The paper develops a framework that explores structured continuity in interactions with AI systems, suggesting nuanced understanding of model behaviors.
Key Points
This work aims to establish a framework to analyze continuity in interactions with stateless large language models.
Developed Recursive Constraint Configuration (RCC) to analyze interactional continuity.
Conducted longitudinal qualitative research on a human–AI archive with one participant.
Compared various conditions and explanations within episode-level matrices.
Demonstrated that continuity does not rely on persistent internal agents but rather on structured interactions.
Identified seven configuration movements to differentiate types of continuity and constrained interactions.
Highlighted that prior structure's availability is variable rather than binary.
Cite This Study
Herlon Heros Vermeuln (2026) studied this question.