Emergent Collective Reproduction via Evolving Neuronal Flocks

Nam H. Le, R. W. G. Watson, Mike Levin, Christopher L. BuckleyarXiv 2024 · DOI · arXiv

Abstract

This study facilitates the understanding of evolutionary transitions in individuality (ETIs) through a novel artificial life framework, named VitaNova, that intricately merges self-organization and natural selection to simulate the emergence of complex, reproductive groups. By dynamically modelling individual agents within an environment that challenges them with predators and spatial constraints, VitaNova elucidates the mechanisms by which simple agents evolve into cohesive units exhibiting collective reproduction. The findings underscore the synergy between self-organized behaviours and adaptive evolutionary strategies as fundamental drivers of ETIs. This approach not only contributes to a deeper understanding of higher-order biological individuality but also sets a new precedent in the empirical investigation of ETIs, challenging and extending current theoretical frameworks.

Introduction

One of the most profound transitions in biological history is the Evolution of Transition in Individuality (ETI), where independent units coalesce into a new, higher-level individual. Understanding how "collective reproduction" emerges from individual-level selection remains a significant challenge. Most existing models either assume this transition or impose top-down constraints. This paper introduces VitaNova, a bottom-up Artificial Life framework where collective reproduction emerges spontaneously from individual agents—"boids"—evolving to survive in a predator-filled environment.

Methodology

The VitaNova environment is populated by autonomous agents controlled by evolving neural networks.

  • Neural Architecture: Each boid is governed by a neural network with 11 inputs and 7 outputs, allowing for dynamic steering and role switching between "workers" (green) and "soldiers" (blue).
  • Evolutionary Mechanics: Boids reproduce asexually upon reaching an energy threshold, passing on mutated neural weights. There is no global fitness function; evolution is driven by local survival.
  • Selection Pressures: Predators (red triangles) incentivize group formation. Specifically, predators avoid clusters containing three or more soldiers, creating a natural selection pressure for cohesive, defensive social structures.

Emergent Collective Dynamics

The most striking result of the VitaNova simulations is the spontaneous emergence of ring-shaped flocks that exhibit a life cycle reminiscent of biological mitosis. These rings grow as internal boids reproduce and eventually bifurcate into two stable daughter rings once a critical size is reached.

Figure 5: Emergent ring-shaped flocks at simulation step 4287, illustrating the formation of multiple distinct collectives.

Quantitative analysis confirms a repeatable cycle of growth and reproduction. As illustrated in the tracking data, the system undergoes phases where flock size increases steadily (growth), followed by a sharp drop in average size and a corresponding increase in the total number of flocks (reproduction).

Figure 6: Temporal analysis of flock count and average flock size, demonstrating the cyclical nature of emergent collective reproduction.

These findings suggest that higher-order individuality can emerge without central control. The "offspring" flocks inherit the structural and behavioral strategies of the "parent" flocks, ensuring generational stability of collective traits.

Conclusion and Future Directions

VitaNova provides a powerful computational model for studying ETIs, demonstrating that complex collective reproduction can emerge from simple individual-level interactions. Future work will involve introducing more complex resource dynamics and co-evolving predators to observe potential "evolutionary arms races." Additionally, the researchers plan to apply thermodynamic metrics, such as entropy production, to further analyze the transitions between levels of individuality.

Summary updated

Abstract

This study facilitates the understanding of evolutionary transitions in individuality (ETIs) through a novel artificial life framework, named VitaNova, that intricately merges self-organization and natural selection to simulate the emergence of complex, reproductive groups. By dynamically modelling individual agents within an environment that challenges them with predators and spatial constraints, VitaNova elucidates the mechanisms by which simple agents evolve into cohesive units exhibiting collective reproduction. The findings underscore the synergy between self-organized behaviours and adaptive evolutionary strategies as fundamental drivers of ETIs. This approach not only contributes to a deeper understanding of higher-order biological individuality but also sets a new precedent in the empirical investigation of ETIs, challenging and extending current theoretical frameworks.
BibTeX
@misc{le2024emergent,
  author = {Nam H. Le and R. W. G. Watson and Mike Levin and Christopher L. Buckley},
  title = {{Emergent Collective Reproduction via Evolving Neuronal Flocks}},
  year = {2024},
  eprint = {2409.13254},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2409.13254},
}
Figure