TASK 5: MRI, PET, FMRI
STRUCTURAL & FUNCTIONAL IMAGING
Functional imaging (PET, fMRI) – detect changes in metabolism / blood flow in the brain
while participants are engaged in cognitive tasks
Don’t directly measure neural events BUT metabolic changes correlated with
neural activity
Neurons require oxygen & glucose to generate energy
When brain area is active, more oxygen & glucose are provided by increasing
blood flow to that region
Measures moment to moment variable characteristics of the brain
Structural imaging (CT, MRI)
Different types of tissues have different physical properties – these different
properties can be used to construct detailed static maps of the physical structure
of the brain
Different magnetic properties, because of the amount of water they have in
their tissue
CT – different types of tissues absorb X-rays differently able to differentiate
between tissues
Spatial resolution – accuracy with which one can measure where a cognitive event is
occurring
Increasing voxel size lower spatial resolution more sensitive to BOLD
response
Decreasing voxel size higher spatial resolution less sensitive to BOLD
response
Optimal voxel size – 3-4mm
Temporal resolution – accuracy with which one can measure when a cognitive event is
occurring
PET
THE MECHANISMS BEHIND IT
1. Radioactive substance introduced into bloodstream (tracer)
Radiation omitted from tracer is monitored by PET instrument
, Tracer administered at least twice (during control & experimental condition)
2. Radioactive isotopes within injected substance rapidly decay release a positron from their
atomic nuclei
Most common isotopes (administered while person is engaged in cognitive task)
Oxygen-15; O (administered in form of water)
Fluorine-18 (administered in form of glucose sugar)
3. Positron collides with an electron two photons / gamma rays are created
Travels 2-3mm before collision
BUT need to average across participants in PET spatial resolution a bit worse
(10mm)
4. Two photons move in opposite directions at speed of light
5. PET scanner detects gamma rays & determines where the collision took place
Results reported as change in regional cerebral blood flow (rCBF)
Spatial resolution – PET can resolve metabolic activity to regions (voxels) of approx. 5-
10mm3
Sufficient to identify cortical & subcortical areas of enhanced activity & functional
variation
Temporal resolution (bad) – 30s, time it takes for radiation to peak to its maximum
Specialised molecules that might serve as biomarkers of particular neurological
disorders (e.g., PiB)
PiB – binds to beta-amyloid (biomarker of Alzheimer’s disease)
If used as PET tracer in might be able to diagnose Alzheimer’s disease
BLOCK DESIGN EXPERIMENT
Used for PET, because participants must be engaged continually in single
experimental task ≥ 40s
Recorded neural activity integrated over a block of time – participant is either presented a
stimulus / performs a task
Recorded activity pattern compared to other blocks that have been recorded while
doing the same task / stimulus, different task / stimulus, nothing at all
Advantage: able to detect small effects
(DIS-)ADVANTAGES
Less susceptible to signal distortion around air cavities than fMRI
Poor temporal resolution compared to e.g., ERPs
Difficult to interpret data – huge data sets, comparisons of experimental & control
conditions produces many differences
Difficult to make inferences about each area’s functional contribution correlation ≠
causation
STRUCTURAL & FUNCTIONAL IMAGING
Functional imaging (PET, fMRI) – detect changes in metabolism / blood flow in the brain
while participants are engaged in cognitive tasks
Don’t directly measure neural events BUT metabolic changes correlated with
neural activity
Neurons require oxygen & glucose to generate energy
When brain area is active, more oxygen & glucose are provided by increasing
blood flow to that region
Measures moment to moment variable characteristics of the brain
Structural imaging (CT, MRI)
Different types of tissues have different physical properties – these different
properties can be used to construct detailed static maps of the physical structure
of the brain
Different magnetic properties, because of the amount of water they have in
their tissue
CT – different types of tissues absorb X-rays differently able to differentiate
between tissues
Spatial resolution – accuracy with which one can measure where a cognitive event is
occurring
Increasing voxel size lower spatial resolution more sensitive to BOLD
response
Decreasing voxel size higher spatial resolution less sensitive to BOLD
response
Optimal voxel size – 3-4mm
Temporal resolution – accuracy with which one can measure when a cognitive event is
occurring
PET
THE MECHANISMS BEHIND IT
1. Radioactive substance introduced into bloodstream (tracer)
Radiation omitted from tracer is monitored by PET instrument
, Tracer administered at least twice (during control & experimental condition)
2. Radioactive isotopes within injected substance rapidly decay release a positron from their
atomic nuclei
Most common isotopes (administered while person is engaged in cognitive task)
Oxygen-15; O (administered in form of water)
Fluorine-18 (administered in form of glucose sugar)
3. Positron collides with an electron two photons / gamma rays are created
Travels 2-3mm before collision
BUT need to average across participants in PET spatial resolution a bit worse
(10mm)
4. Two photons move in opposite directions at speed of light
5. PET scanner detects gamma rays & determines where the collision took place
Results reported as change in regional cerebral blood flow (rCBF)
Spatial resolution – PET can resolve metabolic activity to regions (voxels) of approx. 5-
10mm3
Sufficient to identify cortical & subcortical areas of enhanced activity & functional
variation
Temporal resolution (bad) – 30s, time it takes for radiation to peak to its maximum
Specialised molecules that might serve as biomarkers of particular neurological
disorders (e.g., PiB)
PiB – binds to beta-amyloid (biomarker of Alzheimer’s disease)
If used as PET tracer in might be able to diagnose Alzheimer’s disease
BLOCK DESIGN EXPERIMENT
Used for PET, because participants must be engaged continually in single
experimental task ≥ 40s
Recorded neural activity integrated over a block of time – participant is either presented a
stimulus / performs a task
Recorded activity pattern compared to other blocks that have been recorded while
doing the same task / stimulus, different task / stimulus, nothing at all
Advantage: able to detect small effects
(DIS-)ADVANTAGES
Less susceptible to signal distortion around air cavities than fMRI
Poor temporal resolution compared to e.g., ERPs
Difficult to interpret data – huge data sets, comparisons of experimental & control
conditions produces many differences
Difficult to make inferences about each area’s functional contribution correlation ≠
causation
