2. Electroencephalography & Magnetoencephalography
1. M/EEG – TECHNICAL BASICS
o EEG – Basic Idea
Neural communication has an electric component: When neurons are active, they change their
membrane potentials (i.e., differences between intra- and extracellular voltage)
Depending on differences in neural activity > see tiny differences in electric fields emitted by brain
If we could sensitively measure these electric field over time > could see what’s going on in the
brain!
o Recording Electric Brain Signals on the Scalp - EEG was first established in the 1920s
German psychiatrist Hans Berger measured electric voltage difference between two scalp electrodes
He for the first time established human cortical rhythms – the alpha and beta oscillations
By amplifying the signal between the EEG electrode and the reference, we
can record the voltage difference across time
o Components of a Modern EEG System – 3 key components
A cap of electrodes mounted on the participants’ head, usually containing
many EEG channels
- Typically consist of 16 – 128 electrodes
- Electrodes are arranged in a standardized system (10-20-system), where
electrodes are positioned between reference points (L-R: left to right periauricular
points; P-A: Inion to Nasion) in steps of 10% - 20% - 20% - 20% - 20% - 10%.
- Occipital electrodes (O), Parietal electrodes (P), Temporal electrodes (T), and
Frontal (polar) electrodes (F), Auricular electrodes (A)
- Odd numbers are left, even number are right
An amplifier that enhances the recorded signal
A battery to power the amplifier and a computer receiving the signal
The EEG Signal
- EEG signals do not just contain information from
the brain
- Indeed, some of the most prominent EEG signals
are related to artefacts coming from muscular
activity, the cardiovascular system, and even line
power
- Typically, EEG data need to be cleaned (e.g., by deleting contaminated data
segments) prior to further analyses
o MEG – Basic Idea
Changes in electric fields are always accompanied by magnetic fields orthogonal to them
Instead of measuring the electric field generated by neural activity, we can also
measure the magnetic field
MEG signals can thus be interpreted in very similar ways as EEG signals
Changeable magnetic fields induce currents in receiver coils. These currents are the
MEG signal.
Problem: The magnetic field changes, and thus the induced currents, are extremely small…
Solution: Make the receiver coils superconductive by cooling them with liquid helium (this makes
MEG comparably expensive).
o MEG Scanners
MEG systems contain hundreds of superconductive magnetic field measuring sensors (SQUIDS)
The system needs to be placed in a magnetically shielded room
The participant’s is placed underneath a helmet containing these sensors
The recorded signal looks highly similar to the signal recorded from an EEG system
o EEG vs MEG
EEG… - Is attached to the participants head,
- Is cheap and highly available thus less prone to movement
, - Suffers from differences in volume - Participants can move inside the MEG
condition, which makes it spatially helmet
imprecise - Magnetic fields are not so much
- Unreliable in higher signal frequencies hampered by volume conduction, thus
spatially more precise
-Higher reliability in higher signal
MEG… frequencies
- Is expensive in only available in few
places
o BCI: Brain Computer Interface – Locked-in syndrome
Rare neurological disorder
Complete paralysis of voluntary muscles
Patients are conscious, can think and reason, but are unable to speak or move
The hope is to enable people with locked-in syndrome to communicate by
- Using either ERPs or specific frequencies
2. THE NEURAL BASIS OF EEG SIGNALS
o EEG and Concerted Neural Activation
EEG signals are mainly driven my large electric dipoles that yield strong electric voltage differences
EEG is thus only sensitive to electric field changes of concerted ensembles of neurons; only when
multiple neurons’ electrical activity stacks up, it can be recorded with EEG!
Single action potentials are too fast and often too asynchronous to be measurable with EEG; we
rather see concerted changes in membrane potentials
The primary source of the EEG signal are dipoles formed by pyramidal neurons
When pyramidal neurons receive presynaptic inputs to their dendrites, their membrane
potentials change
For example, when there is an excitatory input, positive currents travel towards the
soma (intracellular) and negative currents travel in the opposite direction
(extracellular), forming a dipole
Pyramidal cells form such dipoles, because in layer IV of the neocortex, many of
these cells are organized in a rigid spatial arrangement, with their axons reaching
deep and their apical dendrites placed more superficially
Additionally, pyramidal cells are well interconnected and show synchronized
activation patterns, allowing electric charges to stack across neurons
o EEG responses versus Microelectrode Recordings
M/EEG signals mirror the properties of extracellular local field potentials (with
the exception that they are pooled across many neurons)
Changes of local field potentials can be correlated with neural spiking, but they
do not need to – this ultimately depends on the synchrony of neural firing
o Excitatory and Inhibitory Inputs – problem1
Different EEG signals can be generated based on whether the inputs are excitatory
(EPSPs) and inhibitory (IPSPs), and depending on where the inputs terminate
It is thus not immediately clear whether EEG reflects excitatory or
inhibitory activations!
o Spatial Precision of EEG Signals – problem2
Neural activation does not always show up in the nearest electrode
Where the EEG response can be measured depends on the orientation of the dipole
Because of cortical folding, activations can be strongest in electrodes far away from
the activation!
