Cognitive flexibility assessment with a new Reversal learning task paradigm compared with the Wisconsin card sorting test exploring the moderating effect of gender and stress.

Cognitive flexibility plays a crucial role in psychological health and this research aimed to investigate its assessment. We developed a novel Reversal learning task (RLT) paradigm adding pure reward (+ 100 points, 0) and punishment (- 100 points, 0) conditions to the classic reward-punishment condition (+ 100, - 100); we also analyzed the RLT convergent validity with approach-avoidance questionnaires (BIS-BAS and Approach-Avoidance Temperament questionnaire) and the Wisconsin card sorting test (WCST) scores through a Principal component analysis. In a sample of 374 participants, we found that these three conditions differently assess flexibility and that high RLT reward sensitivity in the punishment condition (0; - 100) is related with high BAS reward responsiveness. Moreover, we found that RLT and WCST flexibility scores, although associated, detect different facets of cognitive flexibility. Finally, in a second sample (N = 172), we explored the impact of stress, moderated by gender, on RLT and WCST. Whereas, WCST was not impacted by these variables, in RLT stressed women showed increased perseverative errors in punishment condition (- 100, 0) and reduced punishment sensitivity in reward condition (+ 100, 0).Overall, our newly developed RLT paradigm and the WCST seem to provide different ways to assess cognitive flexibility and to be differently affected by moderators, such as gender and stress.