W3: fMRI Methods Flashcards
define
Cognitive Subtraction
A method, whereby brain activity is measured first in a “control state”, before a stimulus is presented, and again while the stimulus is presented. It is necessary to determine the activity that is due to manipulation, by subtracting the control activity from the stimulation activity
Pure Insertion
define
Inverse Problem
researchers must trace back from an underspecified signal measured at the scalp to a single postulated set of neural generators
list
2 Categories Cognitive Neuroscience Techniques can be Divided Into
1) Measurement
2) Manipulation
define + limitation
Measurement Techniques
Measure changes in brain function while a research participant engages in some cognitive activity
BUT cannot show that a region is necessary for that function
aka correlational approach
define + how
Manipulation Techniques
Examine how perturbations of the brain’s function change cognitive function / behaviour
either by transiently changing neuronal firing rates or neurotransmitter levels or by permanently damaging tissue
aka causal approach
list
3 Factors Determining Which Research Meethods to Apply to a Given RQ
1) Temporal Resolution
2) Spatial Resolution
3) Invasiveness
define, good - intermediate - poorest resolution + examples
Temporal Resolution
frequency in time with which measurements or manipulations can be made
- GOOD: techniques that record neuronal activity directly through electrophysiological means (e.g. msec precision)
- INTERMEDIATE: techniques that measure indirect metabolic correlates of neuronal activity (e.g. seconds to minutes)
- POOR: techniques that manipulate brain function through drug effects / brain lesions
define, highest - intermediate - lowest resolution + examples
Spatial Resolution
Ability to distinguish adjacent brain regions that differ in function
- HIGHEST: techniques that position electrode sensors directly within the brain (e.g. individual neurons or better)
- INTERMEDIATE: techniques of functional neuroimaging (e.g. mm-cm)
- LOWEST: techniques that measure electrical signals that spread diffusely (e.g. cm - entire brain)
(Non-)Invasive Techniques
Can they make measurements without damage to or disruption of the brain or other body tissue
- NON-INVASIVE: record endogenous brain signals using sensors outside the body and therefore can be conducted repeatedly in human volunteers with no major risk
- INVASIVE: introduce a chemical or reecording device into the body. While some can only be used in human patients (e.g. prior to neurosurgery) and / or non-human animals (e.g. single-unit recordings)
state
2 Types of Neuronal Information Processing
1) Axonal Signalling
2) Dendritic Integration
Invasive Neurophysiology
Single-Unit Recording: How Does It Work?
Measuring APs directly
CAVEAT: difficult to distinguish the firing of a single neuron from the collective firing of several neurons in close proximity thus called single-unit, refers to a single functional unit which may or may not include more than one neuron
Invasive Neurophysiology
Single-Unit Recording: Technology
- Electrodes (very fine metal wires) inserted into neural tissue, immediately adjacent to the neurons of interest
- Surgery is requires to implant electrodes which is risky!
- Cog. neuroscientists cannot target a specific neuron in humans / non-human primates; neurons are simply too small and organised in too idiosyncratic a fashion
- Instead, they lower the electrodes into a brain region of interest as ID’ed using stereotaxic coordinates
- Experimental localiser tasks may be used to evoke activity in that brain region so experimenters know when their electrode is correctly positioned
- Following the experiment, structural MRI / another method may be used to verify the track by the electrode
Invasive Neurophysiology
Single-Unit Recording: Procedures
**1) once electrode positioned, experiment can begin! **
- Data can usually be collected form a single unit for a period of minutes to hours and during this time the experimenter can collect data from hundreds of experimental trials
2) Results from recordings can be displayed in both relatively raw and averaged forms
- By convention, researchers often show raw data from a single “representative neuron” along with the average activity from all neurons that have met some criterion for inclusion
Invasive Neurophysiology
Single-Unit Recording: Advantages
- provides direct information about the rate and the timing of APs within a region
- critical in ID’ing the core functions of brain regions
- data provides grounds for many computational models of brian function, both by…
1) ID’ing the processing associated with individual neurons and
2) by helping map out supporting local circuitry
- helps reveal diversity of processes within a brian region, often showing several populations of intermixed neurons that have qualitatively different response properties (which would be largely invisble in to other techniques)
Invasive Neurophysiology
Single-Unit Recording: Limitations
- Invasive nature
- Data collection generally slow and labour-intensive with data often collected serially from one neuron / unit at a time
- Added cost of maintaining an animal colony
- Most published papers focus on neurons in a single brain region which limits the inferenes that can be drawn about complex cognitive processes
Non-Invasive Neurophysiology
EEG: How It Works
EEG signal arises form synchronous changes in the membrane potentials of the dendrites of many neighbouring neurons.
