Learning to trust: trust and attachment in early psychosis.

BACKGROUND: Distrust and social dysfunction are characteristic in psychosis and may arise from attachment insecurity, which is elevated in the disorder. The relationship between trust and attachment in the early stages of psychosis is unknown, yet could help to understand interpersonal difficulties and disease progression. This study aimed to investigate whether trust is reduced in patients with early psychosis and whether this is accounted for by attachment avoidance and attachment anxiety. METHOD: We used two trust games with a cooperative and unfair partner in a sample of 39 adolescents with early psychosis and 100 healthy controls. RESULTS: Patients had higher levels of attachment anxiety, but the groups did not differ in attachment avoidance. Basic trust was lower in patients than controls, as indicated by lower initial investments. During cooperation patients increased their trust towards levels of controls, i.e. they were able to learn and to override initial suspiciousness. Patients decreased their trust less than controls during unfair interactions. Anxious attachment was associated with higher basic trust and higher trust during unfair interactions and predicted trust independent of group status. Discussion Patients showed decreased basic trust but were able to learn from the trustworthy behaviour of their counterpart. Worries about the acceptance by others and low self-esteem are associated with psychosis and attachment anxiety and may explain behaviour that is focused on conciliation, rather than self-protection.

Reduced brain reward response during cooperation in first-degree relatives of patients with psychosis: an fMRI study.

BACKGROUND: Psychosis is characterized by a profound lack of trust and disturbed social interactions. Investigating the neural basis of these deficits is difficult because of medication effects but first-degree relatives show qualitatively similar abnormalities to patients with psychosis on various tasks. This study aimed to investigate neural activation in siblings of patients in response to an interactive task. We hypothesized that, compared to controls, siblings would show (i) less basic trust at the beginning of the task and (ii) reduced activation of the brain reward and mentalizing systems. METHOD: Functional magnetic resonance imaging (fMRI) data were acquired on 50 healthy siblings of patients with psychosis and 33 healthy controls during a multi-round trust game with a cooperative counterpart. An a priori region-of-interest (ROI) analysis of the caudate, temporoparietal junction (TPJ), superior temporal sulcus (STS), insula and medial prefrontal cortex (mPFC) was performed focusing on the investment and repayment phases. An exploratory whole-brain analysis was run to test for group-wise differences outside these ROIs. RESULTS: The siblings' behaviour during the trust game did not differ significantly from that of the controls. At the neural level, siblings showed reduced activation of the right caudate during investments, and the left insula during repayments. In addition, the whole-brain analysis revealed reduced putamen activation in siblings during investments. CONCLUSIONS: The findings suggest that siblings show aberrant functioning of regions traditionally involved in reward processing in response to cooperation, which may be associated with the social reward deficits observed in psychosis.

Childhood maltreatment, adult attachment and psychotic symptomatology: a study in patients, siblings and controls.

PURPOSE: The association between childhood maltreatment (ChM) and psychotic disorders is well established. However, there is an ongoing debate about which factors account for this relationship. One explanation is that the relationship between ChM and psychosis is mediated by adult attachment style. Therefore, in this study, we aimed to investigate whether adult attachment style mediates the relationship between ChM and positive and negative symptomatology. METHODS: We investigated the relation between ChM and psychotic symptoms, taking into account levels of (insecure) attachment, in 131 patients with psychotic illness, 123 siblings and 72 controls. ChM was assessed with the Childhood Trauma Questionnaire (CTQ). Attachment dimensions of anxiety and avoidance were measured using the Psychosis Attachment Measure (PAM). RESULTS: In both patients and siblings, ChM predicted positive symptoms and this relationship was partly mediated by attachment style. This relationship was found to be stronger for siblings than for patients. ChM predicted negative symptoms in patients and siblings. In the patient sample, attachment style did not mediate the relationship between ChM and negative symptoms, whereas attachment style was found to be a mediator in the sibling sample. CONCLUSIONS: ChM was associated with positive and negative symptomatology in both patients and siblings. Particularly in siblings, the relationship between ChM and psychosis seems to be mediated by adult attachment style. Perhaps attachment style may play a more prominent role on a subclinical level.