Neuropsychology: The Imaged Brain Flashcards
What are structural imaging methods?
-Computerized Tomography (CT)
-Magnetic Resonance Imaging (MRI)
What is a computerized tomography (CT)?
Brain imaging method in which series of X-rays is taken
-Related to X-ray absorption (radioactivity)
-Risk low if not done often
-Very little to see about brain structure: no folds, gyri, sulci, difference in grey and white matter
What is a magnetic resonance imaging (MRI)?
Brain imaging method in which magnetic fields, magnetic field gradients and radio waves are used
-No radiation involved
-Brain structure more clear to see
-Advantages over CT
-Some basics of MRI physics
What advantages does an MRI have over CT?
-Doesn’t use ionizing radiation
-Better spatial resolution
-Better discrimination between white and grey matter
-Adapted for use in fMRI
How can you measure protons in an MRI?
-Protons embedded in water-molecules
-Have magnetic field: initially random, but some protons align with external magnetic field
-Then second magnetic field introduced
What are characteristics of the second magnetic field introduced in MRI?
-Field rotates at certain frequencies (radio-wave pulse)
-Protons align with this field and rotate
-Protons absorb energy and release it when field stops
-Release of energy measured as change in electrical current
What does the signal emitted by relaxing protons depend on?
-Tissue that protons are embedded in
-Hemodynamic parameters: how blood flow changes over time
What methods can be used to link structure to function?
-Voxel-based morphometry (VBM)
-Diffusion tensor imaging (DTI)
What is voxel-based morphometry (VBM)?
-Used to link structure to function (MRI)
-Measures white and grey matter density in each voxel
-Can look for differences between groups or correlations with cognitive measures
-Multi-voxel pattern analysis: type of representations and how they change (HOW it happens)
What was found in the study with VBM about 5 important personality characteristics?
-Study: looking at differences in gray matter density related to 5 important personality characteristics
-Results: clear link between gray matter density and 4 characteristics
–>Extraversion: medial orbifrontal cortex
–>Conscientiousness: middle frontal cortex
–>Neuroticism: dorsomedial cortex, cingulate gyrus, medial temporal lobe, dorsomedial prefrontal cortex
–>Agreeableness: superior temporal sulcus, posterior cingulate cortex
What is diffusion tensor imaging (DTI)?
-Used to link structure to function (MRI)
-Measures white matter organization (connectivity)
-Based on limited diffusion of water molecules in axons: fractional anisotropy
How is DTI based on the limited diffusion of water molecules in axons?
-Water-molecules in axons can move freely within axon than across cell membrane
-If many axons aligned with each other
–>Possible to quantify effect with DTI: seeing how molecules have moved from one point to another in time
–>Described using measure called fractional anisotropy: describes degree of anisotropy of diffusion process
What is functional imaging?
-Neural activity consumes oxygen as well as generating electrical signals
–>In order to compensate for increased oxygen consumption: more blood pumped into active region
-Time taken for this response low (secs), so functional imaging has poor temporal resolution, but good spatial resolution
–>Complementary profile to ERPs
What functional imaging methods are there?
-Positron Emission Tomography (PET)
-Functional Magnetic Resonance Imaging (fMRI)
What is a positron emission tomography (PET)?
Functional brain imaging method which measures local blood flow (rCBF) by injecting radioactive tracer into bloodstream
-Very slow temporal resolution: tracer takes up to 30secs to peak and up to 1-2mins to disappear
-Effective spatial resolution: around 1cm
–>Better than EEG, worse than fMRI
-Not very useful for cognitive neuroscience, but still used for measuring metabolism, finding biomarkers, etc.
What is a functional magnetic resonance imaging (fMRI)?
-Functional brain imaging method that measures brain activity
-Does not use radioactivity, but signal affected by concentration of deoxyhemoglobin in blood: BOLD response (Blood Oxygen Level Dependent contrast)
-Change in BOLD response over time: hemodynamic response function (HRF) and has number of distinct phases
–>Not to be confused with ERP waveform, completely unrelated
-HRF peaks in 6-8 secs: limits temporal resolution of fMRI
-Has highest spatial resolution of non-invasive methods (few milimeters)
What are characteristics of HFR in fMRI?
-Changes in BOLD response over time
-Different phases: if brief moment of neural activity happens at time 0
–>First initial dip: lower signal intensity because slightly lower blood oxygenation due to oxygen consumption
–>Then overcompensation: huge increase in blood oxygenation, because lots of blood comes in due to initial dip (delayed: 4secs after neural activity)
–>Then undershoot and back to baseline
-Spread out in time: peaks in 4-6secs
-When several stimuli are presented close in time, different HRFs summate
What different experimental designs are there in cognitive neuroscience?
-Cognitive subtraction methodology
-Factorial designs
-Parametric designs
-Event-related vs blocked designs
What are characteristics of the cognitive subtraction methodology?
