Toward shared situational awareness in human-AI hybrid intelligence: A survey in high-risk aviation environments

As Artificial Intelligence (AI) becomes increasingly embedded in high-risk aviation operations, deeper and more effective hybrid intelligence team collaboration is required to sustain safety and operational performance. Shared Situational Awareness (SSA) has emerged as a key concept for achieving aligned and compatible situational representations between human operators and AI systems. However, the mechanisms through which SSA forms and adapts remain insufficiently understood. This study systematically reviews research from the past decade to synthesize evidence on how SSA emerges and evolves within hybrid human-AI intelligence systems. Across the dimensions of information exchange, interaction processes, and influencing factors,the analysis indicates that SSA is established through cognitive alignment and subsequently evolves through interconnected processes of divergence, negotiation, and repair. Building on this synthesis, an explanatory framework for human-AI SSA formation is proposed, highlighting how adaptive coordination processes, mutual prediction, and shared understanding shape the maintenance of robust and resilient collaboration. This work enriches existing perspectives on SSA and frames safety as an emergent and resilient property of hybrid human-AI cognition. It provides an integrative conceptual basis to inform the design, evaluation, and safety management of human-AI systems supporting coordinated action in dynamic aviation contexts.