Dual-Route Processing of Social Interactions: Dissociating Domain-General and Domain-Specific Mechanisms in Visual Search

Efficiently perceiving social interactions is fundamental to human social cognition, yet the underlying mechanisms remain contested. Domain-specific theories propose specialized processing for social content, while domain-general accounts attribute perceptual advantages to low-level visual features. This debate has been critically hampered by an overreliance on static stimuli, which obscure crucial temporal dynamics and introduce configural confounds. The present study addressed these limitations using a dynamic visual search paradigm that differentiated physical-contingency interactions (defined by synchronized, perceptually coupled movements) from communicative-contingency interactions (defined by sequential, meaning-based exchanges). Experiment 1 revealed a sharp dissociation: physical-contingency interactions were detected more efficiently than both non-interactive dyads and communicative-contingency interactions, while communicative-contingency interactions showed no search advantages. Experiment 2 employed eye-tracking to elucidate underlying mechanisms. Physical-contingency interaction search showed overall search advantages through enhanced target recognition and efficient distractor rejection, despite equivalent attentional guidance compared to non-interaction search. Communicative-contingency interaction search exhibited enhanced target recognition and efficient distractor rejection but suffered from attentional guidance deficits that eliminated overall search benefits. Additionally, communicative-contingency interactions as distractors required the longest fixation durations, indicating strongest attentional stickiness. These findings challenge existing theoretical frameworks by demonstrating that social interaction search advantages cannot be explained by either domain-specific or domain-general mechanisms alone. We propose a dual-route processing model wherein physical-contingency interactions benefit primarily from domain-general perceptual mechanisms that efficiently process motion synchrony and gestalt properties, while communicative-contingency interactions rely on domain-specific social processing mechanisms that handle complex interpersonal meaning. This integrated framework reconciles competing theoretical perspectives and advances understanding of how different types of social information are prioritized in dynamic visual environments.