Jumping to delusions? Paranoia, probabilistic reasoning, and need for closure

Abstract

Introduction. The present study was designed to replicate and extend the findings of Bentall and Swarbrick (2003). It was hypothesised that patients with a history of persecutory delusions would display higher need for closure and a more extreme jumping to conclusions bias than healthy control participants.

Methods. Twenty-two patients with a history of persecutory delusions and nineteen healthy control participants were administered a probabilistic reasoning task, along with self-report measures of depression and need for closure.

Results. The clinical group scored higher on need for closure than the controls, but showed no greater tendency to jump to conclusions. No relationship was found between need for closure and a jumping to conclusions bias.

Conclusions. The results confirm an association between persecutory delusions and need for closure, yet suggest that persecutory delusions in an outpatient sample can be seen in the absence of a jumping to conclusions bias.

Acknowledgements

This work was supported by a Training Fellowship in Psychiatric Research from the New South Wales Institute of Psychiatry, awarded to the first author, and also by the Neuroscience Institute of Schizophrenia and Allied Disorders (NISAD), utilising infrastructure funding from NSW Health. Presentation of this work at the Canadian Society for Brain, Behaviour and Cognitive Science 15th annual meeting was funded by an Overseas Conference Grant from the British Academy, awarded to the first author.

Notes

One participant from the clinical group chose the incorrect jar during the beads task, and was therefore excluded from the analyses of results for this task.

Two clinical participants did not complete the NFC questionnaire validly and were excluded from the analysis.

³Due to a concern about the homogeneity-of-variance assumption, a nonparametric equivalent of this analysis was run, with virtually identical results.

⁴This analysis was repeated after excluding four participants with an NFC “lie score” greater than 15, which is part of the standard procedure for scoring the NFC Scale with nonclinical populations as recommended by Kruglanski. The pattern of results was unchanged.