Effects of metacognitive training on the neurophysiological correlates of the jumping-to-conclusions bias in schizophrenia

The jumping-to-conclusions (JTC) bias, i.e. the tendency to arrive on decisions based on limited evidence, occupies a central place among theories of delusion generation. Since delusions constitute one of the core symptoms of schizophrenia, the JTC bias might represent an ideal target for treatment. Although not affected by dopaminergic antagonists, the JTC bias is amenable to specific psychotherapeutic interventions, which have been shown to be efficient in treating delusional symptoms as an adjunct to antipsychotic medication treatment in patients with schizophrenia. However, the neurobiological bases of JTC in schizophrenia have received little interest so far. A few studies have investigated data gathering in healthy subjects and have shown it to be associated with activity in brain areas different than (or additional to) those involved in decision making and reward processing. However, so far there have been no studies on the neurophysiological correlates of disturbed data gathering in patients with schizophrenia. The present study aims to provide a biological marker for the JTC bias and other aspects of data gathering abnormalities in patients with schizophrenia. This will be achieved by comparing brain activation and functional connectivity patterns in 24 unmedicated patients with schizophrenia and 24 healthy control subjects (matched in gender, age and education) during two different tasks assessing data gathering. Moreover, changes in brain activation/connectivity patterns in patients following a course of metacognitive training (MCT), an intervention specifically designed to address reasoning biases such as the JTC. The study aims to provide insights into the pathophysiology of schizophrenia that will further the development of new treatment options for the disorder.

Reasoning biases such as the jumping-to-conclusions bias (JTC) are thought to significantly contribute to delusions in schizophrenia, and metacognitive interventions targeting these biases, such as metacognitive training (MCT), may improve delusions. So far, it is not clear whether JTC depends on dopaminergic reward areas that constitute the locus of action of antipsychotic drugs, or rather on dopamine-independent cortical areas. Activations across widely distributed brain areas including the prefrontal, parietal, medial temporal cortex, thalamic/striatal areas and the insula have been reported in connection with specific aspects of probabilistic reasoning, but previous findings are difficult to integrate into one coherent account. The objective of the present project was to investigate the neurophysiological correlates of probabilistic reasoning, and their changes following MCT, in patients with delusions. Functional MRI brain activation and functional connectivity patterns during a probabilistic reasoning paradigm (Box Task) were examined in 27 healthy controls and in 30 patients with delusions. Only medication-free or medication-resistant patients were included in the project in order to minimize antipsychotic medication effects. A second fMRI was carried out in patients after a course of MCT. In healthy controls, conclusion events were associated with extensive activations in widely distributed brain areas associated with the task-positive network (TPN), such as the dorsal anterior cingulate cortex, right prefrontal cortex, medial occipital areas, thalamus, bilateral (mainly inferior) parietal areas, and anterior insula. In contrast, draw events were characterized by higher activation in brain areas assumed to be part of the tasknegative network (TNN), i.e. inferior medial prefrontal cortex, ventral anterior cingulate cortex, posterior cingulate cortex, angular gyrus, and right posterior insula. There were no differences between patients and controls in brain activation and connectivity patterns. However, presence of JTC (irrespective of group) was associated with lower activity in the dorsal striatum, and lower connectivity between this area and inferior parietal cortex, during final conclusion events. In addition, MCT responders exhibited higher activity in areas of the taskpositive network and higher connectivity between inferior parietal cortex (BA40) and salience/reward areas. In summary, the present study indicates that probabilistic reasoning in the absence of a reward context is dependent on the balance between the task-positive and task-negative network, and that shifts in effective connectivity between the two networks may be crucial for data-gathering. Although JTC is associated with dysregulated activity in reward system areas that constitute the main target of antipsychotic drugs, its improvement may be dependent on cortical areas belonging to the task-positive network, and their ability to regulate the reward system. These findings point to additional pathophysiological mechanisms, supplementary to dopaminergic antagonism, which may help improve JTC and delusions.