Since the rankings of neuropsychological variables varied widely across different parameters of the random forest classification, these features were statistically compared across the participant groups using one-way analysis of variance (ANOVA) to determine the level of significance (see Table 5). Only two variables from the visual span test (i.e., TotCor_Fw and Span_Fw) were significant after Bonferroni corrections. The score “TotCor_Fw” (total number of correctly performed forward trials), which showed the highest significance level, was also identified by some of the parameters of the random forest as a variable contributing to group classification.
