Syncing Minds and Machines: Hybrid Cognitive Alignment as an Emergent Coordination Mechanism in Human–AI Collaboration
提出混合认知对齐理论，解释人类与AI如何在协作中动态协调任务与角色，通过四种对齐路径实现功能兼容，为组织设计人机协作提供新视角。
As humans and AI increasingly work together in organizations, how can they dynamically allocate tasks and roles amid evolving task demands? Humans and AI represent the world in distinct yet complementary ways, creating both performance opportunities and coordination challenges for human–AI collaboration (HAIC). Tackling this, we advance a novel theory of “hybrid cognitive alignment” (HCA) as an emergent coordination mechanism that explains microprocesses leading to a functional compatibility between human and AI, enabling both parties to anticipate and adapt to each other. We apply the taskwork–teamwork framework to explain what needs to align, and create an AI-typology to elucidate how HCA can be achieved. We delineate how humans collaborate with “tool-like,” “assistant-like,” “rigid-teammate-like,” and “teammate-like” AI through four distinct pathways to develop instrumental, contextualized, prescribed, and reciprocal alignments. Our theory complements the current top-down organization design approach with a bottom-up, emergent perspective. We highlight AI’s material properties as a distinctive driver of emergent coordination in HAIC, in parallel to human–AI’s iterative exchanges. Our work creates a new theoretical frontier for the coordination literature, helps to synthesize mixed findings regarding HAIC effectiveness, and generates implications for designing and deploying AI as a collaborator.