Analysis of Accuracy and Timing in Decision-Making Tasks

The present study investigates the separation abilities by age and gender based on raw data of two-alternative force choice decision-making task in visuo-motor experiment. The applied methodology is based on machine learning procedure for finding, assessing, and interpreting existing dependencies in interested data spaces. The procedure applies fuzzy cluster analysis to discrimate the biosignal data of the visual task where the location of the pattern center is determined by form cues, motion cues, or by their combination. The obtained grouping results are assessed according to the participants’ age and gender. Further, these results are compared against the results obtained of statistical parameters data of a hierarchical drift-diffusion model (HDDM) processed by the same machine learning methodology. Differences in the subjects’ capabilities to perform the visuo-motor task are summarized. It was found that age groups could be recognized with similar success by both raw and HDDM data clustering analyses. Between factors analysis strongly underlines the informativity of the reaction time. Dynamic conditions are better performed for age distinction in both cases. However, the gender is better recognizable in HDDM data space. The group of young people is characterized by low reaction time and middle value of accuracy in their responce, whereas the reverse is valid for the middle-aged participants.