EEG

Backlinking

EEG History

EEG (update!)

Electroencephalography (EEG) consist of recording of the spontaneous electrical activity generated by active neurons in the brain, mainly the cortex. In general, the brain produces electrical activity divided into two parts – simultaneous rhythmic components, often called brain waves, and event related potentials, related to sensory stimulation or task-related thinking (Furth & Ph, 2018).

Figure 87. EEG temporal and spatial resolution.

EEG mainly records the activity of pyramidal cells of the cortex (layers VI, V, IV, III/II) (Thomson et al., 2007). The EEG does not record APs, but it records fPSPs, either excitatory or inhibitory. A single PSP is not strong enough to be registered by the EEG but fPSPs generate a dipole which is strong enough to be registered by the EEG. The temporal summation of several PSPs, which reflects the synchronous activity of the cortex, contributes to the intensity of the signal to be picked up by the EEG; in fact, it is necessary that there be at least $6 c m^{2}$ of synchronous activity for there to be a signal. The orientation of the neurons and of the fPSP also contribute to signal intensity. If the orientation of EPSPs and IPSPs overlap, then there is charge neutralisation and the signal will be weaker. Finally, the tissues which the signal must transverse, such as fat, skin, bone, CSF, and muscle, modulate the signal itself. To measure deep brain activity from the scalp, those deep brain sources must produce powerful fields, and there should be many trials for averaging. For example, studies on brainstem-generated potentials generally have thousands of trials to obtain sufficient signal-to-noise. The second reason why activity from deep brain structures is difficult to measure from the scalp is that populations of neurons in subcortical structures are not often arranged in a geometrically parallel orientation - leading to a cancellation of potentials.

The temporal resolution of EEG is determined by the sampling rate of the acquisition. It is generally between hundreds and a few thousands of samples per second. The temporal precision, in contrast, depends on the analysis applied. temporal accuracy is extremely high because brain electrical activity travels instantaneously (within measurement possibilities) from the neurons generating the electrical field to the electrodes that are measuring those fields.

Figure 88. Bull’s-eye illustration of the differences among resolution (R), precision (P), and accuracy (A). Up-and down arrows indicate high and low levels. Resolution is illustrated by the number of dots, precision is illustrated by the spread of the dots, and accuracy is illustrated by the distance of the dots away from the center of the bull’s-eye.

Although EEG has high temporal precision, resolution, and accuracy, its spatial precision, resolution, and accuracy are all relatively low compared to high-spatial-resolution imaging techniques such as fMRI. The spatial resolution of EEG is determined by the number of electrodes. The spatial precision of EEG is fairly low but can be improved by spatial filters such as the surface Laplacian or adaptive source-space-imaging techniques. The spatial accuracy of EEG is low. Activity recorded from one electrode does not reflect only activity from neurons directly below that electrode, but rather, from a complex mixture of activities from many brain regions close to and distant from that electrode.

Lastly, the signals captured by the EEG reflect meso- and macroscopic scales, but not microscopic scales which involve individual neurons or columns of neurons.

Figure 89. Orientation of the signal influences the amplitude captured in the EEG. Negative values are the result of the electromagnetic field imposed by the dipole in which there is electronegativity closer to the electrode (fEPSPs).

Comments

06/07/2024 - 100 years of EEG!

Info

A single minicolumn or even a single macrocolumn (containing about $1000$ minicolumns or $100, 000$ pyramidal cells) is not expected to generate a dipole moment of sufficient strength to produce scalp potentials in the recordable range of a few microvolts.

As a general "rule of head," about $6 c m^{2}$ of cortical gyri tissue (containing about $600, 000$ minicolumns or $60, 000, 000$ neurons forming a dipole layer) must be "synchronously active" to produce recordable scalp potentials without averaging (Cooper et al. 1965; Ebersole 1997).

The tissue label "synchronously active" in this context is based on cortical recordings with macroscopic electrodes and must be viewed mainly as a qualitative description. In the case of dipole layers in fissures and sulci, tissue areas larger than $6 c m^{2}$ are apparently required to produce measurable scalp potentials.

As such diffusion physics is still restrained on the amount of cortical activation in order to produce measurable electrical activity in the range of $μ$ $V$ . Neuron arrangement geometry is also thought to aid signal strength, i.e. the parallel arrangement of Pyramidal cell dendritic axes encourages superposition of fields from many individual synaptic sources ,by forming large Dipole layers.

Summary

EEG signal magnitude determinants:

Neuron geometry
Synaptic Distribution
Source Depth

As such, EEG may originate with a mixture of network and non-network activity.

Clean EEG

Clean EEG should be stationary (On preprocessing EEG data). However, its ambiguity makes it an unpopular concept to assess whether the data is clean.

Cost

EEG setups can cost as much as fMRI or EMG ones (around 150,000$) (cohenAnalyzingNeuralTime2014).

References

Links

https://www.learningeeg.com/

Notes

list from [[]] and !outgoing([[]]) and (#baby or #child or #adult or #topic or #domain) 
sort file.name asc

Articles

list from [[]] and !outgoing([[]]) and #article 
sort file.name asc

Others

list from [[]] and !outgoing([[]]) and !(#baby or #child or #adult or #topic or #domain or #article) 
sort file.name asc