Purpose: Isolate the signal in a specific frequency band.
Method: FIR filter (Finite Impulse Response) using the Parks–McClellan algorithm (firpm).
Filter parameters:
- Transition band: ±0.5 Hz from the passband boundaries
- Stopband attenuation: 80 dB
- Filter order: calculated automatically
Filter types:
- Type 0: No filtering
- Type 1: Band-pass filter (single band)
- Type 2: Ratio of two bands (band1 / band2)
Available ranges (Type 1):
- Theta (4–8 Hz)
- Theta1 (4–6 Hz)
- Theta2 (6–8 Hz)
- Alpha (8–13 Hz)
- Alpha1 (8–10 Hz)
- Alpha2 (10–13 Hz)
- Beta (13–30 Hz)
- Beta1 (13–20 Hz)
- Beta2 (20–30 Hz)
- Gamma (30–100 Hz)
Band ratios (Type 2):
- Alpha/Theta (8–13 Hz / 4–8 Hz)
- Alpha/Beta (8–13 Hz / 13–30 Hz)
Use case: Isolation of standard EEG bands, their sub-bands (theta1, theta2, alpha1, alpha2, beta1, beta2), and computation of band ratios.
Physiological explanation of frequency bands:
Delta (0.5–4 Hz):
- Physiology: the slowest rhythms; associated with deep sleep (stages 3–4), coma, and pathological conditions
- Neural mechanisms: synchronization of large cortical regions, reduced metabolism
- When it appears: in adults, only during sleep; in infants, also during wakefulness (normal)
- Clinical significance: excessive delta during wakefulness may indicate pathology
Theta (4–8 Hz):
- Physiology: associated with light sleep, meditation, creative thinking, and learning
- Neural mechanisms: synchronization of the hippocampus and cortex; important for memory consolidation
- When it appears: with eyes closed, in relaxation, during REM sleep
- Clinical significance: elevated theta may indicate drowsiness or pathology
Theta sub-bands:
- Theta1 (4–6 Hz): slower theta, more often associated with drowsiness and transitional states
- Theta2 (6–8 Hz): faster theta, more often associated with attention and cognitive processing
Alpha (8–13 Hz):
- Physiology: the “resting rhythm”; dominant during relaxed wakefulness with eyes closed
- Neural mechanisms: synchronization of occipital and parietal cortex; associated with the absence of visual stimulation
- When it appears: with eyes closed, in a calm state; disappears when the eyes open or during mental effort
- Clinical significance: absence of alpha may indicate pathology; alpha asymmetry between hemispheres may indicate problems
Alpha sub-bands:
- Alpha1 (8–10 Hz): slower alpha, more often associated with relaxation and calmness
- Alpha2 (10–13 Hz): faster alpha, more often reflects readiness for information processing
Beta (13–30 Hz):
- Physiology: active wakefulness, attention, problem solving
- Neural mechanisms: cortical desynchronization, active information processing
- When it appears: with eyes open, during mental activity, movement
- Clinical significance: increased beta may indicate stress, anxiety, and tension
Beta sub-bands:
- Beta1 (13–20 Hz): associated with active thinking and attention
- Beta2 (20–30 Hz): higher beta, more often associated with tension and stress
Gamma (30–100 Hz):
- Physiology: high-level cognitive processing, binding information from different areas
- Neural mechanisms: rapid synchronization of distant cortical regions; “binding” features into a unified perception
- When it appears: during complex cognitive tasks, perception, and attention
- Clinical significance: gamma activity abnormalities are associated with schizophrenia and autism
Band ratios (Type 2):
- Alpha/Theta: an index of relaxation relative to drowsiness/slower processes
- Alpha/Beta: an index of the balance between relaxation and cognitive activity
Why filtering is important:
The EEG signal contains all frequencies simultaneously. Filtering makes it possible to isolate the frequency range of interest and analyze it separately. For example, alpha activity is needed to study relaxation, while beta activity is used to study concentration.