BACKGROUND AND OBJECTIVES The inferential confusion hypothesis postulates that obsessive doubt is perpetuated by a subjective form of reasoning characterized primarily by a distrust of reality and an overreliance on imagined possibilities. However, experimental evidence for this hypothesis may be compromised by a potential confound between type of information (reality vs. possibility) and its valence (danger vs. safety). In the present study we aimed to untangle this potential confound. METHODS Forty OCD and 40 non-clinical participants underwent two versions of the Inferential Processes Task (Aardema, F., et al. (2009). The quantification of doubt in obsessive-compulsive disorder. International Journal of Cognitive Therapy, 2, 188-205). In the original version, the reality-based information is congruent with the safety hypothesis, whereas the possibility-based information is congruent with the danger hypothesis. In the modified version incorporated in the present study, the reality-based information is congruent with the danger hypothesis, whereas the possibility-based information is congruent with the safety hypothesis. RESULTS Our findings did not support the inferential confusion hypothesis: both OCD and control participants changed their estimations of the probability of unwanted events based on the type of information they received (whether it conveyed danger or safety) regardless of whether it was framed as reality or possibility. LIMITATIONS The design of the present study does not lend itself to examining alternative explanations for the persistence of doubt in OCD. CONCLUSIONS The hypothesized inferential confusion in OCD requires further validation. It is particularly important to demonstrate that findings do not reflect a prudential reasoning strategy.