自动观点采择中内隐心智化和潜心智化的分离

utomatic perspective taking, a crucial cognitive process in social interactions, refers to the spontaneous ability to compute others viewpoints. Two primary theoretical explanations currently exist for this phenomenon: the implicit mentalizing account and the submentalizing account. The implicit mentalizing account posits that automatic perspective taking is a domain-specific cognitive ability unique to humans, involving the rapid and unconscious attribution of mental states to others. In contrast, the submentalizing theory suggests that this process stems from domain-general attentional orienting mechanisms, which do not necessarily involve mental state attribution. The present study aimed to investigate whether these two mechanisms operate independently or collaboratively in driving automatic perspective taking. To address this question, a series of experiments were conducted, including two behavioral experiments and one event-related potential (ERP) experiment. These experiments employed the dot perspective taking task, a widely used paradigm in the field. The study manipulated two key factors: social factors and attentional load. In Experiment 1, the social relevance of the central stimulus was manipulated (avatar vs. arrow). In Experiments 2 and 3, the sociality of prompts was varied. Across all experiments, attentional load was manipulated to examine its impact on automatic perspective taking. Results from the behavioral experiments indicate that implicit mentalizing and submentalizing operate in parallel during automatic perspective taking. Experiment 1 revealed only a main effect of attentional load, with no significant effect of the social relevance of the central stimulus. This finding initially supports the submentalizing account. However, Experiments 2 and 3 demonstrated that both social prompts and attentional load enhanced the automatic perspective taking. Importantly, there was no interaction between these factors, suggesting that they may influence perspective taking through independent mechanisms. The ERP experiment (Experiment 3) provided further insights into the neural correlates of these processes. Regarding to TP450 component, which is associated with perspective computation, showed an effect of prompt sociality. This finding supports the implicit mentalizing account, suggesting that social information is processed rapidly and automatically. Additionally, interactions between attentional load and congruency were observed, indicating the impact of cognitive resources on perspective computation. This result lends support to the submentalizing account, highlighting the role of domain-general attentional processes. In conclusion, both behavioral and ERP results suggest that implicit mentalizing and submentalizing mechanisms jointly drive automatic perspective taking while operating independently. This finding offers a novel perspective on understanding the complex cognitive processes underlying perspective taking. It suggests that rather than being mutually exclusive, these two mechanisms may complement each other in facilitating social cognition. The study provides further empirical evidence to address the ongoing debate between implicit mentalizing and submentalizing explanations, proposing an integrated view of automatic perspective taking. Future research could further explore the specific conditions under which each mechanism predominates and how they interact in more complex social scenarios.