Cognitive flexibility in women who recovered from anorexia nervosa – a model-based approach

Research findings on cognitive flexibility (CF) functioning in women who recovered from anorexia nervosa (RAN) were found to be inconsistent. This was attributed to the multiple definitions of CF and the diverse measuring tools used to assess it. Applying a deductive approach to explore CF function may address these inconsistencies; thus, we used a model that divides CF into three subtypes, namely, stimulus-response mapping, switching sets and task switching. Additionally, we explored the association between CF subtypes and the disorder's clinical measures to assess the relation of CF to recovery. Forty-three RAN and 54 healthy controls performed tasks designed to assess CF subtypes based on the model's division, and the RAN group completed the Eating Disorder Examination Questionnaire. The results showed that the RAN group performed significantly worse than controls only in the stimulus-response mapping subtype. Additionally, there were no correlations between CF subtypes and clinical symptoms or the disorder measures – current and nadir body mass index, age of onset, time since recovery, and disorder duration. In conclusion, the study revealed CF impairment after recovery from AN, specifically in stimulus-response mapping. The variability in performance of the CF subtypes supports the application of a theory-driven perspective viewing CF as a modular ability in RAN. Additionally, CF is unrelated to clinical measures post-recovery and thus may not be used as a criterion for evaluating recovery.