Replication of individual differences in the Pirates and Planets conditioned punishment task, but no correspondence to behavior in a non-clinical sample

Open Access
Authors
Publication date June 2026
Journal Journal of Behavior Therapy and Experimental Psychiatry
Article number 102086
Volume | Issue number 91
Number of pages 6
Organisations
  • Faculty of Social and Behavioural Sciences (FMG) - Psychology Research Institute (PsyRes)
Abstract
Individuals differ in their responses to negative consequences. While some adjust quickly to avoid punishment, others persist in maladaptive behaviors despite adverse outcomes. Such differences in punishment sensitivity have been implicated in psychological disorders, including substance use, obsessive-compulsive, and mood disorders. The Pirates and Planets task (Jean-Richard-dit-Bressel et al., 2023) identified three punishment sensitivity profiles differing in punishment contingency awareness and responsiveness. However, the task's external validity and clinical relevance remain unclear.Here, we replicate and extend their findings using a within-subjects design with a larger sample (N = 188), a broader array of trait and psychopathology measures, and an additional task block to enhance profile interpretability.
We replicated the three profiles: (1) “sensitive” individuals who adapt to avoid punishment; (2) “unaware” individuals who adapt only after explicit punishment contingency information; and (3) “compulsive” individuals who persist despite such information. Crucially, no significant associations emerged between these profiles and symptom measures (i.e., obsessive-compulsive behaviors, alcohol use, depression, anxiety), suggesting these patterns may reflect context-dependent punishment learning processes that do not directly correspond to self-reported clinical symptoms, at least in this non-clinical sample.
Taken together, these findings indicate that, in its current implementation, the Pirates and Planets task is valuable for studying punishment learning mechanisms, but it may not yet be suitable for addressing clinically relevant individual differences or for translation to clinical populations. Establishing which task parameters reliably produce behavior that maps onto clinically relevant outcomes would be an essential next step for future research.
Document type Article
Language English
Published at
Other links
Downloads
Supplementary materials
Permalink to this page
Back