Research

Independent research publications on human behavior, cognitive architecture, and deterministic governance in complex human and AI systems.

Independent research publications by Didomi Research.

---

Publications

The Agentic Inhibitory Governance Standard (A.I.G.S.)

A Control-Layer Blueprint for Top-Down Executive Governance in Autonomous AI Systems
May 2026

This paper presents that today’s AI agents can be very intelligent but still lack the behavioral stability needed to act responsibly over time. In behavioral science terms, they have capability without executive control: they can reason, generate language, use tools, and follow instructions, but they do not have a persistent sense of identity, boundaries, priorities, or decision principles. A.I.G.S. proposes giving AI agents a “digital prefrontal cortex,” inspired by the human brain’s executive control system, so the agent can consistently answer four basic questions: who am I, what matters to me, how do I decide when values conflict, and how should I present myself. Instead of relying only on prompts that fade or can be manipulated, A.I.G.S. defines these answers in a structured identity profile that is checked at runtime, helping agents remain coherent, auditable, and aligned with their intended role even during long or adversarial interactions.
→ View paper

Simulative-Elasticity Model (SEM)

A Systems Theory of Intelligence: Navigating the Space Between Habit and Adaptation
February 2026

This paper presents that intelligence is not mainly about how much a person or AI system knows, but about how well it can regulate its own behavior when the world changes. In behavioral science terms, SEM describes intelligence as the ability to move between efficient habits and deeper adaptive reasoning at the right moment. Most of the time, humans and machines rely on learned scripts because they save energy, but when reality no longer matches expectation, the system needs a kind of internal “clutch” that detects the mismatch, interrupts autopilot, and re-engages flexible reasoning. SEM calls this mismatch the Delta, the gap between what the system predicted and what actually happened. A truly intelligent system is not one that always uses deep reasoning, and not one that blindly repeats past habits, but one that knows when to stay efficient, when to adapt, when to test new explanations, and when to turn successful adaptations into new automatic skills.
→ View paper

Didomi Behavioral Model (DBM)

A Systems Biology Framework for Neurobehavioral Dynamics and Allostatic Regulation
February 2026

The Didomi Behavioral Model presents that human behavior is not driven by one simple force like “motivation,” but by a set of internal regulatory systems trying to keep the person stable, safe, energized, connected, and functional. In behavioral science terms, DBM describes the mind as an inner control panel made of different systems that manage energy, recovery, attention, status, and connection. When these systems are balanced, people can think clearly, engage with others, pursue goals, and adapt to challenges. But when one system is overloaded, depleted, threatened, or ignored, behavior can shift in ways that look like laziness, anxiety, avoidance, defensiveness, impulsivity, or low motivation. DBM reframes these behaviors not as character flaws, but as signals from internal systems trying to protect the person or restore balance. The goal is to help people understand what system is driving their behavior, regulate it more consciously, and make better choices without shame, force, or relying only on willpower.
→ View paper