Fig. 3

Intra- and Inter-Subject Similarity. (A) Probability distributions for a representative segment of randomly selected 5 min data (Set 1), along with the subsequent 5 min (Set 2). The blue and green boarders denote representative subjects from Fig. 2. (B) The average inter-subject confusion matrices, computed from 50 iterations, are presented across the three kinematic domains for both feedback groups. Rows denote a subject as a predictor, while columns represent a subject as the test data. The bolded diagonal elements indicate the intra-subject similarities between Set 1 and Set 2 (blue and green squares denote representative subjects in (A)). Violin plots illustrating the intra- subject (C) and inter-subject (D) similarity of data in (B). The left side of each violin plot represents the data spread, while the right side shows a histogram of the data, and the white circle marks the median. Outliers, defined as points more than 1.5 interquartile ranges above the upper quartile or below the lower quartile, were excluded for visualization. Significant difference (p < 0.05) is indicated by *. These results suggest that intra-subject position distributions are less similar for the VF group, while inter-subject position distributions are more similar