A Promise Theory Perspective on the Role of Intent in Group Dynamics

TL;DR

The paper advances a Promise Theory framework to explain how human groups organize into hierarchical scales (Dunbar’s numbers) around seed intents, balancing attention work against contention. By combining dimensional analysis with simple statistical mechanics, it derives probabilistic rules for group sizes and shows consistency with large-scale Wikipedia editing data, including the role of bots. The approach reframes trust as a work-like resource and accounts for the emergence of hub-centric grooming as a scalable, energy-like constraint on coordination. The findings offer a principled link between social cognition, neural processing, and information-theoretic limits, with implications for understanding human-AI collaboration in coordinated social systems.

Abstract

We present a simple argument using Promise Theory and dimensional analysis for the Dunbar scaling hierarchy, supported by recent data from group formation in Wikipedia editing. We show how the assumption of a common priority seeds group alignment until the costs associated with attending to the group outweigh the benefits in a detailed balance scenario. Subject to partial efficiency of implementing promised intentions, we can reproduce a series of compatible rates that balance growth with entropy.

Figures (4)

• Figure 1: Groups form either because agents come together independently attracted to contribute to a common cause (like fighting a common enemy or working on a common product), or they form emergent clusters by pairwise percolation of promise relationships. In our model, we assume the left hand picture of attraction in which 'mistrust' of the central 'seed' promise drives increased attention and potentially proximity as a secondary effect.
• Figure 2: A basic cooperation/calibration triangle in Promise Theory allows two agents to work together on behalf of a third, or allows a third to act as a seed effectively bringing them into alignment $X=Y$. From Promise Theory, one would expect opportunistic dyadic structures $N=2$ for compositional or symbiotic specialization, with more important coordinated structures built from equilibrated/cross-checked triads $N=3$.
• Figure 3: Curve fit of data using the formula in equation \ref{['formula']}. The crosses approximate error uncertainty. The model fit is expected to be worst for small $n$ due to integer effects.
• Figure 4: The group equilibrium law plotted for $\langle N\rangle_{\overline T} = 4,8,30,150$ illustrating the flattening of group probability curves with increasing number. The amplitude gives an approximate magnitude for the attention power rate required to maintain each level.