Behavioural Friction Theory: Toward a Common Currency for Behavioural Science

Abstract

Behavioural science has produced robust findings across dozens of research traditions, yet lacks the integrative architecture needed to connect them — a common unit or map for relating constructs across domains and anticipating cross-field interactions. Behavioural Friction Theory (BFT) proposes friction — the cost a regulatory system assigns to a potential action in a given situation — as a candidate common currency for behavioural science, a programmatic proposal whose status is contingent on independent field-level operationalisation. It offers a candidate decomposition of that cost into four fields (Safety, Meaning, Ability, Effort) across five layers (Biological, Emotional, Inner, Cognitive, External), with resolution modelled as a pressure-dependent race. BFT adds no new mechanism: its components are inherited and cited at the point of use (accumulation-to-threshold, expected value of control, hierarchical control, ordered appraisal). The contribution is threefold: (1) a field × layer map on which traditions such as polyvagal theory, self-efficacy research, cognitive load theory, psychological safety, and nudge theory are located as candidate descriptions of regions of a proposed shared landscape; (2) design heuristics motivated by the decomposition (baseline pressure as an intervention precondition, weakest-field prioritisation, and a bounded-compensation prediction); and (3) thirty formal, individually-testable propositions, each with a minimal test design, framed as a pre-registerable programme for testing whether the four fields are functionally dissociable. The central requirement is stated plainly: field-level friction has no validated measurement instrument yet, so several propositions are not presently runnable — they are tests to be run, not results in hand. The decomposition is falsified if the four fields do not dissociate under manipulation, or if a unitary-cost model — expected value of control, active inference, or a resource-rational model that collapses to a single cost — accounts for the same variance. BFT is offered as an architecture within which existing theories can be compared, connected, and extended, not as a replacement for them. v15 (August 2026) — positioning update. The framework is situated relative to Resource-Rational Analysis (Lieder & Griffiths, 2020), and the common currency is stated as a candidate, programmatic account rather than a substrate-universal law; the architecture and propositions are unchanged in substance. Earlier versions remain accessible through the version history. v16 (August 2026) — positioning and attribution revision. The four-field taxonomy is now positioned against the health-behaviour frameworks a reader would actually test it against — the Health Belief Model, Protection Motivation Theory and the Theory of Planned Behaviour — and the distinction drawn is structural rather than numerical. Those frameworks enter every factor with one fitted monotone form; the claim here is that the four differ in form and that the differences are stated in advance: Safety and Effort absence-structured with the zero point as the optimum, Meaning and Ability appetitive, Ability alone carrying an intrinsic inverted-U, and Effort the aggregate load of running the other three rather than a fourth sibling. Subjective norm is decomposed as a worked case, with a falsifier and a single stated placement for each component, and the corollary that it is consistently the theory's weakest predictor now carries its source: Armitage and Conner's (2001) meta-analysis of 185 studies, whose own diagnosis — poor measurement and the need for expansion of the normative component — is the expansion this decomposition supplies a structure for. The felt side of the Effort field is credited to Kurzban, Duckworth, Kable and Myers's (2013) opportunity-cost model of subjective effort, with the difference stated: theirs is the value of the next-best use the recruited machinery is denied, and the field proposed here is the summed load of the other three fields' races, present whether or not it is felt. Earlier versions remain in the version history.