A Physiology-First, Execution-Efficiency Framework for Psychiatric Disorders: Retraining Intact Learning Trajectories

Current psychiatric diagnosis and intervention rely on a reductionist paradigm: symptoms are abstracted into static categories (DSM, ICD), and features such as neurotransmitter abnormalities, cognitive distortions, or sleep disruption are treated as independent causes. This reductionist approach overlooks the temporal dynamics by which these variables accumulate, interact, and reinforce one another, thereby obscuring mechanisms and constraining interventions. We propose a paradigm shift toward dynamic system psychiatry. In this model, psychiatric symptoms are reconceptualized as emergent outcomes of intact learning mechanisms operating under maladaptive stress trajectories. Two computational modules provide the foundation: the Learning Resonance Index (LRI), a dimensional metric that integrates physiological regulation, behavioral execution efficiency, and cognitive-emotional variability; and the stress function W(t), a recursive model capturing cumulative stress exposure, recovery slope, and modulators such as circadian rhythm and environmental load. Together, they enable continuous, interpretable assessment of system state rather than categorical labeling. Intervention follows a physiology-first, efficiency-centered sequence: rapid active responses to stress deviations, auxiliary supports as needed, repetition-based consolidation, and quantitative efficiency monitoring (latency, adherence, repetition, generalization). Crucially, the framework is compatible with existing therapies, providing a unifying scaffold for CBT, behavioral activation, pharmacological treatment, and digital health tools. By reframing psychiatry from extractive classification to dynamic iterative modeling, this approach enhances mechanistic interpretability, improves alignment of intervention with system state, and lays the foundation for precision psychiatry that is both explanatory and adaptable.