Bottom-up robust modeling for the foraging behavior of Physarum polycephalum Journal: Journal of The Royal Society Interface
Damiano Reginato, Daniele Proverbio, Giulia Giordano · 2024
Read Bottom-up robust modeling for the foraging behavior of Physarum polycephalum Journal: Journal of The Royal Society Interface by Damiano Reginato, Daniele Proverbio, Giulia Giordano (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-19790.txt.
Introduction
Damiano Reginato, Daniele Proverbio, and Giulia Giordano model, in a 2024 paper, how a slime mold hunts for food and builds networks. The mold Physarum polycephalum can link food sources with a web of tubes, all without a brain. Older models copy the web well but say less about the smallest rules that make the feat robust.
The team built three agent-based models of rising complexity, so they could add rules one step at a time. They scored how well the agents linked food sources on poor ground, and tested how the models cope with varied settings and mixed agent traits. Robust web patterns emerged, matching lab evidence. The work also speaks to engineers who want self-building networks without central control.
Catalog record
| CREATOR | Damiano Reginato, Daniele Proverbio, Giulia Giordano |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 50 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-19790
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Bottom-up robust modeling for the foraging behavior of Physarum polycephalum Journal: Journal of The Royal Society Interface Damiano Reginato, Daniele Proverbio, Giulia Giordano Abstract The true slime mold Physarum polycephalum has the remarkable capability to perform self-organized activities such as network formation among food sources. Despite well reproducing the emergence of slime networks, existing models are limited in the investigation of the minimal mechanisms, at the microscopic scale, that ensure robust problem-solving capabilities at the macroscopic scale. To this end, we develop three progressively more complex multi-agent models to provide a flexible framework to understand the self-organized foraging and network formation behaviors of Physarum . The hierarchy of models allows for a stepwise investigation of the minimal set of rules that allow bio-inspired computing agents to achieve the desired behaviors on nutrient-poor substrates. By introducing a quantitative measure of connectedness among food sources, we assess the sensitivity of the model to user-defined and bio-inspired parameters, as well as the robustness of the model to parameter heterogeneity across agents. We ultimately observe the robust emergence of pattern formation, in line with experimental evidence. Overall, our study sheds light onto the basic mechanisms of self-organization and paves the way towards the development of decentralized strategies for network formation in engineered systems, focusing on trade-offs between biological fidelity and computational efficiency. 1 Introduction Self-organization is the spontaneous ability of organisms to perform global tasks through an ensemble of local actions and adaptation. A leading specimen for studies about emergent problem-solving ability and self-organization is the true slime mold Physarum polycephalum , a member of the Myxogastria class of organisms, also known as acellular slime molds; in fact, in its plasmodium (vegetative) stage, P. polycephalum is composed of a single cellular membrane enveloping a multitude of nuclei. In biology and complex systems science, Physarum is extremely valuable to study cell cycle [ 1 ] and DNA replication [ 2 ] , as well as motility [ 3 ] , environmental sensing [ 4 ] , and response to chemical and physical stimuli [ 5 ] . In its vegetative stage, the plasmodium moves through a peculiar process called shuttle-streaming [ 6 ] . The cytoplasm differentiates into two forms: rigid gel-like ectoplasm (gel, a stiff sponge-like matrix composed of actin–myosin fibers) and fluid endoplasm (sol) flowing within the matrix. Inside vein-like structures of ectoplasm, periodic oscillations of contractile fibrils cause back-and-forth movements of the endoplasm, making the plasmodium elongate and move. When a local concentration of nutrients is sensed, the external membrane softens in its proximity, making the streaming more effective in that direction and thus allowing the organism to reach the food source more quickly [ 7 ] . Initially, a fine mesh of tubules (well approximated by a planar sheet) covers the area until the food source is engulfed by the plasmodium. Then, the network structure changes to efficiently distribute