VI and MI values for clustering solutions (k = 2 to 10) using only TD participants. The dashed line denotes the optimal number of solutions based on the same criteria as in Fig. 2. The error bars are standard errors of the mean. Each kind of arrows points to the optimal number of solutions based on “similarity” and “dissimilarity” criteria. “n.s.” indicates that the MI or VI values are not significantly different between the 2 bracketed points. (PDF 1125 kb)