Author Information
JaeHwan Byun, Wichita State University, United StatesMara Alagic, Wichita State University, United States
Abstract
The rapid integration of Generative AI (GenAI) into education has created anomalies in the Kuhnian sense—systematic mismatches that established learning paradigms (Behaviorism, Cognitivism, and Constructivism) cannot fully explain. While these frameworks account for conditioning, information processing, and social negotiation in learning, they do not capture the dynamics of interacting with a semi-autonomous, non-human agent that produces fluent yet decontextualized artifacts. This gap calls for a fundamental shift in how learning is conceptualized. This study proposes Generativism as a distinct learning paradigm for the GenAI era. Drawing on Kuhn’s theory of scientific revolutions, we analyze limitations of existing paradigms and articulate Generativism’s ontological, epistemological, and methodological principles. Ontologically, learning is conceptualized as a distributed enactment within a “Human–AI–Artifact system,” rather than solely an internal or social process. Epistemologically, validation shifts from accuracy or consensus to epistemic reflexivity, in which learners negotiate meaning by situating AI-generated outputs within their own reasoning and relevant alternatives. Methodologically, learning is operationalized through the “interaction episode,” an iterative cycle of prompting, generation, critique, and refinement through which meaning is negotiated rather than merely corrected. Central to this paradigm is the learner as an “Orchestrator” who directs the generative process. Generativism provides a coherent conceptual language to design and evaluate learning experiences beyond tool use, fostering critical thinkers capable of navigating AI-assisted meaning-making. We conclude by outlining implications for educational research and instructional design in GenAI-mediated learning environments.
Comments
Powered by WP LinkPress