Work in progress. This is exploratory research into multicriticality. Multicritical points themselves are a well-established topic in statistical physics, and a recent literature actively studies how systems self-organize toward them (e.g. Sormunen, Gross & Saramäki 2025, arXiv:2506.04275). What is provisional here is the cross-system synthesis: the skeleton assignments, the pairwise mechanism claims, and the system-by-system profiles are hypotheses developed through computational experiments and literature review. They likely contain errors. The goal is to make the questions visible, not to present finished answers.

Multi-Critical Systems

Six physical systems where independent phase transitions converge

This is a series of six interactive tutorials examining a pattern that appears across condensed matter, biophysics, neuroscience, and plasma physics: multiple independent phase transitionsSudden qualitative changes in a system's behavior at threshold values — like water freezing or magnets losing magnetism. Characterized by power laws, diverging correlations, and universal behavior. converging in a single physical system. At the convergence point, the transitions couple, the universality classA category of phase transitions sharing the same critical exponents regardless of microscopic details. Water boiling and iron demagnetizing share a universality class — different physics, identical mathematics at the transition. changes, and new structure emerges that no single transition contains.

A single critical transition is a point in parameter space where one order parameterA measurable quantity that distinguishes phases. Zero on one side of the transition, nonzero on the other. Magnetization for magnets, density difference for liquids. At criticality, it fluctuates at all scales. diverges — the system reorganizes. This is well understood. A multi-critical pointA point in parameter space where two or more independent phase transitions converge. The physics at this point is qualitatively different from either transition alone — new exponents, new correlations, new phenomena. is where two or more independent transitions converge. At the coupling point, the physics changes qualitatively: the coupled system produces correlations, failure modes, and early-warning signals that exist only in the interaction between transitions, not in either one alone.

Each tutorial below teaches one physical system, starting from plain language and building to the multi-critical structure. Technical detail, falsification criteria, and citations are available in expandable sections for readers who want depth. The tutorials are ordered from simplest (two transitions, one coupling parameter) to most complex (four independent frameworks converging), and form three natural pairs connected by shared physical mechanisms.

The central question: why do multiple independent transitions converge in the same physical system — and what does the coupling produce that no single transition contains?

00 What Is Multicriticality?

START HERE

Interactive introduction. What happens when tipping points interact — explained from first principles with animated phase diagrams, a threshold coupling demo, and a side-by-side detection comparison. No physics background required. If you’re new to multicriticality, start here before the six system tutorials below.

1. The Tutorials

01 Glass Transition

T_MCT dynamic T_K thermodynamic

The cleanest bicritical case. Two transitions — one weakly first-order, one second-order — resolved as distinct specific heat peaks with independent topological signatures. In strong glass formers, they nearly merge. Fragility measures distance from the codimension-twoA point requiring two independent parameters tuned simultaneously to reach. Rare and structurally constrained — finding the intersection of two lines rather than just one point on one line. point where they meet.

Paired with: Protein Folding — connected through the Gardner/jamming chain

02 Active Matter

Manna onset braid selection novel η=1.04

Three critical phenomena with independent exponents. Defect onset follows Manna universality on closed surfaces and directed percolationA universality class for spreading processes with a preferred direction (like water flowing downhill through random soil). The most common class for transitions between absorbing and active states. on open ones — boundary topology selects the universality class. Confinement geometry selects braid class: golden braid in confinement, trivial braids on flat torus. A novel universality class at the periodic-to-chaotic boundary.◆ our readingThe measured η=1.04 exponent is a published result (Guldager Andersen et al. 2025); reading it as a multicritical coupling signature is this program’s hypothesis, not established consensus.

Paired with: Cardiac Tissue — connected through KT defect unbinding

03 Cardiac Tissue

Feigenbaum cascade KT unbinding RAF closure

Three transition frameworks in one tissue. Full period-doubling cascade before fibrillation. Kosterlitz-ThoulessA phase transition driven by topological defects (vortices) binding and unbinding. No divergent order parameter — instead, the character of the defect population changes qualitatively. Nobel Prize 2016. defect unbinding maps flutter to bound vortex pairs and fibrillation to unbound. The calcium signaling network is a maximal autocatalytic set — 17 of 17 reactions, zero individually essential, catastrophic only at class-level removal.◆ our resultAn original, not-yet-peer-reviewed result from this research program — distinct from established consensus.

Paired with: Active Matter — connected through KT defect unbinding

04 Protein Folding

thermodynamic jamming δ=9/2 SOC kinetics edge-of-chaos

Four independent phase-transition frameworks converge in one system. Cooperative two-state folding. Jamming at φ_c ≈ 0.55 with an anomalous exponent found nowhere else.◆ our resultAn original, not-yet-peer-reviewed result from this research program — distinct from established consensus. Power-law folding kinetics across three universality classes. Edge-of-chaos dynamics in disordered proteins. Four frameworks that normally describe unrelated physics, meeting here.

