Cognitive biases, especially jumping to conclusions (JTC), are ascribed a vital role in the pathogenesis of schizophrenia. This study set out to explore motivational factors for JTC using a newly developed paradigm.

Twenty-seven schizophrenia patients and 32 healthy controls were shown 15 classical paintings, divided into three blocks. Four alternative titles (one correct and three lure titles) had to be appraised according to plausibility (0–10). Optionally, participants could decide for one option and reject one or more alternatives. In random order across blocks, anxiety-evoking music, happy music or no music was played in the background.

Patients with schizophrenia, particularly those with delusions, made more decisions than healthy subjects. In line with the liberal acceptance (LA) account of schizophrenia, the decision threshold was significantly lowered in patients relative to controls. Patients were also more prone than healthy controls to making a decision when the distance between the first and second best alternative was close. Furthermore, implausible alternatives were judged as significantly more plausible by patients. Anxiety-evoking music resulted in more decisions in currently deluded patients relative to non-deluded patients and healthy controls.

The results confirm predictions derived from the LA account and assert that schizophrenia patients decide hastily under conditions of continued uncertainty. The fact that mood induction did not exert an overall effect could be due to the explicit nature of the manipulation, which might have evoked strategies to counteract their influence.

Worry is considered a key feature of generalized anxiety disorder (GAD), whose neural correlates are poorly understood. It is not known whether the brain regions involved in pathological worry are similar to those involved in worry-like mental activity in normal subjects or whether brain areas associated with worry are the same for different triggers such as verbal stimuli or faces. This study was designed to clarify these issues.

Eight subjects with GAD and 12 normal controls underwent functional magnetic resonance imaging (fMRI) mood induction paradigms based on spoken sentences or faces. Sentences were either neutral or designed to induce worry. Faces conveyed a sad or a neutral mood and subjects were instructed to empathize with those moods.

We found that the anterior cingulate and dorsal medial prefrontal cortex [Brodmann area (BA) 32/23 and BA 10/11] were associated with worry triggered by sentences in both subjects with GAD and normal controls. However, GAD subjects showed a persistent activation of these areas even during resting state scans that followed the worrying phase, activation that correlated with scores on the Penn State Worry Questionnaire (PSWQ). This region was activated during the empathy experiment for sad faces.

The results show that worry in normal subjects and in subjects with GAD is based on activation of the medial prefrontal and anterior cingulate regions, known to be involved in mentalization and introspective thinking. A dysregulation of the activity of this region and its circuitry may underpin the inability of GAD patients to stop worrying.

There are few theoretical proposals that attempt to account for the variation in affective processing across different affective states of bipolar disorder (BD). The Interacting Cognitive Subsystems (ICS) framework has been recently extended to account for manic states. Within the framework, positive mood state is hypothesized to tap into an implicational level of processing, which is proposed to be more extreme in states of mania.

Thirty individuals with BD and 30 individuals with no history of affective disorder were tested in euthymic mood state and then in induced positive mood state using the Question–Answer task to examine the mode of processing of schemas. The task was designed to test whether individuals would detect discrepancies within the prevailing schemas of the sentences.

Although the present study did not support the hypothesis that the groups differ in their ability to detect discrepancies within schemas, we did find that the BD group was significantly more likely than the control group to answer questions that were consistent with the prevailing schemas, both before and after mood induction.

These results may reflect a general cognitive bias, that individuals with BD have a tendency to operate at a more abstract level of representation. This may leave an individual prone to affective disturbance, although further research is required to replicate this finding.

Many studies have used negative mood induction techniques to investigate the effect of emotional state on cognitive performance but positive mood induction paradigms have been used less frequently. The objective of this study was to investigate the effect of positive mood induction on emotional processing in euthymic individuals with bipolar disorder (BD) and controls.

Previously, we reported that positive mood induction using a novel technique based on feedback produced a longer-lasting effect in euthymic individuals with BD than controls (Farmer et al.2006). Here we report the effect of mood induction on two tests of emotional processing, the Affective Go/No-go test (AGNG) and the Cambridge Gamble task (CGT), on which BD patients in the manic phase differ in their performance from controls.

Following positive mood induction, bipolar cases exhibited a positive emotional bias on the AGNG and performed more slowly than controls on the CGT, particularly when making more difficult decisions.

These data confirm that positive mood induction is more effective in individuals with BD than controls. They also suggest that alterations in decision making and attentional biases occur even with transient and subtle changes in mood in bipolar disorder.