Lecture 5 EEG Flashcards by Anna Wilson

Which neurologist developed the first technique of recording electrical activity at the scalp?

A) Wegner
B) Berger
C)

B) Berger (1929)

How well did you know this?

Not at all

Perfectly

What causes the electrical activity at the scalp measured by EEG?

When neurons fire they are changing electrical polarisation

When lots of neurons fire at the same time- the electrical activity sums and we are able to measure it

The activity of populations of neurons firing together produced a change in electrical polarisation in the brain

This can be measured by electrodes on the scalp in EEG

How well did you know this?

Not at all

Perfectly

What makes signal diffusion a limitation of EEG?

The electrical signals have to pass through lots of tissue (including the skull) before you can measure them
This gives EEG very poor spatial resolution
Nearby electrodes are often highly correlated with each other
It is hard to tell where a signal is coming from

How well did you know this?

Not at all

Perfectly

How are problems with sulci limitations of EEG?

Human brains are very convoluted (they have many folds)
Opposite potentials across a sulcus can cancel each other out
But the magnetic field should be OK because it is orthogonal
So MEG can sometimes detect signals that are lost to EEG

How well did you know this?

Not at all

Perfectly

In EEG we are measuring a difference in electrical charge (the difference between two places)

What is the technical term for this difference?

Potential difference or voltage

How well did you know this?

Not at all

Perfectly

Voltage is always measures as a relative.

There must always be a reference electrode.

Where can these electrodes be located?

Behind the ear, nose, leg, arm

How well did you know this?

Not at all

Perfectly

EEG voltages are typically very large

True or False?

False

EEG voltages are very small.

The voltages we measure are tiny, typically in the order of tens of microvolts
A microvolt is a millionth of a volt
So much much less than a battery produces

How well did you know this?

Not at all

Perfectly

Give examples of components of EEG

Electrodes
Electrode caps
Amplifiers
Triggers
Filters

How well did you know this?

Not at all

Perfectly

What are electrodes traditionally made of?

Silver/ silver chloride

How well did you know this?

Not at all

Perfectly

Are electrodes in EEG MRI or MEG compatible?

Usually no, but you can get specialised systems with electrodes that are compatible

How well did you know this?

Not at all

Perfectly

What is a montage?

A standard referencing system to determine electrode placement

How well did you know this?

Not at all

Perfectly

What is the most common montage?

The 10-20 system

How well did you know this?

Not at all

Perfectly

In a 10-20 system, odd numbers are located on the right and even on the left, true or false?

True

How well did you know this?

Not at all

Perfectly

In a 10-20 system, what do the following letters stand for?

O
P
T
F
C
A
M
z

O – Occipital
P – Parietal
T – Temporal
F – Frontal
C – Central
A – Anterior*
M – Mastoid
z – zero
* or sometimes earlobe

How well did you know this?

Not at all

Perfectly

When placing a cap, what are the two reference points that determine where certain electrodes should sit?

The naison (just above the bridge of the nose)

The inion (the bump at the base of the skull)

How well did you know this?

Not at all

Perfectly

What is electrical impedance?

Opposition to the electrical current

How well did you know this?

Not at all

Perfectly

How can electrical impedance be lowered in EEG?

By using a conductive gel or paste

How well did you know this?

Not at all

Perfectly

Is an impedance of <5 ohms considered good?

Yes

How well did you know this?

Not at all

Perfectly

Why do EEG voltages need to be amplified?

Because they are too small

How well did you know this?

Not at all

Perfectly

Online monitoring- contemporary systems give a live trace of all electrodes- what sorts of normal activities give off a response?

Blinking and clenching teeth

How well did you know this?

Not at all

Perfectly

What are low pass filters?

Smoothing filters that make the waveforms easier to interpret

How well did you know this?

Not at all

Perfectly

EEG has a high temporal resolution, true or false?

True

Data analysis- why is it important to average waveforms over many trials?

Lots of noise- true signal can be revealed by averaging over multiple trials/ subjects

Data analysis- what is artifact removal?

Removal of trials where data was corrupted by some artefact e.g. electrical equipment such as a fan

Different ways of getting rid of this data:
Various labs use different rejection criteria: e.g. removing trials where some measure exceeded ±3SE of the mean
Other sophisticated analysis techniques exist for detecting things like blinks, such as Independent Components Analysis (ICA)

Blink detection

* Blinks show up on the electro-oculogram at the bottom of the plot * But there are also voltage changes at some scalp electrodes * Some are positive, some are negative * ICA can use these to detect and remove blinks

What is an event-related potential?

A complex waveform, averaged over many trials/ stimulus presentations It has multiple positive and negative components Everything is relative to stimulus onset in EEG

In ERPs are negative voltages usually plotted upwards or downwards?

Upwards! Always check!

Common ERPS ERP components can also be labelled by the (approximate) time they occur in milliseconds What is P100?

P100- Basic visual response (occipital)

Common ERPS ERP components can also be labelled by the (approximate) time they occur in milliseconds What is N100?

