[Paper Review] Self-Organization and Artificial Life: A Review

Carlos Gershenson, Vito Trianni+2 more|arXiv (Cornell University)|Apr 3, 2018

Modular Robots and Swarm Intelligence3 citations

TL;DR

This paper provides a comprehensive review of self-organization in Artificial Life (ALife), examining its theoretical foundations, applications across soft (simulated), hard (robotic), and wet (biochemical) domains, and its role in emergent complexity. It clarifies conceptual ambiguities, synthesizes key research, and outlines future directions for guided self-organization in ALife systems.

ABSTRACT

Self-organization has been an important concept within a number of disciplines, which Artificial Life (ALife) also has heavily utilized since its inception. The term and its implications, however, are often confusing or misinterpreted. In this work, we provide a mini-review of self-organization and its relationship with ALife, aiming at initiating discussions on this important topic with the interested audience. We first articulate some fundamental aspects of self-organization, outline its usage, and review its applications to ALife within its soft, hard, and wet domains. We also provide perspectives for further research.

Motivation & Objective

• To clarify the concept of self-organization, which is often misinterpreted or inconsistently defined across disciplines.
• To examine the role of self-organization in the three domains of Artificial Life: soft (simulations), hard (robotics), and wet (biochemical systems).
• To synthesize key research on self-organization in ALife, including collective behavior, morphogenesis, and pattern formation.
• To highlight the importance of self-organization in enabling emergent, decentralized, and adaptive behaviors in artificial systems.
• To stimulate discussion and guide future research by identifying open challenges and opportunities in guided self-organization.

Proposed method

• Surveying foundational literature on self-organization from cybernetics, statistical mechanics, and information theory.
• Classifying ALife research into soft, hard, and wet domains to analyze domain-specific applications of self-organization.
• Analyzing case studies such as flocking in birds and robots, cellular automata, and chemical pattern formation (e.g., Belousov-Zhabotinsky reaction).
• Reviewing theoretical frameworks such as entropy minimization, statistical complexity, and information-theoretic measures of organization.
• Examining self-organizing systems in artificial societies, self-replicating robots, and swarm robotics using agent-based models.
• Integrating insights from guided self-organization, where dynamics are shaped to achieve specific attractors or functional outcomes.

Experimental results

Research questions

• RQ1 What are the core characteristics and definitions of self-organization, and why is it often misunderstood or inconsistently applied?
• RQ2 How does self-organization enable the emergence of complex, global patterns from local interactions in ALife systems?
• RQ3 In what ways has self-organization contributed to progress in soft, hard, and wet ALife domains?
• RQ4 What mechanisms allow self-organization to be guided or steered toward desired outcomes, and how can this be formalized?
• RQ5 What are the open challenges and future research directions for self-organization in artificial life and complex systems?

Key findings

• Self-organization is a central mechanism in ALife, enabling the emergence of complex, functional patterns from decentralized, local interactions.
• In soft ALife, self-organization underlies phenomena such as flocking in agent-based models, pattern formation in cellular automata, and evolutionary dynamics in Boolean networks.
• In hard ALife, self-organizing swarms of robots achieve collective tasks like foraging, navigation, and decision-making without centralized control.
• In wet ALife, self-organization drives morphogenesis in biological systems and chemical pattern formation, such as in the Belousov-Zhabotinsky reaction.
• Guided self-organization, where system dynamics are shaped to achieve specific attractors, is a promising direction supported by information-theoretic frameworks.
• Despite conceptual ambiguity, self-organization remains a powerful and practical concept in ALife, with broad applicability across simulation, robotics, and biochemical systems.