HC2, 3, 4 EEG Flashcards
Why use electroencephalogram (EEG)?
- Reaction time is the final outcome of sensory, decision and motor processes
- EEG can track the time course of these stages with millisecond precision
- EEG can inform us about cognitive processes when there is no behavioral response
Which potentials are measured by EEG?
- Post synaptic potentials at apical dendritic trees of pyramidal cells
- For brain electrical activity to be detectable through skull, must be strong signal summed over many neurons
- Pyramidal Cells in the cortex have the right properties
Post synaptic potentials at apical dendritic trees of pyramidal cells
o Action potentials can not be added up so not useful for EEG
o Post synaptic potentials can be added up so useful for EEG.
For brain electrical activity to be detectable through skull,
must be strong signal summed over many neurons
o All behaving similarly at same time
o All oriented in same way
▪ This way the potentials can be added up without canceling each other out
▪ The sum is 0 in the amygdala because of the dendritic architecture that causes the PSPs to cancel each other out.
o So negative and positive don’t cancel each other out when summed
Electromagnetic field
Is a physical field produced by electrically charged objects (e.g., a piece of brain tissue)
- It has properties of both electricity and magnetism
- Electric and magnetic field are oriented perpendicular
How is EEG measured?
- Voltage difference between two electrodes
o One electrode on the scalp
o Other non-cortical (reference) electrode (e.g. earlobes)
o Compare voltage from electrode at the brain to neutral voltage of reference electrodes (at place with no brain) - Result in rhythmic fluctuations in voltage
Reference and ground electrode locations
- These electrodes pick up noise from monitors and lights
o Ground electrodes necessary for keeping (a lot of) the noise from the outside from the EEG signal - Mastoid is the thickest point of skull, this makes it a good reference point
Electrode placement - the international 10-20 system
- Odd to the left, even to the right
- The higher the number, the more lateral placed (more towards the ears)
- Regions:
o FP= prefrontal region
o F= frontal region
o P= parietal region
o O= occipital region
o T = temporal region
o C= central region (NOT an brain region) - A1 and A2 are the electrodes at the mastoids
- Cz is the central electrode, located at the vertex
- Two lines used for the placement
o A line connecting the nasion to the inion
o A line connecting both preauricular to each other
Electrooculography (EOG)
Eye movements mess up the EEG therefore EOG necessary to “clean up” the EEG afterwards
- Large vertical lines in EEG indicates blinking
o Eyes can be seen as high voltage batteries while the brain is a low voltage battery
- Both horizontal as vertical movements can be seen in EEG
EOG - electrode locations
- Left of left eye (checks for horizontal movement)
- Right of right eye (checks for horizontal movement)
- Above the eye (checks for vertical movement)
- Underneath the eye (checks for vertical movement)
How electrodes work
- A scalp electrode picks up electric signal from the brain. Contact point is often silver chloride (AgCl)
- Signals have to travel through the skull and scalp = high resistance: lower the resistance (impedance) with conductive gel
- A ground electrode reduces electrical environmental noise.
- A reference electrode provides a biological baseline.
- EOG (eyes), ECG (heart), and EMG (muscle) electrodes are attached to monitor the artifacts.
Experimental EEG set up - you need..
o Brain with sensors (electrodes) on it
▪ The signal (voltage difference) is picked up at the electrodes attached to the scalp…..
o Converter
▪ …… Augmented by an amplifier and digitized by an analogue/digital (A/D) converter…..
▪ Amplifier increase the amplitude of the EEG signal
o Computer
▪ ……. Stored, displayed and analyzed on a computer
Analogue to digital (AD) conversion
Analogue EEG signal (micro volts) is digitized (numbers) to have time series that represent the voltage values
Sampling frequency
The rate of digitization in Hertz (Hz)
o Typical to use a sample frequency of 512 Hz
Sample rate of 500 Hz:
500 x per second = each 2 ms, take one sample (per electrode)
Nyquist-Shannon sampling theorem:
: Sample rate should be at least 2x the fastest frequencies in the signal
o If highest frequency of EEG is 100 –> use a frequency of at least 200
AD level
o Amount of information in each sample
o Bit depth → 1 bit = 2^1 (0 1) , 2 bits = 2^2 (00 01 10 11), 3 bits (000 001 010 011 100 101 110 111) etc.
Properties of the field signal
- Amplitude
- Frequency
- Phase
- The field activity oscillates in time. Thus, the signal is
represented as a time wave. The wave can also be represented as a rotation. Properties of oscillation, such as frequency, phase and amplitude, are used to describe and analyze the signals.
