the level of de-oxygenated blood per se, only its level compared to the level of oxygenated blood: more oxygenated blood means lower BOLD
• This would mean that greater brain activity is only detected by BOLD if all oxygenated blood is used up

Question 16

16.

What are the three ingredients needed for BOLD to be effective?

Answer 16

1. ►CMRO2: cerebral metabolic rate of oxygen
2. ►CBF: cerebral blood flow
3. ►CBV: cerebral blood volume

Question 17

17.

What is the Hemodynamic response function (HRF)?

Answer 17

• initial dip: as neurons consume oxygen -> small rise in amount of deoxyhaemoglobin -> reduction of BOLD signal
• overcompensation: in response to the increased consumption of oxygen, blood flow to the region increases. Increased blood flow is greater than increased consumption -> BOLD signal increased
• undershoot: blood flow and oxygen consumption dip before returning to original levels. This may reflect a relaxation of the venous system.
• % signal change: 0.5-3%
• time to peak: signal peaks 4-6 sec after stimulus begins

Question 18

18.

fMRI

►Brain is always …………………………

►In imaging, ‘active’ area refers to ……………….. that is greater relative to some other condition

►Need for ………………………, well-matched to experimental task

Answer 18

►Brain is always physiologically active

►In imaging, ‘active’ area refers to physiological response that is greater relative to some other condition

►Need for baseline response, well-matched to experimental task

Question 19

19.

Example: Petersen, Fox, Posner, Mintun and Raichle (1988) PET study using cognitive subtraction.

Answer 19

Study brain activity involved in word recognition, phonology and retrieval of word meaning – cognitive subtraction.

Need to think about what stages that can be differentiated.

Question 20

20.

Define - Cognitive subtraction.

Answer 20

• cognitive substraction is a method to measure which areas in our brain are active while doing a certain activity or seeing specific words.
• We know that by “substracting” the activity of the brain before hearing a stimuls and after hearing it results in an activity in the brain connected with this word.

Question 21

21

Example: Petersen, Fox, Posner, Mintun and Raichle (1988) List the 3 conditions of the study.

Answer 21

1. Passive viewing of written words vs baseline (passive viewing of x)
2. Read aloud a written word vs baseline (passive viewing of a written word)
3. Generate an action (eg. see cake sat “eat”) vs baseline (readaloud a written word)

All were compared to the baselines which left only information related to word regognition.

Question 22

22.

Name problems with cognitive subtraction.

Answer 22

►Assumption of “pure insertion” (or “pure deletion”): adding an extra component does not affect the operation of earlier ones in the sequence.

BUT: interactions are likely to occur

►Baseline task: should be as similar to the experimental task as possible

Question 23

23.

Another approach to making sense of fMRI is to look at cognitive conjunctions and factorial designs.

Answer 23

►Cognitive conjunction requires a set of tasks that has a particular component in common. Look for regions of activation that are shared across several different subtractions.

►Example: why can’t we tickle ourselves (Blakemore, Rees, and Frith,1998).

2 Factors: ‘touch’ (felt/not) and ‘self movement’ (moved/not)

Question 24

24. Cognitive Conjunctions and factorial designs

Example: why can’t we tickle ourselves (Blakemore, Rees, and Frith,1998).

Answer 24

2 Factors:

1. ‘touch’ (felt/not)
2. ‘self movement’ (moved/not)

Conjunctions:

A-B and C-D tactile sensation

A-C and B-D motor production

(A-B)-(C-D) efference copy and reduced ticklishness

-> Somatosensory cortex and cerebellum

Question 25

25. Parametric design

►Variable of interest is treated as a ……………… rather than a ……………………

………………. between brain activity rather than ………………..between …… or ………. conditions.

Answer 25

►Variable of interest is treated as a continuous dimension rather than a categorical distinction.

Associations between brain activity rather than differences between two or more conditions.

Question 26

26.

Give an example of parametric design.

Answer 26

• passive listening to spoken words at six different rates
• Different brain regions show different response profiles to different rates of word presentation. Adapted from Price et al. (1992), and Friston (1997).
• no baseline necessary

Question 27

27.

Functional integration:

Answer 27

►Functional integration: the way in which different regions communicate with each other.

• Model how activity in different regions is interdependent.
• ‘effective connectivity’ or ‘functional connectivity’ between regions when performing a task.
• Use techniques like principal component analysis
• Example: word production vs repeating letters in schizophrenics and controls

Question 28

28.

How can a study be evaluated? Should PET or fMRI be used?

Answer 28

One has to make assumptions about the most appropriate method.

