Threshold-free cluster-enhancement (TFCE)

Source: R/compute_tfce_array.R

Compute the TFCE statistics and the corresponding p-values.

compute_tfce_array(distribution, graph, alternative, E, H, ndh)

Arguments

distribution

An 3d array representing the null distribution of multiple signal. The first dimension is the permutations, the second the samples, the third is the channels.
graph

A igraph object representing the adjacency of the channels.
alternative

A character string indicating the alternative hypothesis. Either "greater", "less" or "two.sided".
E

A numeric indicating the parameter associated to the cluster extend. Suggestion E = 0.5.
H

A numeric indicating the parameter associated to the cluster height. Suggestion for t statistic: H = 1, for F statistic: H = 2.
ndh

A numeric indicating the number of step for the approximation of the integral. Suggestion ndh = 500.

Value

A list containing an igraph and a data.frame, with the results for each sample, channel.

References

Smith, S. M., & Nichols, T. E. (2009). Threshold-free cluster enhancement: addressing problems of smoothing, threshold dependence and localisation in cluster inference. Neuroimage, 44(1), 83-98, https://doi.org/10.1016/j.neuroimage.2008.03.061.

See also

Other MCP: compute_clusterdepth_array(), compute_clustermass_array(), compute_stepdownmaxT_array(), compute_troendle_array()