关键词: 人工智能, 应急决策与处置, 经验阻断效, 替代式 AI 使用, 人智协同

Abstract: Artificial intelligence (AI) is increasingly embedded in key stages of emergency decision-making and response, including information assessment, situation evaluation, plan generation, action coordination, and after-action reflection. Although AI can broaden information coverage, improve situational integration, and support complex decisions, emergency managers’ experience-based judgment is not an automatic by-product of task completion. It develops through first-hand exposure to raw situational cues, independent judgment generation, calibration against action consequences, and reflective transformation. When AI shifts from supporting managers to substituting for these activities, current task performance may become decoupled from subsequent experiential growth. We introduce the artificial intelligence experience-blocking effect (EBE), defined as a proximal, multidimensional task-process state in which AI assistance compresses, displaces, weakens, or renders superficial managers’ exposure to raw situations, generation of judgments, ownership of consequences and feedback-based calibration, and reflective transformation. We develop an integrated framework linking substitutive AI use, EBE, experiential learning processes, and experience consolidation across three interrelated studies. Study 1 clarifies the meaning and boundaries of EBE, identifies four formative dimensions, and develops a measurement instrument. Study 2 operationalizes substitutive AI use as a human–AI–task configuration, tests whether it increases the four EBE dimensions, and treats current task performance as a parallel comparison outcome. Study 3 uses longitudinal, repeated, high-fidelity simulations to test whether EBE contributes to deficits in individual experience internalization by constraining situational-cue perceptual learning and action–outcome associative learning, and to deficits in team experience externalization by reducing the experiential material that members can articulate and by limiting its transformation through team learning. The framework extends the evaluation of AI-assisted emergency management beyond immediate task performance to experiential learning and the consolidation of individual and team experience, offering theoretical and practical guidance for human–AI collaboration that preserves both task effectiveness and the development of managers’ experience-based expertise.

Key words: artificial intelligence, emergency and crisis management, experience-blocking effect, substitutive AI use, human-AI collaboration