Neural responses to AIGC based on the associative propositional evaluation model
Dong Lv, Rui Sun, Shukun Qin, Qiuhua Zhu, Yue Cheng · Scientific Reports · 2026
Artificial Intelligence-Generated Content (AIGC) can improve production efficiency and reduce creation costs, but consumer evaluations of AIGC applications remain mixed. Based on the Associative-Propositional Evaluation (APE) model, we used event-related potentials (ERPs) to examine consumers’ automatic neural responses and behavioral judgments of AIGC applications across different levels of emotional engagement.Pairing positive adjectives with AIGC content elicited more negative N2 amplitudes, which reflects early cognitive conflict or mismatch during associative-dominant processing. In high-emotional-engagement scenarios, negative adjectives triggered more negative N2 amplitudes than in low-emotional conditions, showing that early evaluative processing is modulated by the application context.Behaviorally, acceptance rates for negative adjectives were higher than for positive adjectives in low-emotional scenarios, meaning implicit responses differ across emotional contexts.These findings offer preliminary neurophysiological evidence for AIGC evaluation, but the results should be interpreted with caution. The N2 component primarily indexes early cognitive conflict, and is not a process-specific marker of definitive attitude valence. The study uses a small, non-probability volunteer sample, which limits the generalizability of the findings. Future research needs to replicate these effects with larger, more representative populations to validate the results.