To Be FAIR: Theory Specification Needs an Update

Open science innovations have focused on rigorous theory testing, yet methods for specifying, sharing, and iteratively improving theories remain underdeveloped. To address this limitation, we introduce <i>FAIR theory</i>, a standard for specifying theories as findable, accessible, interoperable, and reusable digital objects. FAIR theories are findable in well-established archives; accessible in terms of their availability and ability to be understood; interoperable for specific purposes, such as selecting control variables; and reusable in that they can be iteratively and collaboratively improved on. This article adapts the FAIR principles for theory; reflects on current FAIR practices in relation to psychological theory; and discusses FAIR theories' potential impact in terms of reducing research waste, enabling metaresearch on theories' structure and development, and incorporating theory into reproducible research workflows-from hypothesis generation to simulation studies. We present a conceptual workflow for FAIRifying theory that builds on existing open science principles and infrastructures. More detailed tutorials, worked examples, and convenience functions to automate this workflow are available in the <i>theorytools</i> R package. FAIR theory constitutes a structured protocol for archiving, communicating about, and iteratively improving theory, addressing a critical gap in open scholarly practices and potentially increasing the efficiency of cumulative knowledge acquisition in psychology and beyond.