Abstract

Organizations increasingly think, decide, remember, and act through evolving networks of persistent representations: strategies, decisions, policies, plans, models, work products, AI-generated recommendations, operational traces, outcomes, and organizational memory. Existing theories explain important aspects of organizational cognition, including learning, distributed cognition, sensemaking, memory, information processing, and feedback. Yet they do not explicitly explain how organizations preserve meaningful semantic and causal relationships across continuously evolving organizational representation networks. This paper develops Correspondence Theory, a representation-network-centered theory of organizational cognition. The theory introduces a representation-network ontology, defines Organizational Correspondence as the emergent organizational property by which meaningful semantic and causal relationships among organizational representations are preserved over time, defines Cognitive Integrity as the dynamic organizational condition associated with sustained Organizational Correspondence, and introduces the Correspondence Loop as the recursive, event-driven network maintenance mechanism through which representation networks are maintained, revised, restored, or degraded. The paper presents a proposition architecture and research agenda for future empirical work while preserving a clear distinction between theory, observable traces, indicators, variables, measurement, and validation. The central contribution is a conceptual architecture for studying whether organizational cognition remains meaningfully and causally connected as organizations become more distributed, AI-mediated, and representationally dense.