Dynamics of social interactions, in the flow of information and disease spreading in social insects colonies: Effects of environmental events and spatial heterogeneity

Xiaohui Guo, Jun Chen, Asma Azizi, Jennifer H. Fewell, Yun KangJournal of Theoretical Biology 2020 · DOI · arXiv

Abstract

Social insect colonies operate as decentralized systems where group-level behaviors emerge from local interactions. This study utilizes agent-based modeling to investigate how spatial heterogeneity—the non-random distribution of individuals based on task specialization—regulates the flow of information and pathogens. We find that spatial clustering serves a dual purpose: accelerating information spread within functional groups while providing "organizational immunity" by slowing transmission across the broader colony.

Introduction: The Spatial Layout of Social Spread

In a social insect colony, there is no central authority. Decisions regarding nest relocation, defense, or foraging emerge from physical interactions between individuals. However, these interactions present a trade-off: high contact rates facilitate rapid information sharing but also increase vulnerability to pathogens.

Spatial clustering of ants into task-specific zones: food processing, brood care, and trash maintenance.

The researchers explored how spatial heterogeneity—the tendency of workers to stay near specific task zones (spatial fidelity)—impacts these dynamics. By quantifying the "Spatial Heterogeneity Degree" (SHD), the study examines how organized nesting patterns act as a regulatory mechanism for agent spreading.

Methodology: Agent-Based Modeling of Colony Dynamics

The study developed an Agent-Based Model (ABM) using a discrete-time Markov chain on a $K \times K$ grid. The model simulates workers with specific tasks, walking styles (random vs. drifted toward task zones), and information statuses.

Flowchart of the agent-based model illustrating worker selection, movement logic, and the mechanics of information spread through contact.

Three environmental scenarios were tested:

  1. Random-Mixing (RM): Uniform random movement with no spatial preference.
  2. Random-Initial-Distribution (RID): Workers start randomly but move toward task zones, simulating recovery from a disturbance.
  3. Aggregated-Initial-Distribution (AID): Workers start and remain near their task zones, representing a stable, healthy colony.

Key Insights: The Dual-Functionality of Spatial Fidelity

The research revealed a strong linear correlation between spatial clumping (SHD) and average contact rates ($R(t) = -0.1033 + 11.895 \cdot SHD(t)$). A critical finding is the context-dependent nature of spatial fidelity:

  • In the RID scenario, high spatial fidelity promotes spread. As workers move from random positions toward their designated zones, the resulting movement creates a "mixing" effect that facilitates contact.
  • In the AID scenario, high spatial fidelity inhibits spread. Task groups become isolated "silos," creating biological barriers that prevent an agent (information or pathogen) from jumping between clusters.

This phenomenon, termed organizational immunity, suggests that spatial segregation is a primary defense mechanism, shielding the colony from localized outbreaks.

Results and Mathematical Modeling

The spread of agents was described using a modified nonlinear logistic growth model: $$\frac{dI}{dt} = R_{RM} \cdot Q_p \cdot e^{-I(t)} \cdot I(t) \cdot (1 - I(t))$$ This model accounts for "local saturation," where the rate of new "infections" slows as an area becomes crowded with informed individuals. In the AID scenario, increasing spatial fidelity from 20% to 98% caused the transmission rate ($Q_p$) to drop significantly (from ~0.0025 to ~0.0007), whereas it increased in the RID scenario.

Conclusion and Future Directions

While the current model provides a robust framework for understanding spatial dynamics, it assumes static task roles. Future research will incorporate task-switching dynamics and attenuation networks, where individuals adjust behavior based on social cues. These findings bridge the gap between theoretical modeling and the biological reality of social immunity in insect societies.

Summary updated

BibTeX
@article{guo2020dynamics,
  author = {Xiaohui Guo and Jun Chen and Asma Azizi and Jennifer H. Fewell and Yun Kang},
  title = {{Dynamics of social interactions, in the flow of information and disease spreading in social insects colonies: Effects of environmental events and spatial heterogeneity}},
  journal = {Journal of Theoretical Biology},
  year = {2020},
  doi = {10.1016/j.jtbi.2020.110191},
}
Figure