Random stimuli should perhpas use Event-related designs and fMRI while A blocked design could use either PET or fMRI.

Question 29

29.

Blocked versus event-related designs

Answer 29

►Block design: stimuli in one condition grouped together

►Event-related design: different stimuli or conditions are interspersed with each other (efMRI)

• Intermingled conditions are subsequently separated out for the purpose of analysis.

Question 30

30.

Event related designs can be used when:

Answer 30

►If stimuli need to be presented randomly

►If stimuli are infrequent

►If conditions defined by participant (e.g. correct/incorrect trials; presence of an hallucination - see right)

Question 31

31.

Blocked Design - pros and cons

Answer 31

+ strong BOLD contrast: higher signal-to-noise ratio

simple design and analysis

• practice/fatigue effects cannot be used when participants should not know which condition is coming next -

Question 32

32

Event-related designs - pros and cons

Answer 32

+ no practice/fatigue effects

+ can be used when participants should not know which condition is coming next: randomisation

+ can be used when trials can only be classified after experiment

• weaker BOLD contrast: lower signal-to-noise ratio more complex design and analysis

Question 33

33.

Define Sessions:

Answer 33

►Session: a scanning session, all the data collected from a participant. Usually consists of structural scan, and a number of runs of functional scans

Question 34

34.

Define volumes:

Answer 34

►Volume (image): Set of slices taken in succession: a 3D spatial image, with a temporal dimension. Expressed in RT (Repetition Time): how long does it take to acquire a volume

Question 35

35.

Define Runs:

Answer 35

►Run: a continuous period of scanning, consists of a specified number of volumes

Question 36

36.

Define Epoch

Answer 36

►Epoch: a period when a certain condition is presented. Conditions (epochs) can be grouped together (blocked design) or randomly intermixed (event related design)

Question 37

37.

Analysis - what are the key considerations?

Answer 37

►Individual differences – averaging over many participants

►Correction for head movement

►Stereotactic normalisation

►Smoothing

►Statistical comparison

Question 38

38.

Correction for head movements:

Answer 38

►Spatial resolution -> small spatial distortions

►Individual differences in brain size and shape – stereotactic normalisation

►Individual head aligned differently in scanner over time due to movement

• Regions harder to detect
• False positive result

Physically restraining head and participant instructions

Correction

Question 39

39.

Stereotactic normalisation:

Answer 39

►Mapping regions on each individual brain onto a standard brain (brain template is squashed or stretched until it fits).

• Tailarach and Tournoux (1988). Brain Atlas (based on one brain), Tailarach coordinates
• X left/right
• Y front/back
• Z top/bottom
• Alternative: Montreal Neruological Instute (average of 305 brains)

►Voxels (volume elements), 3-D coordinates (x,y,z)

Question 40

40.

What is smoothing?

Answer 40

• Spreads some of the raw activion level of a given voxel to neighbouring voxels.
• See D4 which is ‘switched on’ And L8, which is ‘switched off’
• Enhances signal-to-noise ratio Compensates for individual differences in anatomy
• Assumption: cognition does not occur in single voxels
• Increases the spatial extent of active region -> more likely to find overlap between participants

Question 41

41.

Slices and Voxels

Answer 41

There are slices of the brain that are scanned in.

A voxel is a volumetric pixel (3x3x6mm)

Question 42

42.

Statisitical comparison - with tens of thousands of voxels to compare chance will play a part without careful comparison.

Answer 42

►Tens of thousands of voxels – capitalisation on chance

• Lower significance level (Bonferroni)
• Choosing statistical threshold based on spatial smoothness (random field theory)
• Analyse pre-determined region

Reported ‘corrected’ or ‘uncorrected’ statistical parameters Voxels looked at may be smaller group relating to the hypothesis.

Question 43

43,

List the 6 stages to analysis.

Answer 43

1. Data acquisition
2. Correct for head movement
3. Stereotactic normalisation
4. Smoothing
5. Divide data according to design (i.e. order of trials)
6. Perform statsitical comparison

Question 44

44.

Interpretation - what are 3 key comparisons to think carefully about?

Answer 44

1. Inhibition versus excitation
2. Activation versus deactivation
3. Necessity versus sufficiency

Question 45

45.

Inhibition versus excitation (not to be confused with activation/deactivation)

Answer 45

• Functional imaging signals are assumed to be related to metabolic activity of neurons, and synapses in particular
• However: activity can be excitatory or inhibitory!
• BOLD signal more sensitive to neuronal input into a region than the output from the region
• Unclear whether functional imaging can distinguish between two neural functions

Question 46

46.

