Suggestion: within-subjects confidence intervals

Martin suggests to provide within-subject confidence intervals, citing [this](https://motorbehaviour.wordpress.com/2014/10/31/confidence-intervals-in-within-participant-design-a-tutorial-on-the-cousineau-morey-method/) explanation, but notes that this only makes sense when multiple subjects are given.