The Dopamine-Driven Discounting Model (D3M ): A Conceptual Heuristic Framework for Digital Wellness Intervention Design

Infinite-scroll architectures and gamified application loops leverage variable reward schedules to maximize user retention, often prompting concerns regarding compulsive digital engagement. This paper introduces the Dopamine-Driven Discounting Model (D3M), an exploratory conceptual synthesis and hypothesis-generating framework that integrates behavioral proxies inspired by neurobiological reward prediction error with behavioral economic temporal discounting principles. We present a toy formalization modeling the latent probability of immediate session continuation (Pscroll). To map this heuristic output bounded within a mathematically valid probability space, we apply a standard logistic function sigmoid transform augmented by an uncalibrated baseline bias parameter (b). We provide a conceptual system architecture illustrating how such a framework might hypothetically inform on-device software intervention design by introducing adaptive user interface friction (D). Finally, we outline structural limitations and discuss prospective advanced methodologies—including survival analysis, mixed-effects logistic models, and hierarchical Bayesian estimation—for future empirical validation.