Activation versus deactivation

Answer 46

►Activation/deactivation

►Merely refers to difference between two conditions

►Does not say anything about the direction of the difference

Question 47

47.

Necessity versus sufficiency

Answer 47

►Are active regions critical to the task?

►Sufficiency: functional imaging shows us active regions, but these may not be crucial

►Use methods in conjunction with other methods

Question 48

48. Give an example of a results report from a study:
49. Magnet Strength
50. Number of slices (volume)
51. How many repetitions
52. Which design was used
53. Number of imaging runs
54. alainment, normalisation and smoothing info
55. What analysis package was used?

Answer 48

• fMRI at 2T, 6 subjects, fixed effect model
• Volume: 32 slices (3mm), TR=2.8
• 20 TR per experimental epoch; 10 TR per fixation epoch
• Block design with 8 epochs per run: 4 experimental epochs (2x2 factorial design) alternating with fixation epochs
• 6 imaging runs of 126 TR (8 epochs plus 6 dummy scans)
• Images were realigned, spatially normalised to Talairach space, and smoothed (6mm) using SPM99
• data analysis with SPM99

Question 49

49.

Notes of caution on neuroimaging. State 3.

Answer 49

1. Brain imaging methods can only provide correlational (not causal) evidence for the involvement of a brain area in a cognitive task: the fact that an area becomes activated during performance on a task does not mean it actually does the task, or what exactly it does in the task.
2. Only very rarely does a single area come out of an experiment. If multiple areas, unsure which are critical for the task.
3. This question can only be answered with experimental manipulation of the areas involved: neuropsychology or transcranial magnetic stimulation, where the effect of unavailability of certain brain areas on behaviour can be assessed.

Question 50

50.

Summary:

……………………….. reveals the static physical characteristics of the brain (useful in diagnosing disease), whereas …………………………….. reveals dynamic changes in brain physiology (that might correlate with cognitive functioning).

Answer 50

Structural imaging reveals the static physical characteristics of the brain (useful in diagnosing disease), whereas functional imaging reveals dynamic changes in brain physiology (that might correlate with cognitive functioning).

Question 51

51. Summary

…………… consumes …………… from the blood. This triggers an ……………. in blood flow to that region and a change in the amount of ……………………………….. in that region.

Answer 51

Neural activity consumes oxygen from the blood. This triggers an increase in blood flow to that region and a change in the amount of deoxyhaemoglobin in that region.

Question 52

52. Summary

As the brain is always …………………….. active, functional imaging needs to measure relative changes in physiological activity. The most basic experimental design in ……………………………….. research is to subtract the activity in each part of the brain whilst doing one task away from the activity in each part of the brain whilst doing a slightly different task. This is called …………………………….

Other methods, including ……………….. and factorial designs, can minimize many of the problems associated with ……………………

Answer 52

As the brain is always physiologically active, functional imaging needs to measure relative changes in physiological activity. The most basic experimental design in functional imaging research is to subtract the activity in each part of the brain whilst doing one task away from the activity in each part of the brain whilst doing a slightly different task. This is called cognitive subtraction.

Other methods, including parametric and factorial designs, can minimize many of the problems associated with cognitive subtraction.

Question 53

53

There is no foolproof way of mapping a point on one brain onto the putatively same point on another brain because of individual differences in ……………… and …………….. anatomy. Current imaging methods cope with this problem by mapping individual data onto a common standard brain (……………………………) and by …………………… regions of significance (……………….).

Answer 53

There is no foolproof way of mapping a point on one brain onto the putatively same point on another brain because of individual differences in structural and functional anatomy. Current imaging methods cope with this problem by mapping individual data onto a common standard brain (stereotactic normalization) and by diffusing regions of significance (smoothing).

Question 54

54.

A region of ‘activity’ refers to a local increase in ………………… in the experimental task compared to the ……………. but it does not necessarily mean that the region is essential for performing the task. …………….. studies might provide evidence concerning the necessity of a region for a task.

Answer 54

A region of ‘activity’ refers to a local increase in metabolism in the experimental task compared to the baseline but it does not necessarily mean that the region is essential for performing the task. Lesion studies might provide evidence concerning the necessity of a region for a task.

Question 55

55.

………………….can be used to make …………………………… about what someone is thinking and feeling and could potentially outperform the traditional ……………….. However, it is highly unlikely that they will ever be able to produce detailed accounts of another person’s thoughts or memories.

Answer 55

Functional imaging can be used to make crude discriminations about what someone is thinking and feeling and could potentially outperform the traditional lie detectors. However, it is highly unlikely that they will ever be able to produce detailed accounts of another person’s thoughts or memories.