From EEG to ERP (event-related potentials)
ERPs are EEG changes that are time locked to sensory, motor or cognitive events that provide safe and noninvasive approach to study psychophysiological correlates of mental processes
Design of ERP experiment: Trigger (Marker)
- Two computers necessary (Stimulation computer + digitalization computer)
- Stimulation computer sends marker codes/event codes to the digitalization computer.
o For example, in Stroop task
▪ Color name and color congruent → marker code/event code 1
▪ Color name and color incongruent → marker code/event code 2 - The EEG signal (originating from the ground, active and reference electrodes) is send to the converter (filters + amplifiers) and then to the digitalization computer
- We look at tiny windows of the EEG signal we are interested in: the epoch’s
o We gather the x’s and o’s and average them → ERP
Importance of clean data
- ERPs are tiny
o Many experimental effects are less than a few millionth of a volt - ERPs are embedded in noise that is 20-100 µV
- Averaging is a key method to reduce noise
o S/N (signal/noise) ratio is a function of sqrt(# of trials)
o Doubling # of trials increases S/N ratio by 41% [sqrt(2)=1.41]
o Quadrupling # of trials doubles S/N ratio [sqrt(4)=2]
EEG measures:
- EEG oscillating signal across several electrodes
- 2 dimensions: time and location
Multiple ways to score EEG
o Peak analysis
o Topography
o Source analysis
o Time-Frequency
o Etc
Topography ≠ Neural source
Auditory N1 fronto-central peak but generated mainly in the auditory cortex
Comparison between EEG and MEG
- The common origin
- Spatio-temporal resolution is similar
- MEG: Magnetic field permeates biological tissue, fluid, and air.
o Therefore: less distortion and smearing out of the signal
o MEG is better for localization of neural sources
▪ Peak in auditory cortex found, not in Auditory N1
(as in EEG)
o MEG costs more than EEG - Butterfly plots: all sensor signals placed over each other
Examples of Event-Related Potentials
- Mismatch Negativity (MMN)
o MMN is evoked automatically by a change in a sequence of sounds - ERP components related to language
ERP component related to language
o N400 = ERP “component” related to meaning
▪ Bigger when word’s meaning doesn’t fit context
▪ Bigger for unfamiliar words
▪ May reflect amount of work required to integrate with context
o P600 = ERP “component” related to form
▪ Bigger when word not of expected type for a position in a sentence
▪ May be a type of P300 - Sometimes called Syntactic Positive Shift (SPS)
o N400 and P600 can be evoked both at once
Why ERP’s provide a continuous measure of activity at each moment in time?
This allows us to measure the brain processes that occur between the stimulus and the response instead of just measuring the behavioral response that comes at the end of these processes.
ERP - early peak
More sensory processing
ERP - later peaks
More cognitive, deeper processing
Inverse problem
The ERP’s measured on the scalp are not necessary given away the neural source of the activity.
N170 component
- Face-specific component
- Responds more to faces than non-faces
- Occipital electrode sides
- Somewhat stronger on right hemisphere.
Data analysis ERP
- Preprocessing
- Main signal processing
- Statistical testing
Preprocessing
- (Re-)reference of the EEG signals
- Preprocessing reduces noise and formats the EEG data for the main signal processing
Filtering
Correction of eye movement artefacts
Artefact rejection- Segmentation
Noise
Electric signals not from the brain
* Electromyogram (EMG)
* Electrooculogram (EOG)
* Electrocardiogram (ECG)
* Skin potentials
* Respiration
* Body motion
* Environmental noise, e.g., AC mains
* Measurement noise, bad electrode
* Artefact due to co-registration with fMRI and TMS
Noise reduction
- Visual inspection and artefact detection/rejection
- Topographic interpolation
- Independent component analysis (ICA) for removing eye movement artefacts
- Filtering
Artefact rejection
- Trade-off between settings of artefact rejection and s/n ratio of ERP
- Liberal artifact rejection: many trials in the average but contaminated with artifacts
- Conservative artifact rejection: fewer trials in the average, though less artifacts
Topographic interpolation
- When one or a few electrodes are very noisy throughout the experiment
- Replace the electrode with the mean of the surrounding electrodes
Ocular correction by independent component analysis (ICA)
ICA can be used to remove eye movement (and other) artefacts.
Unmixing - delete - mixing
Filtering - High-cut/ low-pass filter
Remove high-frequency noise (e.g., muscular artifacts, 60-80 Hz).
Filtering - Low-cut/ High-pass filter
Remove low-frequency noise (e.g., drift due to skin conductance change (<0.5HZ).
Filtering - Band-reject/ Notch filter
Filters out noise between the lower and upper thresholds (e.g., 45-55 HZ for the 50Hz AC noise).
Baseline correction
Separately per electrode:
1) Compute the average baseline activity
2) Subtract the average baseline activity from each sample in the epoch
When use mean activity?
- When large latency variation in peaks.
- When there is no peak but more sustained activity.
- When difference between conditions lies between peaks.
Data driven approach for discovering differences between ERPs
Looking at which time at which electrodes the ERPs differ significantly
