Modeling relational state
and behavioral consequences
P/G is a computational framework for representing entity properties, directed relationships, and the affective consequences of change.
Our research examines how these representations can support behavioral modeling in AI agents and multi-agent systems.
The framework
Properties, relationships, and the consequences of change
P/G makes the represented state explicit: what changed, whose properties are involved, and how relationships contribute to modeled affect.
P — Entity property state
P represents an entity’s properties, such as health, resources, capability, and reputation. The declared schema specifies which properties an experiment represents and how they change.
G — Directed relationships and group contexts
G includes directed relationships between entities and explicitly defined group-level relational fields. Group-level G is distinct from giving an abstract group its own P-vector.
The current affective baseline uses member identification and descriptive cohesion/continuity summaries. The valuation framework can separately assign decision weights to changes in group-level G.
Primary affect and derived emotions
In the P/G model, effective property changes generate positive or negative primary contributions associated with particular objects. The proposed derived-emotion layer uses personality, current state, and experience to interpret these contributions and associate them with candidate responses.
Current research
Computational cycle complete; manuscripts in closure
The current computational manuscript cycle is complete: Stage A, B0 engineering, and the first registered Computational Theory Test (CTT) program have closed.
The CTT cycle used prediction registers frozen before literature contact, discriminating model comparisons, and bounded literature correspondence.
One registered challenge remains a documented limitation of the frozen mechanism set. Stage B human-judgment validation is prepared but has not started. Ethics filing and external registration remain outstanding; no recruitment or human data collection has occurred. The manuscripts are now in closure and integration.
Working manuscripts
The theory and its testable predictions
Three companion manuscripts cover the P/G framework, its primary affective core, and relational valuation. The current editions present models and experimental proposals; they report no empirical validation.
Experimental platform
A current core for controlled studies
P/G Engine is an experimental implementation of specified P/G core mechanisms. Its development and verification support the research program.
Games are the first selected domain for future controlled agent experiments. The completed computational manuscript cycle is separate from the prepared Stage B human protocol, which uses structured vignettes and has not started.
Demo coming soon.
Explore the experimental platformResearch applications
Directions for subsequent research cycles
Economics and social dynamics
Research directions include heterogeneous relational states, repeated interaction, relational capital, networks, and institutions. These connections motivate future formal models and evaluation.
Explore the economics research directionHuman–robot interaction
A future direction asks whether explicit relational state can support persistent interaction. Perception, uncertainty, user control, and real-world safety require separate work.
Explore the robot research directionAbout
An independent research and product effort
P/G Dynamics develops P/G Theory and experimental agent systems, connecting explicit state modeling with questions about relationships and behavior.
Contact
Discuss the research
Get in touch about research collaboration, manuscript access, or the experimental platform.