Fear-Driven Collective Topology: Comparing Smart-Boid Vietoris–Rips Graphs to Animal Communication Networks via Persistent Features

Guilherme S. Y. Giardini, Carlo R. da Cunha

Keywords: Smart-Boids, Animal communication networks, Collective behavior, Vietoris–Rips complexes, Persistent homology, Topological data analysis, Nearest-neighbor interactions, Evolutionary neural networks, Fear-driven cohesion, Dispersion relation, Small-world topology, Swarm intelligence

TL;DR: This paper shows that minimal ecological rules, implemented in adaptive Smart-Boid agents, generate communication network topologies that closely match empirical animal systems when analyzed through Vietoris–Rips filtrations and persistent homology.

Abstract: We test whether adaptive agents, a.k.a. ``Smart-Boids'' governed by neural networks under evolutionary pressure, can generate topologies resembling animal communication networks. Using Vietoris--Rips filtrations and persistent homology, we compare $1000+$ empirical networks to simulations via feature-based correlations. Minimal ingredients (fear of isolation, limited perception, inertia, exclusion, noise) reproduce both sparse and small-world topologies observed in diverse animal systems. Results suggest that ecological constraints, rather than complex cognition, drive the emergence of communication networks.

Submission Number: 47