EEG Weighted Phase Lag Index (wPLI)

Computes the Weighted Phase Lag Index and its debiased variant between all EEG channel pairs. The wPLI reduces the influence of volume conduction by weighting the phase differences by the magnitude of the imaginary part of the cross-spectrum, computed via Welch's method.

Usage

eegWPLI(
  x,
  band = c(8, 13),
  window_sec = 2,
  overlap = 0.5,
  assay_name = NULL,
  debiased = TRUE
)

Arguments

x: A PhysioExperiment object with EEG data (2D: time x channels).
band: Numeric vector of length 2 specifying the frequency band in Hz over which to compute wPLI (default: c(8, 13) for alpha band).
window_sec: Window length in seconds for spectral estimation (default: 2).
overlap: Overlap fraction between adjacent windows, from 0 to 1 exclusive (default: 0.5).
assay_name: Name of the input assay. If NULL, the default assay is used.
debiased: Logical; also compute the unbiased debiased wPLI (default TRUE).

Value

A data.frame with columns:

channel1: Character or integer identifier of the first channel.
channel2: Character or integer identifier of the second channel.
wpli: Numeric wPLI value in [0, 1].
wpli_debiased: Numeric debiased wPLI (Vinck et al., 2011, Eq. 6), computed by the shared wpliEstimate estimator so it matches PhysioCrossModal::weightedPLI. It is unbiased and therefore distributes around 0 for independent signals (it is not clamped to be non-negative).

Details

$$wPLI = |mean(Im(S_{xy}))| / mean(|Im(S_{xy})|)$$

The debiased wPLI corrects for sample-size bias: $$wPLI^2_{debiased} = (N \cdot wPLI^2 - 1) / (N - 1)$$

References

Lachaux, J. P., et al. (1999). Measuring phase synchrony in brain signals. Human Brain Mapping, 8(4), 194-208.

Vinck, M., et al. (2011). An improved index of phase-synchronization for electrophysiological data in the presence of volume-conduction, noise and sample-size bias. NeuroImage, 55(4), 1548-1565.

Examples

if (FALSE) { # \dontrun{
pe <- make_eeg(n_time = 5000, n_channels = 4, sr = 500)
wpli_df <- eegWPLI(pe, band = c(8, 13))
head(wpli_df)
} # }