Introduction: A number of studies have suggested that deluded patients show a "jumping to conclusions" reasoning style on probabilistic reasoning tasks. In order to systematically explore the cognitive underpinnings of this task, we compared deluded and nondeluded patients on a number of experimental manipulations to investigate the role of memory and task pragmatics on performance.
Methods: In Study 1, the performance of deluded and nondeluded schizophrenia patient groups was compared to nonpsychiatric controls on a battery of probabilistic reasoning tests. In Study 2, two variants of the standard "beads in jars" task were compared in order to explore the possible role of working memory load on task performance.
Results: In Study 1, there were no significant differences between any of the groups on any of the probabilistic reasoning tasks. In Study 2, we found a significant difference between the two schizophrenic groups and the controls, but no difference in performance between deluded and nondeluded patient groups. The deluded group responded fastest in the memory intensive condition.
Conclusions: Deluded and nondeluded schizophrenic patients perform similarly on probabilistic reasoning tasks and only show the "jumping to conclusions" response pattern under some conditions but not under others. Memory demands may influence the appearance of this pattern of responding in schizophrenia.