Criticality is the point at which a system sits on the exact boundary between two states, where a tiny change can either die out or cascade into something enormous. The term shows up across nuclear physics, thermodynamics, complexity science, and neuroscience, but the core idea is the same: a system balanced at a tipping point where its behavior can shift dramatically.
Criticality in Nuclear Physics
The most familiar use of “criticality” comes from nuclear reactors. Here, it describes whether a nuclear chain reaction is self-sustaining. When an atom of uranium or plutonium splits (fission), it releases neutrons. Those neutrons can hit other atoms and cause them to split too, releasing more neutrons. Whether this chain reaction grows, shrinks, or holds steady depends on a single number: the effective neutron multiplication factor, written as k-effective.
K-effective is the ratio of neutrons produced in one generation of fission to the total number lost through absorption and leakage in the previous generation. When k-effective equals exactly 1, the system is “critical.” Every generation of fission produces exactly enough neutrons to sustain the next generation, no more, no less. A nuclear power plant operates at criticality on purpose. The reaction runs at a steady, controlled rate, producing a constant output of heat.
When k-effective drops below 1, the system is “subcritical” and the chain reaction fizzles out. When it climbs above 1, the system is “supercritical” and the reaction intensifies. Reactor operators adjust control rods and other mechanisms to keep k-effective hovering right at 1. An uncontrolled supercritical state is what makes nuclear weapons work, and it’s also what happens in a criticality accident.
Criticality Accidents and Radiation
A criticality accident occurs when fissile material unexpectedly reaches a self-sustaining or supercritical chain reaction outside of a controlled reactor setting. This can happen in fuel processing plants, research labs, or storage facilities when too much fissile material accumulates in one place. The result is an intense burst of radiation that can last fractions of a second or, in some configurations, persist for hours.
People caught near a criticality accident receive a massive dose of penetrating radiation in a very short time. Radiation exposure above about 0.7 Gray causes acute radiation syndrome, which begins with nausea and vomiting and progresses to bone marrow destruction. At doses above roughly 10 Gray, the lining of the gastrointestinal tract breaks down irreparably, typically causing death within two weeks. Extremely high doses, above 50 Gray, damage the cardiovascular and central nervous systems directly, and death can follow within days. Historical criticality accidents, like those at Los Alamos in the 1940s and the Tokaimura facility in Japan in 1999, exposed workers to fatal doses in seconds.
The Critical Point in Physics
In thermodynamics, criticality refers to the critical point of a phase transition. Think of water: at normal pressures, liquid water and steam are clearly different. But if you increase temperature and pressure simultaneously, the distinction between liquid and gas narrows. At a specific temperature and pressure (the critical point), the boundary between the two phases vanishes entirely. The substance becomes a single, uniform phase called a supercritical fluid.
What makes the critical point so interesting to physicists is that the system’s behavior changes in fundamental ways. Normally, the interactions between molecules only matter over very short distances, on the scale of atomic spacing. At the critical point, correlations between molecules extend across all length scales. The material develops a fractal-like character, where patterns repeat at every scale of magnification. Certain measurable quantities, like how strongly the material responds to small changes in temperature or pressure, spike toward infinity near the critical point.
This behavior turns out to be universal. Completely different materials, a magnet losing its magnetization, a fluid at its liquid-gas boundary, a liquid crystal changing orientation, all follow the same mathematical patterns near their critical points. The specific atoms involved don’t matter. What matters is the system’s dimensionality and the type of order parameter (density, magnetization, molecular orientation) involved. This discovery of universality was one of the major achievements of 20th-century physics.
Self-Organized Criticality
In 1988, physicist Per Bak and colleagues proposed a striking idea: some complex systems don’t need to be tuned to their critical point by an outside force. They drive themselves there naturally. This concept is called self-organized criticality.
The classic example is a sandpile. Imagine dropping grains of sand one at a time onto a table. The pile grows steeper until it reaches a particular slope. At that point, adding a single grain might do nothing, or it might trigger an avalanche that reshapes the entire pile. Most of the time, the disturbance is small. Occasionally, it’s massive. The pile maintains itself right at the boundary between stability and collapse without anyone controlling it. This was later demonstrated experimentally using rice piles.
The mechanism is straightforward. When the system is below the critical state (slopes too shallow), grains accumulate and push it toward criticality. When it overshoots (slopes too steep), avalanches pull it back. This feedback loop keeps the system parked at the critical point indefinitely. The signature of self-organized criticality is a specific statistical pattern: many small events and rare large ones, following a power-law distribution. Earthquakes follow this pattern (many small tremors, few large quakes), as do forest fires, stock market fluctuations, and even the sizes of cities.
Criticality in the Brain
Neuroscientists have gathered substantial evidence that the brain operates near a critical state. In this context, criticality means the boundary between two regimes of neural activity: one where signals between neurons die out too quickly, and one where they amplify uncontrollably. At the critical point between these two extremes, the brain hits a sweet spot.
Operating at criticality gives the brain several measurable advantages. It maximizes dynamic range, meaning the brain can detect and distinguish both faint and intense signals. It maximizes information transmission, allowing signals to travel efficiently across distant brain regions without being lost or distorted. And it maximizes representational capacity, the brain’s ability to encode a wide variety of distinct patterns of activity. Neural activity at criticality shows the hallmarks physicists expect: large fluctuations in synchronization, scale-free patterns, and long-range correlations across space and time.
When the brain drifts away from criticality, function degrades. Too far in the subcritical direction and neural communication becomes sluggish, with signals unable to propagate effectively. Too far in the supercritical direction and activity becomes excessively synchronized, resembling something closer to a seizure. Some researchers now frame neurological and psychiatric conditions partly in terms of where the brain sits relative to its critical point.
Criticality in Medicine
In clinical medicine, “criticality” takes on a more straightforward meaning: how sick a patient is. Intensive care units use standardized scoring systems to quantify patient criticality and predict outcomes. The most widely used is APACHE II, developed in 1985, which assigns a numerical severity score based on physiological measurements like heart rate, blood pressure, temperature, and organ function. Higher scores correspond to higher mortality risk.
Other systems include SOFA (Sequential Organ Failure Assessment), created by the European Society of Intensive Care Medicine, which tracks the function of six organ systems over time to assess whether a patient is improving or deteriorating. These scores help clinicians allocate resources, compare outcomes across hospitals, and make decisions about treatment intensity. In this medical context, “criticality” simply means the severity of a patient’s condition and how close they are to a life-threatening threshold.
The Common Thread
Across all these fields, criticality describes a system at a boundary. In a reactor, it’s the boundary between a dying chain reaction and a growing one. In thermodynamics, it’s the boundary between two phases of matter. In a sandpile, it’s the boundary between stability and avalanche. In the brain, it’s the boundary between silence and seizure. The reason the concept is so powerful is that systems at criticality all share the same mathematical behavior: extreme sensitivity to small perturbations, correlations that span all scales, and a mix of large and small fluctuations that follow predictable statistical laws.