Protocols

The protocols behind the core matrix

Each protocol fixes who was tested, how the data were split, which electrodes and window a method saw, what it was allowed to learn, and what guessing would score. A score is comparable only with other scores under the same protocol: there is no overall ranking.

Core-matrix protocols

Protocol	Dataset	Task	Evaluation	People	Chance level	Methods scored
Motor imagery & rest	ds003810	Rest versus right-hand imagery	Transfer to a new person	10	50.0%	Spectral ridge, CSP+LDA, LaBraM, CBraMod, EEGNet
Idle & command	ds005342	Seated motor imagery · cue-gated replay	False activation × detection	4	none — detection is read with false activations	CSP+LDA, LaBraM, EEGNet, ShallowFBCSPNet, CBraMod, Deep4Net
SSVEP · 8 channels	BETA	40 visual targets · 8 posterior electrodes	Transfer to a new person	70	2.5%	Standard CCA, Spectral ridge, LaBraM, CBraMod, EEGNet
SSVEP · 4 channels	BETA	40 visual targets · 4 posterior electrodes	Transfer to a new person	70	2.5%	Standard CCA, Spectral ridge, LaBraM, CBraMod, EEGNet
Arithmetic & rest	EEGMAT	Serial subtraction versus resting EEG	Transfer to a new person	36	50.0%	Spectral ridge, LaBraM, CBraMod, EEGNet
P300 target ERP	ds006593	Visual target versus nontarget events	Transfer to a new person	21	50.0%	Spectral ridge, Temporal ridge, LaBraM, CBraMod, EEGNet
Semantic target ERP	TMNRED / ds005383	Reading · target versus nontarget events	Transfer to a new person	30	50.0%	Spectral ridge, Temporal ridge, LaBraM, CBraMod, EEGNet
Sleep staging	EESM19 scalp subset	Five stages · balanced scalp-EEG sample	Transfer to a new person	20	20.0%	Spectral ridge, LaBraM, CBraMod, EEGNet

A blank cell in the matrix is a protocol a method was not run on — not a failure. Chance level differs by protocol, so read each score against its own.

The core matrix on the home page → · Core matrix · JSON ↓ · All datasets → · All methods →