Figure 4. .

Train-test classification for task or subject identification. (A) The train-test procedure consists in splitting the sessions in a train set and a test set. The train set is used to optimize the classifier (here find the best boundaries between labels; see dashed gray lines), whose accuracy is then measured on the test set. (B) Accuracy for the classification of test sessions to discriminate movie versus rest (two labels) for the MLR, 1NN, and LDA classifiers, as indicated by the colors. The connectivity measures are indicated on the x-axis: The masked FC is the subset of FC features corresponding to the SC mask used for the MOU-EC model; “heur” corresponds to the symmetric connectivity matrix obtained using the heuristic optimization in Appendix D in the Supporting Information. The classifiers are trained with 80% of the subjects and tested on the remaining subjects, over 40 repetitions. The dashed line indicates the chance level, which corresponds to predicting always movie (three sessions out of five). The MLR gives perfect classification for all connectivity measures except for MAR. (C) Same as panel B for the classification of session to identify the four tasks (three movie sessions and rest). The chance level corresponds to predicting always rest (two sessions out of five). (D) The left confusion matrix indicates the errors of the MLR classifier trained with five labels (two labels for rest here), as indicated by the dark green off-diagonal pixels. As expected from panel B, rest and movie are well separated. The movie sessions (M1 to M3) are distinguishable to some extent, but not the rest sessions (R1 and R2). The right confusion matrix is the equivalent for the 1NN classifier. (E–F) Same as panels C–D for the identification of the 22 subjects. Here 40 splits of the data are considered for two sessions in the train set and three sessions in the test set, irrespective of rest/movie and individually for each subject. (G) Model error for the MOU-EC and heuristic optimizations in the two conditions.