Operationalising the arts exposome.

A recent arts exposome framework identified 50 transdiagnostic causal mechanisms (emotional, cognitive, self-referential, arousal-stress, inflammatory-epigenetic, social, and behavioural) by which arts engagement may affect mental health, organised around patterned daily exposure to active ingredients across diverse modes of arts engagement. Translating this structure into testable predictions depends on two operational prerequisites that current research has under-addressed. First, the active ingredients must be decomposed into mechanism-aligned exposure dimensions: aggregating activities as varied as receptive consumption, active participation, and cultural attendance under "arts engagement" averages mechanistically different effects and yields findings that the framework cannot localise to specific processes. Second, mechanism tractability is unevenly distributed across measurement modalities. Stress-HPA-immune mechanisms are well-served by peripheral measurement, including transcriptomic markers such as the Conserved Transcriptional Response to Adversity. Aesthetic and reward mechanisms are tractable through central measurement and autonomic indices of aesthetic emotion, but not through peripheral biomarkers. Affiliative mechanisms face validity challenges in peripheral measurement that may require pharmacological probes or behavioural proxies. Neither prerequisite suffices alone: a valid biomarker linked to a heterogeneous exposure measures something the framework cannot interpret, and a well-specified exposure linked to an inappropriate modality deteriorates evidence quality. These refinements are best understood as minimum conditions for identifying the core ingredients that matter most for each mechanism, since individual and contextual variation continue to moderate activation in real-world settings. Together, they identify where mechanism research can deliver near-term advances, where alternative methods are needed, and how the framework can move from theoretical mapping toward ecologically grounded testable inference.