Carsten Klein
Sep 9, 2018 · 1 min read

Hi arash sayareh, the 0.2 threshold was chosen arbitrarily in the code to outline the general concept of the analysis. If you lower the number you will find more voxels that pass the threshold, as you mentioned already. If you set it to 0 you will end up with the un-thresholded maps. You might want to have a look at this paper which talks about finding thresholds in fMRI correlation analysis.

In general it would be good if you would use a threshold that is based on some significance parameter, which allows you to make a statement with what probability your active voxels were activated by the task. This link might be also helpful for you.

    Carsten Klein

    Written by

    PhD in neuroscience interested in data analysis and artificial intelligence