Fig. 1: Prediction accuracy.

Distribution of prediction accuracy over different cross-validation iterations is shown for three NIH toolbox outcomes. Permutation test revealed significant predictions for total and fluid cognition (p < 0.05 corrected). The box plot on each violin displays the interquartile range, with the center line representing the median prediction accuracy. The dots are outliers, and the asterisks (*) signify the significance of predictions for total and fluid cognition.