Electroencephalography (EEG): Signals, Uses and Configurations

Scalp EEG measures electrical potential differences generated largely by synchronized postsynaptic activity in neuronal populations. These signals reach the electrodes after propagating through brain tissue, cerebrospinal fluid, skull and scalp, so the recorded activity reflects population-level electrophysiology and is spatially shaped by volume conduction, the spread of electrical fields through brain tissue, cerebrospinal fluid, skull and scalp.

EEG systems commonly analyze spontaneous rhythms, event-related potentials (ERPs), brief voltage changes time-locked to a stimulus or event and changes in spectral power, which describes how strongly EEG activity is represented within a defined frequency band or connectivity. Interpretation depends on montage, reference, electrode quality, filtering, sampling and the experimental or clinical context. [1–3]

Routine and sleep EEG follow formal recording standards because their purpose can include clinical interpretation, especially in epilepsy and related neurophysiology. Long-term video-EEG adds synchronized behaviour and video to electrophysiology. [1,4]

Research EEG may instead prioritize high-density coverage, mobile recording, precise stimulus timing, source analysis or integration with VR, eye tracking, EMG or stimulation. The value of a higher channel count depends on the scientific question, spatial coverage and planned analyses.

EEG is used when the timing of brain activity matters. The same basic measurement can support very different applications depending on montage, recording duration, task design and analysis.

EEG is the dominant non-invasive BCI signal because it is comparatively accessible, portable and fast. For example, motor-imagery BCIs use changes in sensorimotor rhythms—EEG rhythms over sensorimotor cortex that change during real or imagined movement. P300-based BCIs use a positive EEG response that often becomes prominent roughly 300 milliseconds after a task-relevant or infrequent stimulus. Steady-state visual evoked potential (SSVEP) BCIs use rhythmic EEG responses that follow the frequency of a periodically flickering visual stimulus. [5]

The challenge is that EEG is sensitive to eye movements, facial and neck muscle activity, cable movement, poor electrode contact and environmental electrical noise. Good BCI performance therefore depends as much on acquisition quality and experimental design as on the classifier.

EEG systems range from compact few-channel mobile setups to 32- or 64-channel general research systems and 128-channel or higher-density arrays. Configurations also differ in electrode technology, mobility and access to real-time data.