What Is Electroencephalography?

Electroencephalography (EEG) is a neurophysiological technique that records the electrical activity of the brain by measuring voltage fluctuations produced by the synchronous firing of populations of neurons. Electrodes placed on the scalp detect the summed postsynaptic potentials of cortical pyramidal neurons, amplify them to the millivolt range, and digitize the resulting time series for analysis. The technique is non-invasive, requires no injected contrast agents or ionizing radiation, and produces data at millisecond temporal resolution, a combination that makes it uniquely suited to capturing the rapid dynamics of neural processes. EEG belongs to the broader class of bioelectric phenomena, which encompasses any electrically measurable signal generated by biological tissue.

The method was first demonstrated in humans by Hans Berger in 1924, who recorded the alpha rhythm, a 8 to 13 Hz oscillation visible when a person closes their eyes, from scalp electrodes. The International 10-20 electrode placement system, introduced in the 1950s, standardized the positions of 19 to 256 electrodes across the scalp, providing a reproducible spatial reference that enables comparison of recordings across laboratories and clinical sites worldwide.

Signal Acquisition

A complete EEG system comprises electrodes, a conductive paste or gel to reduce skin-electrode impedance, differential amplifiers to suppress common-mode noise, analog filters to remove power-line interference, and an analog-to-digital converter sampling at typically 250 to 2,000 Hz. Active electrodes, which incorporate pre-amplification at the electrode tip, have reduced impedance requirements and simplified the logistics of setup, enabling ambulatory and wireless recording systems that can collect data outside the laboratory or clinic. The spatial resolution of scalp EEG is limited by volume conduction: signals from a cortical source spread and blur as they pass through brain tissue, cerebrospinal fluid, skull, and scalp before reaching the electrodes, confining reliable source localization to regions several centimeters across. Intracranial EEG, in which electrodes are placed on the cortical surface or within brain tissue, bypasses this limitation at the cost of a surgical procedure, and is used when precise localization of epileptic foci is required.

Frequency Bands and Signal Interpretation

EEG signals are conventionally parsed by frequency content into five named bands, each associated with distinct states of brain function. Delta activity (0.5 to 4 Hz) predominates during deep slow-wave sleep. Theta rhythms (4 to 8 Hz) appear during drowsiness, memory encoding, and meditative states. Alpha oscillations (8 to 13 Hz) index visual cortex idling and relaxed wakefulness. Beta rhythms (13 to 30 Hz) accompany active motor planning and sensorimotor processing. Gamma oscillations (above 30 Hz) are linked to high-level cognitive processing and feature binding in sensory cortex. As described in the NCBI Bookshelf introduction to clinical EEG, pathological patterns, including epileptiform spikes, spike-wave discharges, and diffuse slowing, provide the basis for diagnosing seizure disorders, encephalopathies, and structural injuries.

Signal Processing and Analysis

Modern EEG analysis relies on a range of signal-processing tools developed to extract physiologically meaningful features from noisy, non-stationary recordings. A PMC review of EEG signal processing methods covers techniques including independent component analysis for separating artifactual components such as eye blinks and muscle activity from neural signals, time-frequency analysis using short-time Fourier transforms and wavelet decompositions, and source localization algorithms that estimate the cortical generators of scalp potentials. Machine learning classifiers trained on EEG features now underpin real-time brain-computer interfaces that allow individuals with motor impairments to control external devices by modulating their own neural activity. Connectivity metrics derived from multi-electrode EEG, including coherence and Granger causality, quantify the functional relationships among brain regions during cognitive tasks and pathological states. PMC's study of EEG source localization details the inverse problem methods used to recover these spatial estimates from scalp recordings.

Applications

Electroencephalography has applications in a wide range of fields, including:

• Diagnosis and monitoring of epilepsy, including presurgical localization of seizure foci
• Sleep staging in clinical polysomnography and research into sleep disorders
• Brain-computer interfaces for communication and motor rehabilitation
• Intraoperative and intensive-care monitoring of cerebral function during anesthesia
• Cognitive neuroscience research into attention, memory, language, and decision-making