🌿 CuriousWinds

Slime Mold Computing & Physarum Polycephalum Networks

Bio-computational state machines, morphogenetic field signaling, and molecular information processing.

Executive Summary & Key Takeaways
Foundational Synthesis for Practitioners & Research Systems

Physarum polycephalum solves NP-hard problems via hydrodynamic flow, not neural control, its tubular network’s radius dictates efficiency through fourth-power scaling, defying centralized computation myths.

01 Shuttle Streaming

Actomyosin-driven cytoplasmic oscillations (60, 90s cycles) create directional pressure gradients for dynamic, neural-free signal routing.

02 Hydrodynamic Memristors

Tube radius (r⁴) governs flow efficiency; shear stress remodels tubules, self-adjusting conductance via plastic dilation/collapse.

03 Gradient Dependency

Disrupting nutrient gradients halts flow, collapsing tubules into isotropic expansion and breaking directed computation.

04 Environmental Thresholds

Maintain 1, 2 cm nutrient spacing; avoid >20°C or <15°C to preserve actomyosin-driven plasticity and flow.

Slime mold computing provides the ultimate empirical demonstration of decentralized biocomputation in non-neural living matter. The acellular slime mold Physarum polycephalum is a single-celled, multinucleated plasmodial amoeba with no brain, no neurons, no sensory organs, and no centralized executive controller. Yet when placed into complex spatial environments, this brainless blob of yellow slime navigates intricate labyrinths, calculates shortest path graphs, solves NP-hard traveling salesperson problems, and designs transportation networks that rival the work of human civil engineers. How can a single bag of fluid cytoplasm execute graph theory algorithms that challenge modern electronic computers?

01 1 · BIOMECHANICS · CYTOPLASMIC SHUTTLING

Foundations of Slime Mold Computing

Empirical Mechanisms

To understand how Physarum computes, one must examine its microscopic mechanics. The plasmodium consists of an interconnected tubular network through which intracellular cytoplasm streams rhythmically back and forth in a phenomenon known as “shuttle streaming,” with a oscillation period of roughly 60 to 90 seconds.

This streaming is driven by actomyosin contractions in the gel layer surrounding the tubules. When a section of the plasmodium encounters a nutrient attractant (such as oat flakes), local receptors trigger an increase in contraction frequency. Conversely, when exposed to light or cold, contractions diminish. The resulting spatial pressure differentials drive fluid cytoplasm through the tubular network according to the Hagen-Poiseuille equation for viscous laminar flow in cylindrical tubes:

Q = ( pi · r^4 · Delta P ) / ( 8 · eta · L )

Where Q is volumetric flow rate, r is tube radius, Delta P is pressure drop, eta is dynamic cytoplasmic viscosity, and L is tube length. Notice the extraordinary sensitivity to radius: flow rate scales with the fourth power (r^4).

Crucially, the tubules exhibit physical plasticity: high fluid shear stress against the actin cortex causes the tube to dilate (increasing r), while low or stagnant flow causes the tube to atrophy and wither away. Mathematically, the tube acts as a memristor, a resistor with memory, whose conductance changes based on the history of flow that has passed through it. By coupling Hagen-Poiseuille hydrodynamics with shear-induced tubule remodeling, the physical slime mold network behaves as a self-optimizing analog circuit that automatically prunes redundant pathways and reinforces high-efficiency conduit lines.

02 2 · EXPERIMENTAL PHYSICS · MAZE-OPTIMIZATION

Empirical Breakthrough

Nakagaki and the Maze-Solving Amoeba (2000)

The biocomputational capacity of Physarum was revealed to modern science in September 2000 in a legendary paper published in Nature by Japanese biophysicist Toshiyuki Nakagaki, Hiroyasu Yamada, and Ágota Tóth (“Maze-solving by an amoeboid organism”). Nakagaki carved a complex 30-square-centimeter labyrinth out of plastic, containing four possible routes between an entrance and an exit.

Nakagaki placed oat flakes (food sources) at the entrance and exit of the maze and spread Physarum plasmodium throughout the entire labyrinth. Initially, the amoeba filled all available corridors as a uniform sheet. Within hours, however, the slime mold began remodeling its body. In dead-end corridors where no net fluid was flowing between the two food sources, contraction waves lacked directional bias; the tubes atrophied and dissolved.

Along corridors connecting the food sources, fluid surged rhythmically. Because shorter corridors have lower hydrodynamic resistance (L in the denominator), fluid streamed faster along the shortest path, triggering selective r^4 tubule dilation. Within 26 hours, the plasmodium had completely evacuated all dead ends and longer alternate paths, condensing its entire biomass into a single thick, pulsing trunk that traced the exact mathematical shortest path through the maze. The brainless amoeba had computed Dijkstra’s shortest-path graph algorithm using fluid mechanics.

03 3 · NETWORK THEORY · ADAPTIVE PATHFINDING

Replicating the Tokyo Rail Network (Tero et al., 2010)

In January 2010, Atsushi Tero and Nakagaki published an even more astonishing empirical investigation in Science (“Rules for biologically inspired adaptive network design”). The researchers placed oat flakes on a wet agar plate in geometric arrangements matching the geographic locations of Tokyo and 35 surrounding metropolitan cities, with light barriers representing mountains and Tokyo Bay.

As Physarum expanded, it connected the 36 food nodes. The resulting slime mold network was evaluated across three engineering criteria: cost (total tube length), efficiency (average transit time between stations), and fault tolerance (resilience to accidental disconnection). Nakagaki compared the biological network to the actual, human-engineered Tokyo rail transit system, developed over decades by urban planners and civil engineers.

In this triumph of slime mold computing, the biological network matched the real Tokyo rail network with startling quantitative accuracy, achieving identical balance between cost, travel efficiency, and redundant loop resilience. The researchers distilled Physarum’s local feedback rules into a clean mathematical algorithm:

d D_ij / dt = f(|Q_ij|) – r · D_ij

Where D_ij represents the diameter of the tube between nodes i and j, Q_ij is the flow flux, and r is the decay rate. The Tero-Nakagaki algorithm is now used in computer science to design decentralized communication networks, pipeline layouts, and peer-to-peer data routing protocols that self-heal under node failures without centralized coordination.

04 4 · HYBRID COMPUTING · ORGANIC-SILICON INTERFACING

Applied Unconventional Computing

The Adamatzky Bio-Processors

At the Unconventional Computing Laboratory at the University of the West of England, professor Andrew Adamatzky constructs hybrid wetware-silicon bio-processors powered by living Physarum. By placing slime mold on microelectrode arrays, Adamatzky’s team uses the organism’s electrical spikes and tubule growth vectors to build logical NOR gates, Voronoi diagram calculators, and analog controllers for autonomous robots. Slime mold computing demonstrates that complex algorithmic spatial intelligence is a natural property of physical fluid self-organization.

Curated Inquiry Series • 10 Investigations

The Edge of Chaos: How Living Cells Compute Without Microprocessors

An empirical exploration through the foundational dynamics, systemic trade-offs, and practical architecture of the edge of chaos: how living cells compute without microprocessors.

Primary Scholarly Provenance & Empirical Attribution
Strukov, D. B., Snider, G. S., Stewart, D. R., & Williams, R. S. (2008). The missing memristor found. Nature, 453(7191), 80–83; Physical Dynamics & Non-Linear State Transitions.
DOI: 10.1038/nature06932