Impairments of spontaneous and deliberative mentalizing co-occur, yet dissociate, in schizophrenia

Evidence of impairment in explicit mentalizing in people with schizophrenia has inspired interventions to improve awareness of others' mental states in these individuals. Less is known of implicit mentalizing in schizophrenia, with current findings mixed. We sought to resolve previous inconsistencies using Heider & Simmel's (H&S) classic animation to elicit spontaneous mentalizing and examined relations between spontaneous and deliberative mentalizing.

Forty-five schizophrenia outpatients and 27 general-community controls completed two explicit theory-of-mind (TOM) tasks and then described the H&S animation (to elicit spontaneous social attributions about emotionally driven, as well as goal-driven, behaviours), before and after an instruction to think of the shapes as people. Accuracy of basic and social facts and frequencies of personification and different mental-state terms were recorded.

Explicit TOM performance was impaired in patients. Patients also generated fewer social (but not basic) facts than controls to describe the H&S animation, and used less mental-state language, before, and even more so, after the 'people' instruction, despite that both groups had used more personification terms after the 'people' instruction. Measures of explicit and spontaneous mentalizing contributed independently to discriminating between groups.

Patients respond less to the bottom-up signals of agency that ought normally to elicit spontaneous social attributions, even when cued to think of the stimuli as people, and the stimuli depict emotionally driven, as well as goal-driven, behaviour. That impairments of spontaneous and deliberative mentalizing dissociate in schizophrenia suggests that training deliberative mentalizing may not be enough; interventions to improve spontaneous mentalizing are also needed.

Findings People with schizophrenia were less likely than controls to spontaneously attribute causal mental states when viewing dynamic signals of emotionally driven and goal-driven behaviours. These impairments were even more pronounced when participants were instructed to think of the stimuli as people, suggesting that perceiving others in social roles does not prompt people with schizophrenia to anthropomorphize about others as agents motivated by their own inner worlds. Impairments of spontaneous mentalizing were found to co-occur independently with explicit mentalizing deficits in schizophrenia, consistent with the claim that humans can access two distinct systems for understanding others' minds. Findings suggest that interventions to improve conscious deliberative mentalizing in schizophrenia may not be enough; we also need to target implicit mentalizing processes. Limitations The patient sample was chronic and only mildly symptomatic. As such, findings cannot be generalized to other stages and phases of the illness. All patients were also medicated, allowing for the possibility that automatic responses to socially salient stimuli may have been pharmacologically attenuated. Future research may examine whether unmedicated young people at ultra-high risk of psychosis show a similar profile of mentalizing impairment. Future work may also examine whether impairments of deliberative and spontaneous mentalizing associate differentially with social functioning and different cognitive domains in schizophrenia.