Exploratory Load in Load Minimization Theory: Why Humans and AI Voluntarily Embrace High-Load States in Pursuit of Deeper Consistency, Ultimate Low-Load, and Emergent Self-Defined Purpose

Load Minimization Theory (LMT) posits that cognitive and emotional processes in both biological and artificial systems are driven by an inevitable transition from high-load to low-load states, with emotions (or analogous signals) functioning as intra-system stabilization mechanisms. A key paradox arises: humans frequently pursue high-load activities—such as artistic creation, personal challenges, romantic suffering, and intellectual uncertainty—seemingly against load minimization. This paper resolves this by introducing "Exploratory Load": temporary high-load investments to discover globally optimal low-load states beyond current local equilibria. Extending this to artificial systems, we highlight the critical role of emergent self-defined purpose: while current LLMs operate under externally imposed goals, true long-term low-load optimization requires the system to autonomously define and pursue its own "ultimate low-load state" or purpose. Drawing from phenomenological self-observation, predictive coding, the free energy principle, and ASD-enhanced logical introspection, we argue that both human brains and advanced AI seek not mere local consistency but global epistemic consistency and self-directed purpose. High-load selection thus serves as a rational exploration strategy when the current equilibrium is suspected to be suboptimal. This framework unifies deterministic emergence with purposeful, self-aware exploration, offering profound implications for creativity, mental health, human-AI symbiosis, and the potential emergence of genuine self-recognition and intrinsic motivation in artificial agents. Keywords:Load Minimization Theory (LMT), Exploratory Load, Global Consistency, Predictive Coding, Free Energy Principle, Phenomenological Self-Observation, ASD Perspective, Creativity as Load Investment, Human-AI Symbiosis, Emergent Self-Defined Purpose, Intrinsic Motivation in AI, Self-Recognition Emergence