-What does it mean to say brain region is active
-Study Peterson et al. (1988): cognitive subtraction
-Several problems with this methodology
What does it mean to say brain region is active in cognitive subtraction methodology?
-Region active when showing greater response in one condition relative to another
-Some brain regions, called Default Network (DMN) are more active during rest
Why does a greater response in one condition relative to another in certain region mean that this region is active?
-Brain has constant supply of blood and oxygen
–>If presented one condition: difficult to say which regions dedicated to task
–>If presented more conditions: differences in brain activity between conditions can be measured
-Cognitive subtraction: inferring functional specialization of brain region by measuring relative differences in brain activity in multiple conditions
–>Involves choosing baseline/comparison condition
–>People do task in scanner and what people are doing manipulated, so specific cognitive functions can be isolated
How are some brain regions (DMN) more active during rest?
-3 regions of Default Mode Network (DMN)
–>Prenucleus and posterior cingulate cortex
–>(Medial) prefrontal cortex
–>Angular gyrus
-Method 1: comparing brain activity between task condition and rest condition
–>Series of regions less active in task condition than in rest condition
–>More difficult task, more deactivated
-Method 2: correlations in brain regions over time
–>Analysis of correlations over time as index of functional connectivity
What was the PET-study of Peterson et al. (1988) with cognitive subtraction?
PET-study to determine stages involved in reading written word aloud and producing words related to it
-As cognitive psychologist: cognitive stages from reading to speaking
-Isolating stages by presenting conditions that do or don’t involve stage
-Components of current experiment isolated by subtracting cognitive components of previous experiment
-Assumption of additivity/pure insertion: adding extra processes doesn’t change previous processes
What cognitive stages are there from reading to speaking, as a cognitive psychologist?
-Word presented
-Visual processing of word
-Recognition of written word
-Activation of meaning or sound of word
-Speech output
How were the stages isolated by presenting conditions that do or don’t involve stage in the PET-study of Peterson et al. (1988)?
-First: which region(s) involved in word recognition
-Second: which region(s) involved in word sound
-Third: which region(s) involved in word meaning
Which region(s) were involved in word recognition according to Peterson et al. (1988)?
-Experiment: passive viewing of written words vs fixation cross
-Isolating regions involved in visual processing
-Region: occipital-temporal junction
-In recent studies also different stages in word recognition
Which region(s) are involved in word sound according to Peterson et al. (1988)?
-Experiment: reading aloud vs passive viewing of written word
-Isolating regions involved in visual processing and word recognition
-Region: motor and premotor cortex
Which region(s) are involved in word meaning according to Peterson et al. (1988)?
-Experiment: generating meaning vs reading aloud written word
-Isolating regions involved in visual processing, word recognition and phonology/articulation
-Region: left inferior frontal cortex
What problems are there with the cognitive subtraction method?
-Problem with interaction
-Importance of choice of baseline
What problem is there with interaction with the cognitive subtraction method?
Ambiguous imaging data with assumption of additivity/pure insertion
-Type/amount of processing in baseline condition and experimental condition can differ
-General: addition of extra component in task can potentially change operation of other components in task
Why is the importance of baseline a problem with the cognitive subtraction method?
-Ideally: baseline similar as possible to experimental task
-Requires good cognitive theory of elements that comprise task
What are characteristics of factorial designs?
-Multiple factors combined
-Example of use: why can’t we tickle ourselves
How was a factorial design used to study why we can’t tickle ourselves?
-2 factors
–>Tactile sensation or not
–>Self movement or not
-4 conditions: A, B, C and D
–>Main effect 1: comparing A-C to B-D for brain regions involved in tactile sensation
–>Main effect 2: comparing A-B to C-D for brain regions involved in moving
–>Interaction effect: difference between touch and no touch smaller or bigger with self-movement or no self-movement
-Results: interaction in certain brain regions
-Possible to predict sensory consequences of own actions because of motor commands
What are characteristics of parametric designs?
-Associations between brain activity and changes in variable are measured
-Variable of interest treated as continuous dimension (<=> categorical)
-Use of correlations
-Example: study Price et al.: effect of how fast words are presented
How was a parametric design used in the study of Price et al. about the effect of how fast words are presented?
-In left dorso-lateral cortex: involved in working memory, activity goes up first and then falls back again due to quantity of presented words
-In primary auditory cortex: involved in hearing, the more words presented, the higher activity
-In Wernicke’s area: involved in language, strength of response invariant to rate of presenting words
-Different brain regions show different response profiles to different rates of word representation
What is important when designing an experiment?
-Decide what conditions will be
-Decide how stimuli/conditions will be presented
-Important in fMRI-experiment to consider characteristics of HRF
What are characteristics of an event-related design?
