The Theory of Behavior - A Fundamental Comprehensive Algorithm for Human Behavior

This paper proposes a fundamental comprehensive algorithm for human behavior with four distinctive properties: it is invariant across all individuals regardless of physiological, genetic, or cultural variation; it is recursive, treating the decision to consider a decision as itself a behavioral computation; it incorporates Bayesian updating, continuously modifying values through experience; and it enables quantification of previously unquantifiable behavioral states as computable conditions of the algorithm's elements. The model treats behavioral options as sets of perceived, value-laden attributes computed from three inherent factors — Rank, Saturation, and Contingency — whose combined values determine which behavior will be enacted. It differs from Subjective Expected Utility Theory (SEU) by operating at the attribute level with dynamic satiation and extends Multi-Attribute Utility Theory (MAUT) by assigning each attribute its own probability, directional force, and satiation process. It accommodates heuristic constructs in the development of attributes and their pairing to behavioral options. The model precisely specifies the source of many common concepts, such as underachievement, the unrealistic person, "The Boy Who Cried Wolf," and the spoiled child, while enabling quantification of previously unquantifiable concepts: Satisfaction, anxiety, fear, panic, despair, and happiness are defined not as subjective reports but as computable conditions of the algorithm's elements — each with a specific, testable structural definition. The paper includes a formal algorithm specification, mathematical formulation, and working Python implementation demonstrating the computational tractability of the model. A live interactive demonstration of the algorithm is publicly available. Submitted to SJDM 2026 and DSI 2026 Annual Conferences. Critique anticipated and invited.