When splitting the data into different runs, I usually get a warning on the first run
Warning: Your trials are quite long, which may cause slow drifts in arousal within a trial. Try shortening the prediction windowfirst and if that doesn't help the model fits, you may need to fit a second gain in the second half of the trial to capture changes in arousal happening during the ISI.
When splitting the data into different runs, I usually get a warning on the first run
Warning: Your trials are quite long, which may cause slow drifts in arousal within a trial. Try shortening the prediction windowfirst and if that doesn't help the model fits, you may need to fit a second gain in the second half of the trial to capture changes in arousal happening during the ISI.This refers to this comparison in dataAnalysis.m:
maxTrialLength/in.sampleRate{ff} > in.predictionWindow*1.5However, I do not know how this can be TRUE for my data, where in.prediction Window=0.65, in.sampleRate{1}=300 and maxTrialLength=200 (samples).
Overall, the estimation runs fine. Do you have any idea, why this warning occurs?
...of note, the warning does not occur, when the data is formated as a single run