Simulating the Unreal: Neural Mechanisms of Counterfactual Comprehension

Counterfactual reasoning-the ability to construct scenarios that contradict known facts-is central to complex human cognition. However, little is known about how such reasoning unfolds in languages like Mandarin Chinese, which lack explicit morphological markers for counterfactuality. In this study, we used a discourse-based manipulation with tightly controlled syntactic and lexical features to examine the neural mechanisms of Mandarin counterfactual comprehension via functional near-infrared spectroscopy (fNIRS). Activation and graph-theoretical analyses revealed that counterfactuals, compared to hypothetical and causal statements, elicited heightened activity and enhanced network efficiency in sensorimotor regions (PreCG, PoCG), alongside increased engagement of the left inferior frontal gyrus (IFG). Inter-subject representational similarity analysis (IS-RSA) further demonstrated that these effects were modulated by individual differences in working memory and Theory of Mind, with WM supporting dual representation and ToM facilitating higher-level inferencing. Together, these findings provide converging evidence for the Mental Simulation Hypothesis: counterfactual comprehension recruits embodied sensorimotor processes to simulate unreal events, particularly in the absence of overt grammatical cues. By integrating individual-difference measures with network-level analyses, this study advances our understanding of the neural architecture of counterfactual thinking and raises questions about how grammatical marking may shape counterfactual processing across languages.