Phase Lag Index (Stam, Nolte & Daffertshofer 2007) between every channel pair. More...
#include "../connectivity_global.h"#include "abstractmetric.h"#include "../connectivitysettings.h"#include <QSharedPointer>#include <QMutex>#include <Eigen/Core>

Go to the source code of this file.
Classes | |
| class | CONNECTIVITYLIB::PhaseLagIndex |
| Phase Lag Index estimator (Stam et al. 2007); rejects zero-lag volume-conduction mixing. More... | |
Namespaces | |
| namespace | CONNECTIVITYLIB |
| Functional connectivity metrics (coherence, PLV, cross-correlation, etc.). | |
Phase Lag Index (Stam, Nolte & Daffertshofer 2007) between every channel pair.
SPDX-License-Identifier: BSD-3-Clause Copyright (c) 2018-2026 MNE-CPP Authors
The Phase Lag Index measures the asymmetry of the cross-spectral phase distribution about zero,
PLI_{xy}(f) = | E[ sign( Im( S_{xy}(f) ) ) ] |
with output in [0, 1]: 0 means phase differences are symmetric about 0 or pi (i.e. dominated by zero-lag mixing) and 1 means the phase lag always has the same sign (perfect, consistently lagged or leading interaction). Stam, Nolte and Daffertshofer (Human Brain Mapping, 2007) introduced PLI specifically to suppress the spurious connectivity that volume conduction and common-reference effects produce in EEG/MEG sensor-space coherence: any pair of channels driven by the same source with constant gains has a phase difference of exactly 0 or pi and therefore PLI = 0 by construction.
PLI shares the anti-leakage property of imaginary coherence but, being amplitude-independent, is more robust to amplitude artefacts at the cost of being biased upward for small samples and discontinuous around the zero-phase line. The weighted variants WeightedPhaseLagIndex and DebiasedSquaredWeightedPhaseLagIndex (Vinck et al., 2011) and the sample-bias correction UnbiasedSquaredPhaseLagIndex address those shortcomings.
Definition in file phaselagindex.h.