REJTREND Detect linear trends in EEG activity and reject the epoched trials based on the accuracy of the linear fit.

Usage: >> [rej rejE] = rejtrend( signal, winsize, minslope, minR, step);

Inputs:
signal   
3 dimensional signal (channels x frames x trials)
winsize   
integer determining the number of consecutive points for the detection of linear patterns
minslope   
minimal absolute slope of the linear trend of the activity for rejection
minR   
minimal R^2 (coefficient of determination between 0 and 1)
step   
step for the window. Default is 1 point (2 points will divide by two the computation time)

Outputs:
rej   
rejected trials. Array with 0 or 1 for each trial.
rejE   
rejected rows of the rejected trials

Algorithm: Looked for all possible windows of size 'winsize' of each trial if
the linear fit have minimum slope and R^2

Author: Arnaud Delorme, CNL / Salk Institute, 2001

See also: eeglab()

See the matlab file rejtrend.m (may require other functions)

Back to functions