EEG is one of the most powerful tools for investigating brain activity and understanding neurological disorders. But its real value is not in the raw signal — it is in what gets extracted from it: the características (in Spanish, características) that make it possible to analyze brain function in depth and detect potential anomalies.
From these features, researchers and clinicians draw valuable information about the complexity of the human brain. They fall into three broad families: time domain, frequency domain and synchrony.
01Time-domain features
They analyze the EEG signal over time. One key feature is amplitude, which measures the intensity of electrical activity. Another is the mean absolute deviation (MAD), which reports on signal stability. The zero-crossing rate (ZCR) helps identify rapid changes in the waveform.
These features are widely used for real-time monitoring and for spotting artifacts in EEG recordings.
- Amplitude
- Mean absolute deviation (MAD)
- Zero-crossing rate (ZCR)
02Frequency-domain features
They analyze the EEG signal across the frequency spectrum. Power spectral density (PSD) is a common feature that quantifies how power is distributed across frequency bands. Relative power (RP) computes the share of power in specific bands — delta, theta, alpha, beta and gamma. Entropy, which measures signal complexity, helps assess cognitive states.
- Power spectral density (PSD)
- Relative power (RP)
- Entropy
Applications
Frequency features have broad applications in brain-computer interfaces (BCI), sleep analysis and cognitive research. They are central to classifying EEG patterns tied to motor imagery in BCI, identifying sleep stages and telling apart cognitive load during mental tasks.
03Synchrony features
They examine the relationship between different EEG channels or brain regions. Coherence, widely used, quantifies the level of synchronization between two signals in specific frequency bands. The phase locking value (PLV) is another essential measure for assessing the phase synchronization of neural oscillations. These features are valuable for investigating brain connectivity and studying neural networks.
- Coherence
- Phase locking value (PLV)
Applications
Synchrony features have become important in studying disorders such as epilepsy, where abnormal synchronization patterns often appear. They are also key to mapping functional connectivity networks in healthy brains.
What they add up to
EEG features offer valuable information about brain function, cognitive states and neurological disorders. Time-domain features — amplitude, MAD, ZCR — are crucial for real-time monitoring; frequency-domain features — PSD, RP, entropy — help classify brain states; and synchrony features — coherence, PLV — support the study of connectivity and of disorders.
Understanding these features unlocks the enormous potential of EEG, from brain-computer interfaces and sleep analysis to the diagnosis and treatment of neurological disorders.
How we use this at Databrain
In neuromarketing we work with these same features — especially relative power per band and spectral density — to measure, with the Enobio EEG system, how the brain responds to a brand, an ad or an experience. The raw signal decides nothing; what we turn into business evidence is precisely what gets extracted from it.
Glossary
- Amplitude
- The intensity of the recorded electrical activity.
- MAD (mean absolute deviation)
- An indicator of signal stability.
- ZCR (zero-crossing rate)
- How often the signal changes sign; it detects rapid changes.
- PSD (power spectral density)
- How the signal’s power is distributed across frequency bands.
- Relative power
- The share of power in a specific band (delta, theta, alpha, beta, gamma).
- Entropy
- A measure of signal complexity; useful for assessing cognitive states.
- Coherence
- The level of synchronization between two signals in a given band.
- PLV (phase locking value)
- A measure of phase synchronization between neural oscillations.
- BCI
- Brain-computer interface (interfaz cerebro-computadora).
Let’s bring EEG to your next study
At the Databrain Lab we apply EEG and research-grade biosensors to measure attention, arousal and emotional response, in the lab and in the field.
Request a studyFrequently asked questions
What are EEG features?
They are metrics extracted from the EEG signal to analyze brain function and detect anomalies. They fall into three families: time domain, frequency domain and synchrony.
What are the EEG frequency bands?
Delta, theta, alpha, beta and gamma. Relative power in each band is associated with different cognitive states and is one of the most widely used features.
What are EEG features used for?
For brain-computer interfaces (BCI), sleep analysis, cognitive research, the study of brain connectivity and the diagnosis of neurological disorders such as epilepsy.
References
- Niedermeyer, E., & da Silva, F. L. (2004). Electroencephalography: Basic principles, clinical applications, and related fields. Lippincott Williams & Wilkins.
- Hjorth, B. (1970). EEG analysis based on time domain properties. Electroencephalography and Clinical Neurophysiology, 29(3), 306-310.
- Buzsáki, G., & Draguhn, A. (2004). Neuronal oscillations in cortical networks. Science, 304(5679), 1926-1929.
- Stam, C. J., Nolte, G., & Daffertshofer, A. (2007). Phase lag index: assessment of functional connectivity from multi-channel EEG and MEG with diminished bias from common sources. Human Brain Mapping, 28(11), 1178-1193.
Adapted and translated into English, with a Latin American focus, from the article “EEG Features for Comprehensive Brain Analysis” published by Neuroelectrics.
