Random Effect - Basic Stats - Make a plot a difference

While it is important to show the data by plotting conditions, it is also important to show differences between conditions and/or groups. Here I explain how to create such difference between two conditions. If you have multiple conditions, you have to show relevant ones. For instance, for 3 conditions, and all comparisons matter, then you want to show (1 vs 2), (1 vs 3) and (2 vs 3).

What are the CI?

For each summary stat (e.g. the mean) we also plot the 95% Highest Density Intervals. These are generated using a Bayesian bootstrap. This means that for a given value, you can see the 95% probability the value is in that interval. Summary stats for beta values

Pre-requisite

You should have computed summary statistics for Beta values, with one file per condition so then create the difference. Files are named (myname)_single_subjects_Betas.mat. This also works for ERP, using the (myname)_single_subjects(stat1).mat files.

Computing and showing differences

• Step 1: load the random effect menu, select the expected channel files and click the 'Make and plot a difference' button.
• Step 2: select 2 single_subject files (those computed using limo_central_tendency_and_ci), then chose the alpha level for credible intervals and input the name of the file to save.
• Step 3: select the (2nd level) directory that corresponds to the differences you want to show, select the LIMO.mat. A figure with the individual mean values and the difference with N% CI are shown.