Relations between Spontaneous and Explicit Mentalizing in Autism and Typical Development

Background: Recent research on mentalizing—or thinking about others’ thoughts, feelings, and desires—suggests that deliberate mental state reasoning may be dissociable from more spontaneous mentalizing. This possibility is especially relevant to autism spectrum disorder (ASD), where explicit social cognition often outpaces real-world social success. Previous ASD research investigating relations between different types of mentalizing, however, has used a limited set of tasks and produced mixed findings. Quantifying links between explicit and spontaneous mentalizing may illuminate cognitive mechanisms underlying real-world social behaviors in ASD.

Objectives: To evaluate, in both ASD and typical development, whether explicit measures of mental state reasoning are related to the tendency to spontaneously mentalize about naturalistic social stimuli.

Methods: Participants were 26 typical individuals (15 males, average age=19.86 years) and 24 individuals with ASD (17 males, average age=21.70 years) matched on age, sex, and verbal IQ. Participants completed two measures of spontaneous mentalizing: (1) the Spontaneous Theory of Mind Protocol (STOMP; Rice et al., 2014), in which participants described videos depicting complex social scenes and (2) the Describe-a-Friend task (Meins et al., 2008), in which participants described a close friend or family member. Both tasks yielded a measure of spontaneous mental state language (i.e., references to thoughts, feelings, and desires). Further, participants answered explicit questions about the mental states of characters in the videos and control questions about physical events. Participants also completed two conventional explicit mentalizing measures: (1) Mind in the Eyes (Baron-Cohen et al., 2001), involving inferring mental states from photographs of eyes and (2) Strange Stories (White et al., 2009), involving reasoning about mental states from stories.

Results: There were no significant group differences in the amount of spontaneous mentalizing for either the STOMP or Describe-a-Friend task. For the explicit questions about the STOMP videos, however, adults with ASD showed diminished performance when reasoning about mental states, but intact performance when reasoning about physical events. Controlling for age and verbal IQ, performance on those explicit mental state questions was related to the tendency to spontaneously mentalize for participants with ASD (r=.46, p=.048), but not for typical adults (r=-.264, p=.35). Across tasks, levels of spontaneous mentalizing were correlated for typical adults (r=.54, p=.01), but not for adults with ASD (r=-.076, p=.76). For the Mind in the Eyes and Strange Stories, typical adults performed significantly better than those with ASD, and, within each group, the two tasks were significantly correlated. For neither group, however, did the Mind in the Eyes or Strange Stories significantly relate to spontaneous mentalizing.

Conclusions: Overall, the relation between implicit and explicit mentalizing is complex, task-dependent, and varies between typical development and ASD. Intriguingly, only adults with ASD showed a relation between the tendency to offer spontaneous mental state description about naturalistic, socially-complex scenes and explicit measures of accurate mental state reasoning about those same scenes. These findings may suggest different routes to explicit and implicit mentalizing in typical development and ASD, and future research should continue to explore this possibility by including measures of mentalizing during social interaction.