N100- Basic auditory response

Common ERPS ERP components can also be labelled by the (approximate) time they occur in milliseconds What is N170?

N170- produced by face stimuli

Common ERPS ERP components can also be labelled by the (approximate) time they occur in milliseconds What is P300?

P300 reflects decision making

Common ERPS ERP components can also be labelled by the (approximate) time they occur in milliseconds What is LRP (lateralised readiness potential)?

LRP- indicator of motor planning

How are ERPs usually quantified (which two things are measured)?

Peak amplitude Latency

Multivariate pattern analysis is a newer technique for analysing EEG data, what do they show and how are they advantageous over traditional ERP averages?

MPA adds a temporal dimension to the analysis MPA tells you more about information available (??)

What is Fourier Analysis?

A data analysis technique- any waveform can be broken down or constructed from the sum of sine waves of different frequencies Fourier analysis- building up one waveform from lots of sine waves

The EEG waveform and the Fourier spectrum represent the same information in different ways, true or false?

True

Frequency vs time plots represent how the fourier spectrum changes over time, what does the colour represent?

The colour represents the amplitude (or the power, which is amplitude squared) at each frequency

The brain produces synchronised activity in different frequency bands , what are some examples of these standard bands?

– Delta – <4Hz, sleep, some attention tasks – Theta – 4-7Hz, similar to Alpha – Alpha – 7-14Hz, resting, tiredness, attentional lapses – Beta – 15-30Hz, motor behaviour, concentration – Gamma – >30Hz, long range neural synchronization

Steady state

* An alternative technique is to make the brain oscillate at a particular frequency * We do this by making the stimulus oscillate in intensity at that frequency * Measure the ‘steady-state evoked potential’ – the amplitude at the stimulus frequency * Works for vision, hearing, somatosensory etc. * Advantage- you know exact frequency, intensity of stimulus can be changed

What is source localisation (visualising responses)

The inferred location of activity in the brain from the distribution of activity across electrodes

What are some uses and benefits of EEG for cognitive neuroscience (types of experiments)?

* Almost all areas of cognitive neuroscience are investigated using ERP designs, e.g. * Face processing * Semantic processing * Memory * Learning * Motor tasks * Attention * The literature is huge

What are some uses and benefits of EEG in clinical research?

* Originally used to diagnose and monitor epilepsy, EEG monitoring is also used in a range of clinical conditions, including: * Migraine * Movement disorders (e.g. Parkinson’s) * Psychiatric disorders * Diagnosing brain death, coma etc. * It is now a standard clinical tool in some settings

What are some uses and benefits of EEG in sleep research?

* The different stages of sleep are associated with characteristic patterns of EEG activity * EEG can be used to investigate sleep disorders, and study consolidation of learning during sleep

What are some uses and benefits of EEG in basic sensory research? (Steady state EEG is becoming more popular, what is a major advantage of this technique?)

* Used widely in hearing, e.g. to assess hearing deficits, the efficacy of cochlear implants * For vision research, steady-state EEG is becoming more popular * Has the advantage that it can be used on subjects who cannot respond to a behavioural task, e.g. very young children. Developmental studies often use EEG

EEG has excellent temporal resolution but poor spatial resolution, true or false?

True

EEG has excellent spatial resolution but poor temporal resolution, true or false?

False

fMRI has excellent spatial resolution but poor temporal resolution, true or false?

True

Why use EEG? List EEG's major advantages

* It is cheap – consumables cost <£2 per subject * Doesn’t require major investment to set up a lab, and is widely available * It has excellent temporal resolution * Non-invasive, and easily combined with other techniques * Analysis techniques are well developed * But critically, use it because it helps you answer a question – the best research is driven by theory, not by methodology

Give examples of some potential future uses of EEG

Human-computer interaction * We used to interact with computers using keyboards, then mice, now touch, what next? * Most manufacturers are investigating brain-computer interfaces Gaming headsets Neurofeedback * Some people have tried to use EEG as a way of monitoring their own internal mental state * A bit like biofeedback for adjusting your own heart rate, aiding meditation, relaxation, stress or anger management etc. * This may or may not work, but it is an interesting use of the technology, and has produced a range of very cheap, but potentially useable consumer hardware

An EEG waveform measures the change over time of the Current at a point on the scalp Voltage at a point on the scalp Voltage between two points on the scalp Current between two points on the scalp

Voltage between two points on the scalp

Special equipment is required to combine EEG with MRI or MEG because Standard EEG electrodes give off powerful magnetic fields The liquid helium used to cool the MEG sensors reacts with EEG gel Standard EEG systems contain too much metal There is insufficient space inside the MEG/EEG scanner for normal electrodes

Standard EEG systems contain too much metal

A contour plot shows the distribution of activity across sensors at a specific point in time a fourier analysis of all sensors along a single axis the ratio of induced to evoked activity the source location of EEG/MEG signals at different time points

the distribution of activity across sensors at a specific point in time

In regard to the N and P waves of the EEG, N and P stand for neocortical and paleocortical novel and practiced negative and positive nascent and persistent

negative and positive