The Emergence of a Theoretical Framework: Research Dialogue with ChatGPT underlying the development of the paper "Why human-LLM interaction works -- and why it often doesn't"

Abstract

This dialogue accompanies the paper "On the Distinction Between Interpretation and Formal Continuity: Why Human–LLM Interaction Works—and Why It Often Doesn't". Its significance lies in documenting the process of enactive conceptual analysis through which the paper was developed because this process reveals, in part, how the underlying theoretical framework emerged, progressively stabilized and became operationally available for interpretation of human-LLM interaction. Unlike the finished paper, which necessarily presents a coherent and unified argument, the dialogue reveals the exploratory movement through which distinctions were proposed, tested, revised, and integrated into a framework capable of explaining a broad range of phenomena associated with human–LLM interaction. Because the dialogue is both about human–LLM interaction and an instance of it, it also provides a concrete example of many of the processes analyzed in the accompanying paper. As such, it is intended as a companion research document in its own right and forms one stage within a broader programme investigating the role of relational determination and semiotic organization across multiple domains.

Keywords