the nutrients throughout the organism. The more nutrient flows through a vein , the thicker this becomes, in a reinforcing feedback loop that ultimately leads to a mature network configuration [ 8 ] . Shuttle-streaming enables P. polycephalum to showcase emergent computing ability: it can successfully link multiple food sources [ 9 ] while avoiding risky environments [ 10 ] , solve puzzles and mazes by connecting food sources via shortest paths [ 11 ] , and optimize in-vitro replicas of transport networks [ 12 ] . Given its unique ability to solve complex problems without possessing central brain-led coordination, P. polycephalum inspires experiments and models about swarm intelligence, emergent computing and self-organization [ 13 ; 14 ] . It also serves as bio-inspiration for the development of efficient algorithms for optimization, maze exploration and computing [ 15 ; 16 ] , as well as for artificial swarm systems [ 17 ; 18 ] . To capture the mechanisms underpinning the emergence of complex problem-solving behaviors, numerous models of P. polycephalum have been proposed, at various scales and with different scopes [ 6 ] . Fluid-dynamics models studied directed migration and memory brought about by oscillations and peristalsis within the cellular endoplasm [ 19 ; 20 ] ; network formation was reproduced using e.g. Hagen-Poiseuille laws on graphs [ 21 ] or reaction-diffusion models [ 22 ; 23 ] ; optimal path formation was demonstrated using memristor-like network dynamics [ 24 ; 25 ] ; and bio-inspired models of computing systems have been surveyed e.g. in [ 26 ] . These models captured several aspects of Physarum evolution, such as the influence of peristalsis and fluid dynamics in its growth, or the role of nonlinear threshold responses to stimuli in driving optimality in network formation. However, they lack a bridge through scales, to inquire in detail how microscopic interactions lead to the emergence of patterns at macroscopic scale – and how disturbances are coped with, at any scale, leading to robust behaviors. To study macroscopic behaviors emerging from microscopic interactions, multiscale models have been developed, like Cellular Automata [ 12 ; 27 ; 28 ] , having low complexity and thus low biological fidelity, or Multi-Agent Systems (MASs) [ 26 ] , having higher complexity but potentially higher computational cost, to investigate point-of-interest and foraging behaviors of the whole organism, driven by local decision-making at the microscopic level. As a fundamental advantage, MASs allow to add layers of complexity that go beyond simpler individual-based approaches. For instance, the decision-making rules of individual computing units can be made more realistic by incorporating knowledge of the biological and mechanical phenomena; multiple self-organizing behaviors can be integrated; and dynamics over multiple scales can be concurrently tracked [ 29 ] . Going beyond what has been discussed in previous works, MASs can also be fruitfully employed to study the robustness of emergent patterns against uncertainties and disturbances at any scale, and thus assess under which stress conditions a self-organizing entity manages to thrive. However, designing a MAS to model the behavior of P. polycephalum requires a careful interpretation of the modeled agents: no easy parallel can be drawn between a collection of independent agents, having a granular nature, and the whole Physarum organism in its plasmodium phase. Nonetheless, following an approach discussed in [ 12 ; 26 ; 30 ] , agents can be thought of as units of solid cytoplasm being transported through the fluid endoplasm (gel/sol interaction). Particle-based models with a bottom-up approach could help elucidate the basic computing mechanisms required to enable network formation and foraging; however, it is still unclear which are the minimal behavioral rules needed to reproduce foraging as a combination of two self-organized mechanisms – search and networking – along with efficient pruning and nutrient transport. Moreover, recent models [ 26 ] have considered diffusive food sources, to exploit Physarum ’s chemotactical capability of streaming up gradients, while nutrients used in experiments (such as oat flakes) or present in real life are typically little or not diffusive [ 31 ] . Hence, the question remains of how P. polycephalum manages to search a more realistic environment and complete its foraging task. In this work, to investigate the minimal mechanisms leading to self-organized efficient network formation, we develop and test three competing agent-based models of increasing complexity and realism; we