Turning scalp waveforms into frequency-domain measures of oscillatory coupling — powerful when handled carefully, fragile when it is not.
qEEG takes the scalp voltage time series and summarises it as spectral power and phase coupling in the delta, theta, alpha, beta, and gamma bands. The power spectrum reports how much variance sits at each frequency; phase-amplitude coupling (PAC) reports whether the phase of a low frequency tracks the amplitude of a high frequency, the cross-frequency signature often invoked in consciousness research.
The intuition is that oscillations are not noise but a coordination mechanism: synchronising when and where sub-networks communicate. The methods measure this coordination, but the mapping from a scalp statistic to a mechanism is indirect and heavily modelled.
Every qEEG number is a product of a processing chain, and reasonable analysts can produce materially different results from identical raw data. The choices that move numbers most:
The defensible practice is to report the full chain, test that the result is not an artefact of any single link, and to prefer individual-frequency analysis over fixed bands.
qEEG’s strength is temporal resolution and portability: it captures millisecond dynamics in a cap that is cheap and widely available, making it the workhorse for sleep staging, workload, and clinical state monitoring. Its weakness is that it inherits every limitation of scalp EEG — the volume conductor blurs which cortex produced a given rhythm, and referencing and filtering choices dominate the numbers.
Treat a qEEG difference as a hypothesis about coordinated activity, not a localisation claim. When the question is where, reach for fMRI or a source-reconstruction method; when it is when and how states change, qEEG is the right instrument.