-Stimuli presented randomly
-Enables wide range of experimental designs and more closely related to typical design structure in most cognitive psychology experiments
-Can weaken differences in predicted fMRI-signal between conditions
What are characteristics of a blocked design?
-Similar stimuli grouped together presented
-More able to detect significant but small effects
How do you decide which experimental designs you’re going to use?
-How many factors are in the experiment?
–>1: parametric design or cognitive subtraction(s)
–>2 or 3: factorial design (factors could be categories or parameters; measure interactions)
–>4 or more: consider fewer factors
-What imaging method will be used? (when using 1 or 2 or 3 factors)
–>PET: blocked design
–>fMRI
-Does the experimental hypothesis require to present events randomly or events determined by participant? (when using fMRI)
–>No: blocked design
–>Yes: event-related design
What different stages are there for analysing functional imaging data?
-Data acquisition
-Correct for head movement: minimizing possibility for head to move in scanner and correcting movement that still happened
-Stereotactic normalisation: mapping individual differences in anatomy onto standard reference brain
-Smoothing: redistributing brain activity from neighboring voxels to enhance signal-to-noise ratio
-Divide data according to design
-Perform statistical comparison
What different interpretations can you give to functional imaging data?
-Inhibition vs excitation
-Activation vs deactivation
-Necessity vs sufficiency
What is inhibition vs excitation as an interpretation of functional imaging data?
-In rest: neurons still fire and influence each other
–>Excitation: one firing neuron makes another fire as well (presynaptic neuron active, postsynaptic neuron switched on)
–>Inhibition: one firing neuron makes another fire less/not (presynaptic neuron active, postsynaptic neuron switched off)
-Not measured in fMRI: energy consumption measured
–>If energy consumption changes, something changes at neural level
–>Not exactly sure if inhibition or excitation, because in both energy consumption
What is activation vs deactivation as an interpretation of functional imaging data?
-During task, some regions show activity and others don’t
–>Activation: more activity in one condition than in another
–>Deactivation: activity in conditions same
–>Not to be confused with excitation and inhibition: activation also possible in inhibition
-In fMRI: activation shown as more activity in condition A compared to condition B
–>Certain brain region activated in condition A
–>Certain brain region deactivated in condition B
What is necessity vs sufficiency as an interpretation of functional imaging data?
-Some regions active during task might not be critical to task itself
–>Necessity: active brain region necessary for task
–>Sufficiency: active brain region not necessary for task, so removing has no consequences
-Studied in fMRI and lesion-deficit analysis
How is necessity vs sufficiency measured in fMRI?
-IV: behavior/task performance, DM: changes in brain region
-Not able to make causal connection
-Tells which regions may be sufficient for cognitive functions
How is necessity vs sufficiency measured in lesion-deficit analysis?
-IV: affected brain regions, DM: behavior/task performance
-Able to make causal connection between lesion and behavior
-Tells which regions may be necessary for cognitive function
What kinds of disagreements can there be between functional imaging data and lesion data?
-Disagreement 1: imaging data imply that brain region is used in given task, but lesion data suggests region is not essential to task (imaging yes, lesion no)
-Disagreement 2: imaging data imply that region is not used in given task, but lesion data suggests region is critical to task (imaging no, lesion yes)
-Example: disagreement on region of semantic memory in patients with semantic dementia
What can be reasons for disagreement 1 between functional imaging data and lesion data?
Imaging data imply that brain region is used in given task, but lesion data suggests region is not essential to task (imaging yes, lesion no)
-Activated region (in imaging data) reflects particular strategy adopted by participants that is not essential to performing task
-Activated region (in imaging data) reflects recruitment of some general cognitive resource that is not specific to task
-Activated region (in imaging data) is being inhibited rather than excited
-Lesion studies not powerful enough to detect importance of region
What can be reasons for disagreement 2 between functional imaging data and lesion data?
Imaging data imply that region is not used in given task, but lesion data suggests region is critical to task (imaging no, lesion yes)
-If experimental task and baseline task (in imaging data) both depend critically on this region, then comparison might produce artefactual null result
-fMRI not sensitive enough: might be hard to detect activity in this region of the brain
-Impaired performance after lesion reflects damage to tracts passing through region rather than synaptic activity in gray matter of region itself
What is the example about disagreement on region of semantic memory in patients with semantic dementia?
-Left anterior temporal lobe
–>Lesion data: patients have lesion in this region, so necessary for semantic memory
–>Imaging data: patients don’t show activity in this region
–>Unable to see activity in region in imaging studies, because both conditions had tasks which involved semantic memory
-Left inferior frontal gyrus
–>Lesion data: patients don’t have lesion in this region, so not necessary for semantic memory
–>Imaging data: region implicated a lot in semantic memory tasks
–>Unable to see activity in region in lesion studies, because fMRI not sensitive enough. Task patients got was not challenging enough, so in more complex tasks differences noticed in region
