Personal Reinforcement Interval (PRI): A Theoretical Framework for Personalized Reinforcement Timing

This conceptual paper introduces the Personal Reinforcement Interval (PRI), a proposed latent behavioral parameter representing the estimated interval during which an individual can sustain a goal-directed behavior in a given context before external reinforcement becomes optimally beneficial. PRI is not a reinforcement schedule, a personality trait, a general measure of motivation, or an estimation algorithm — it is modeled as a dynamic function of person, behavior, context, and time, estimated from pre-intervention behavioral persistence rather than from the outcomes of any specific intervention. The paper formally separates the latent PRI construct from the intervention-timing decision rule, states causal constraints, gives operational criteria for behavioral episodes and for the class of persistence-capacity signals admissible as inputs, distinguishes PRI from the three adjacent model families most likely to be proposed as substitutes (hazard models, JITAI receptivity, and RL-derived policies), and states testable predictions and explicit falsification conditions. Status: theoretical preprint. The construct is presented as a falsifiable research proposal, not as an established or empirically validated finding. This version (v1.4) constrains the persistence-capacity signal to a checkable class of admissible signals, adds a worked derivation supporting the model's discriminant-validity claim, and condenses the research agenda into five gated stages.