SUMMARY
o EEG signals arise from the coordinated activation of many neurons, which forms electric dipoles
o EEG signals are a reflection of changes in local field potentials
o The magnitude of EEG signals does not straightforwardly map onto (1) neural excitation and inhibition
and (2) the signal’s origin in brain space
1. M/EEG – TECHNICAL BASICS
o EEG – Basic Idea
Neural communication has an electric component: When neurons are active, they change their
membrane potentials (i.e., differences between intra- and extracellular voltage)
Depending on differences in neural activity > see tiny differences in electric fields emitted by brain
If we could sensitively measure these electric field over time > could see what’s going on in the
brain!
o Recording Electric Brain Signals on the Scalp - EEG was first established in the 1920s
German psychiatrist Hans Berger measured electric voltage difference between two scalp electrodes
He for the first time established human cortical rhythms – the alpha and beta oscillations
By amplifying the signal between the EEG electrode and the reference, we
can record the voltage difference across time
o Components of a Modern EEG System – 3 key components
A cap of electrodes mounted on the participants’ head, usually containing
many EEG channels
- Typically consist of 16 – 128 electrodes
- Electrodes are arranged in a standardized system (10-20-system), where
electrodes are positioned between reference points (L-R: left to right periauricular
points; P-A: Inion to Nasion) in steps of 10% - 20% - 20% - 20% - 20% - 10%.
- Occipital electrodes (O), Parietal electrodes (P), Temporal electrodes (T), and
Frontal (polar) electrodes (F), Auricular electrodes (A)
- Odd numbers are left, even number are right
An amplifier that enhances the recorded signal
A battery to power the amplifier and a computer receiving the signal
The EEG Signal
- EEG signals do not just contain information from
the brain
- Indeed, some of the most prominent EEG signals
are related to artefacts coming from muscular
activity, the cardiovascular system, and even line
power
- Typically, EEG data need to be cleaned (e.g., by deleting contaminated data
segments) prior to further analyses
o MEG – Basic Idea
Changes in electric fields are always accompanied by magnetic fields orthogonal to them
Instead of measuring the electric field generated by neural activity, we can also
measure the magnetic field
MEG signals can thus be interpreted in very similar ways as EEG signals
Changeable magnetic fields induce currents in receiver coils. These currents are the
MEG signal.
Problem: The magnetic field changes, and thus the induced currents, are extremely small…
Solution: Make the receiver coils superconductive by cooling them with liquid helium (this makes
MEG comparably expensive).
o MEG Scanners
MEG systems contain hundreds of superconductive magnetic field measuring sensors (SQUIDS)
The system needs to be placed in a magnetically shielded room
The participant’s is placed underneath a helmet containing these sensors
The recorded signal looks highly similar to the signal recorded from an EEG system
o EEG vs MEG
EEG… - Is attached to the participants head,
- Is cheap and highly available thus less prone to movement
, - Suffers from differences in volume - Participants can move inside the MEG
condition, which makes it spatially helmet
imprecise - Magnetic fields are not so much
- Unreliable in higher signal frequencies hampered by volume conduction, thus
spatially more precise
-Higher reliability in higher signal
MEG… frequencies
- Is expensive in only available in few
places
o BCI: Brain Computer Interface – Locked-in syndrome
Rare neurological disorder
Complete paralysis of voluntary muscles
Patients are conscious, can think and reason, but are unable to speak or move
The hope is to enable people with locked-in syndrome to communicate by
- Using either ERPs or specific frequencies
2. THE NEURAL BASIS OF EEG SIGNALS
o EEG and Concerted Neural Activation
EEG signals are mainly driven my large electric dipoles that yield strong electric voltage differences
EEG is thus only sensitive to electric field changes of concerted ensembles of neurons; only when
multiple neurons’ electrical activity stacks up, it can be recorded with EEG!
Single action potentials are too fast and often too asynchronous to be measurable with EEG; we
rather see concerted changes in membrane potentials
The primary source of the EEG signal are dipoles formed by pyramidal neurons
When pyramidal neurons receive presynaptic inputs to their dendrites, their membrane
potentials change
For example, when there is an excitatory input, positive currents travel towards the
soma (intracellular) and negative currents travel in the opposite direction
(extracellular), forming a dipole
Pyramidal cells form such dipoles, because in layer IV of the neocortex, many of
these cells are organized in a rigid spatial arrangement, with their axons reaching
deep and their apical dendrites placed more superficially
Additionally, pyramidal cells are well interconnected and show synchronized
activation patterns, allowing electric charges to stack across neurons
o EEG responses versus Microelectrode Recordings
M/EEG signals mirror the properties of extracellular local field potentials (with
the exception that they are pooled across many neurons)
Changes of local field potentials can be correlated with neural spiking, but they
do not need to – this ultimately depends on the synchrony of neural firing
o Excitatory and Inhibitory Inputs – problem1
Different EEG signals can be generated based on whether the inputs are excitatory
(EPSPs) and inhibitory (IPSPs), and depending on where the inputs terminate
It is thus not immediately clear whether EEG reflects excitatory or
inhibitory activations!
o Spatial Precision of EEG Signals – problem2
Neural activation does not always show up in the nearest electrode
Where the EEG response can be measured depends on the orientation of the dipole
Because of cortical folding, activations can be strongest in electrodes far away from
the activation!
SUMMARY
o EEG signals arise from the coordinated activation of many neurons, which forms electric dipoles
o EEG signals are a reflection of changes in local field potentials
o The magnitude of EEG signals does not straightforwardly map onto (1) neural excitation and inhibition
and (2) the signal’s origin in brain space