Paired with: Glass Transition — connected through the Gardner/jamming chain

05 Brain

avalanche criticality Griffiths phase spiral defects

Near-consensus on avalanche criticality from 140 datasets. Modular-hierarchical cortical architecture stretches the critical point into a Griffiths phaseA region near a phase transition where rare, strongly ordered clusters create anomalously slow dynamics. The system is technically disordered but behaves critically because rare regions act as local nucleation sites. — no fine-tuning. Cortical spiral waves carry conserved topological charge. The architecture that produces the Griffiths phase is the same architecture that supports topological defect dynamics.

Paired with: Tokamak — connected through Griffiths phase architecture

06 Tokamak H-Mode Pedestal

SOC avalanches magnetic topology predator-prey loop

The non-biological case. SOC sandpile dynamics produce the pedestal and ELMs. Magnetic islands mediate ELM suppression. A closed autocatalytic loop — gradients drive turbulence, turbulence generates zonal flows, zonal flows suppress turbulence, suppression steepens gradients — self-sustains at the edge of chaos. Multi-criticality is not biology-specific.

Paired with: Brain — connected through Griffiths phase architecture

2. Three Pairs

The six systems form three natural pairs, each connected by a shared physical mechanism that crosses domain boundaries. These are not analogies. In each pair, the same mathematical structure appears in both systems with measurable, quantitatively comparable signatures.

Glass ↔ Protein

The Gardner transitionA secondary transition within the glass phase where the energy landscape shatters from smooth basins into a fractal hierarchy of sub-basins. Discovered theoretically in spin glasses, now confirmed experimentally in colloids. sits below the Kauzmann temperature. Gardner produces jamming scalings. Jamming sets the protein packing fraction φ_c ≈ 0.55. The chain runs: glass bicriticality → Gardner → jamming → protein packing. Two systems, one causal chain through three transitions.

Tutorials 01 & 04

Active Matter ↔ Cardiac

Both systems contain topological defects that unbind through Kosterlitz-Thouless physics. In active nematics: ordered → periodic braiding → chaotic. In cardiac tissue: sinus rhythm → flutter (bound vortices) → fibrillation (unbound). Same three-regime structure, same topological mechanism, different substrates.

Tutorials 02 & 03

Brain ↔ Tokamak

Both are modular-hierarchical systems where the architecture stretches a critical point into a Griffiths phase. In the brain: cortical modules create an extended critical phase without fine-tuning. In the tokamak: nested transport barriers create a self-sustaining critical region. Same architectural principle, same consequence — robust criticality from structure, not parameter tuning.

Tutorials 05 & 06

3. The Four Skeletons

Across these six systems, the same mathematical structures recur. We decompose each system's multi-critical profile into four independent skeletonsIndependent mathematical frameworks — topology, dynamics, closure, computation — that recur across multicritical systems. Each was discovered independently by multiple research communities in different fields. — each a mathematical framework that was independently discovered by multiple communities. The tutorials above show where these skeletons appear in physical systems. The pages below teach each skeleton on its own terms.

Skeleton 2 (Dynamics / Criticality) is the anchor — the best-understood of the four, and the one running through every tutorial above. The other three are less familiar and have their own dedicated pages.

Skeleton 1: Topology

Protection → Computation. Braids, knot invariants, topological protection, fault-tolerant computation. Tracer: φ = 1.618...

7 communities, one mathematics →

Skeleton 2: Dynamics

Spontaneity → Coherent transport. Phase transitions, power laws, scale invariance. The anchor for everything else.

Taught throughout the six tutorials above

Skeleton 3: Closure

Self-sustenance → Self-creation. Catalytic loops, RAF theory, autopoiesis. No tracer — structural test only.

8 formalizations, 6 independent discoveries →

Skeleton 4: Computation

Universality → Coordination. Distributed computation, edge of chaos, emergence. Bridge to Skeleton 2 at the critical point.

8 fields, one framework →

4. Compare

Every system has a skeleton profile: how much of each mathematical structure it exhibits, scored at two depths (has the structure vs. uses the structure). The matrix below lets you overlay profiles for up to three systems and see what emerges — and what's missing.

5. The Question

Each of these systems is well-studied within its home domain. Cardiac electrophysiologists know the Feigenbaum cascade. Active matter physicists know the golden braid. Glass physicists know the two transitions. Plasma physicists know the SOC pedestal.

What none of these communities has asked is: why do multiple independent transitions converge in the same physical system? And what does the coupling between transitions produce that no single transition contains? Recent work on self-organization to multicriticalityA 2025 line of work (Sormunen, Gross & Saramäki) showing that some systems tune themselves toward multicritical points without external fine-tuning — a possible mechanism for why such points recur across domains. (Sormunen, Gross & Saramäki 2025) begins to ask a related question from the dynamics side.

That is what these tutorials examine.

Part of Domain Convergence Research