Answers are grounded in this paper and cite their sources.
Digital mental health has become increasingly precise about the individual and the moment. It is less explicit about the relationship through which support is requested, offered, refused, delayed, or redistributed. This paper argues that a distinct methodological layer is needed for such decisions: relational infrastructure. We define relational infrastructure as a bounded digital architecture in which one or more relationships are formalised as explicit decision-bearing structures, with support permissibility, role asymmetry, burden constraints, recency, uncertainty, and admissible output classes governed by inspectable rules. Building on REMA I and REMA II, we distinguish this layer from communication tools, care coordination, conversational agents, generic multi-user systems, and dyadic or social-support interventions that do not formalise the relationship itself. We then describe six design requirements for relational infrastructure: explicit relational state, role asymmetry, burden-aware admissibility, temporal validity, bounded outputs, and glass-box decision logic. A worked example shows how an unchanged primary-user signal can lead to different support decisions when supporter capacity, permissions, or data validity change. We argue that this architecture creates a tractable object for empirical study rather than treating relational context as an informal modifier of individual-level intervention. The paper closes with a falsifiable research programme spanning construct validation, burden modelling, ecological robustness, selector calibration, comparative transparency, and implementation. REMA III is a conceptual and methodological proposal, not an efficacy claim. Its value depends on whether explicit relational state and bounded decision rules improve explanation, safety, and coordination beyond individual-level models.
No takes yet. Share an insight, caveat, or question.
Fisher et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: