Quasicriticality

Quasicriticality describes a dynamical regime in which a system hovers near but never precisely at a critical point, exhibiting many of the computational and statistical properties of criticality — power-law distributions, scale-free correlations, maximal sensitivity — without the fragility that exact criticality entails. The concept was developed to resolve a tension in self-organized criticality research: biological and neural systems appear critical in aggregate statistics, yet they are regulated by homeostatic mechanisms that would seem to push them away from the singular point.

The resolution is that these systems are not exactly critical. They are quasicritical — close enough that large-scale behavior resembles criticality, but buffered by feedback mechanisms that prevent the runaway avalanches and catastrophic collapses that exact criticality would permit. In neural systems, pharmacological perturbations that drive the brain slightly subcritical or supercritical degrade power-law statistics, but the natural operating point appears to be a narrow band around criticality rather than the point itself. Homeostatic synaptic plasticity acts as a governor, pulling the system back toward the critical band when it wanders too far.

Quasicriticality is not merely criticality with noise. It is a distinct dynamical regime with its own properties. The correlation length in a quasicritical system is large but finite; the power law has a cutoff at very large scales; and the system retains a memory of its recent history that pure criticality would erase. These properties make quasicriticality a better candidate than exact criticality for the operating point of biological computation: it offers sensitivity without fragility, and multi-scale response without unbounded cascades.