EEG can be used to either analyse..
1) Oscillations (general)
2) ERPs (time-locked)
Non-Invasive Neurophysiology
EEG: Technology
Most EEG studies record changes in electrical potential using electrodes positioned on the scalp.
EEG signals can be recorded with as few as 2 electrodes, though modern high-density electrode arrays position 64, 128 or more electrodes on the scalp, improving inferences about spatial distance of elecrtical activity.
Before analysing EEG for ERP or brain oscillations, clean data by removing noise and artifacts.
Non-Invasive Neurophysiology
EEG: Technology, Source Localisation
estimating where in the brain the activity comes from since EEG electrodes pick up signals from the scalp
Non-Invasive Neurophysiology
EEG: Procedures
1) participant fills out paper work and preactices the experimental task while electrodes put into place
2) performs task repeatedly in a series that is 30-120mins. long, taking advantage of EEG’s good temporal resolution
3) After data collection, experimenter applies processing algorithms ti minimise data quality issues and extracts trial-by-trial responses in each electrode for subsequent analyses
Non-Invasive Neurphysiology
EEG: Advantages
- non-invasive
- high temporal resolution access to electrical activity of the brain
- can be used to separate changes in brain function that occur over several hundred msecs
- relatively inexpensive (acquisition less than tenth of fMRI)
- incremendtal costs of running system (e.g. replacement of electrodes + gel) are minimal
Non-Invasive Neurphysiology
EEG: Limitations
- Imprecise spatial localisation (more severe than just “poor”)
- Inverse Problem applies thus have to focus on understanding properties of particuarl well-studied components (changes in activity over time) rather than differentiating functions of brain regioons (changes in activity over space)
- Can take a remarkably long time to set up + make sure electrodes are in place
Non-Invasive Neurphysiology
MEG: How It Works
like a middle ground b/w EEG and fMRI
When neurons are active, they generate electrical currents that also produce magnetic fields. Electrical currents run parallel to the scalp while magnetic fields extend perpendicularly into the space around the head –> MEG measures these fields
Non-Invasive Neurphysiology
MEG: Technology
(SQUIDs, structure + function)
MEG systems rely on SQUIDs
SQUID = Superconducting Quantum Inference Devices
basically specialised electrical coils which, when cooled to very low temperatures, become superconductors extraordinarily sensitive to changes in the magnetic field
measuring extremely weak magnetic fields from the brain
- SQUIDs are sensitive to all magnetic fields not jus tthose from the brain, thus MEG systems are placed in magnetically shielded rooms to attenuate contribution of external magnetic fields
- MEG systems arrange often several hundred SQUIDs in a large helmet-like device that surrounds the part’s head
- Each coil records magnetic field changes simultaneously at very high temporal res.
Non-Invasive Neurphysiology
MEG: Procedures
1) participant sits upright in the MEG system
2) participant views experimental stimuli on a screen in front of them (projector placed outside room to minimise magnetic interference)
3) responds using button box / keyboard positioned on lap
Non-Invasive Neurphysiology
MEG: Advantages
- non-invasive
- well-tolerated by human participants
- can be used with a wide range of experimental paradigms
- records data from brain simultaneously
- can provide insights into combined location and timing of a cortical activity with precision
- Like EEG: captures msec-level timing and can analyse activity as oscillations or ERPs
- Unlike EEG: magnetic fields are not significantly weakened by the skull
Non-Invasive Neurphysiology
MEG: Limitations
- Magnetic field changes produced by neuronal activity are very small, thus a challenge for detection
- Substantial cost roughly similar to those of fMRI, though MRI cost largely covered through patient scans, but not case for MEGs (couple hundred exist)
- MEG data also limits on spatial sensitivity - MEG considered to be sensitive only to neurons oriented in parallel to the skull as they generate magnetic fields perpendicular to the scalp
Metbaolic Neuroimaging
PET: How It
