CRITICALITY AND PERCOLATION
| Collection | International Encyclopedia of Systems and Cybernetics |
|---|---|
| Year | 2004 |
| Vol. (num.) | 2(1) |
| ID | ◀ 766 ▶ |
| Object type | General information, Methodology or model |
While criticality in a composite system corresponds to its tendency to turn back to a defined critical state which acts as an attractor (BAK et al., 1988, p.365), conversely percolation corresponds to its tendency to runaway behavior when a percolation threshold is reached.
This is made clear by P. BAK, C. TANG and K. WIESENFELD who write: “The system will become stable precisely at the point when the network of minimally stable clusters has been broken down to the level where the noise signal cannot be communicated through infinite distances” (Ibid, p. 367).
Most of the percolation processes tend to spend themselves quickly because they destroy their own conditions of permanence. For example, a population crash is the customary result of a population explosion. The composite system whose this population constitutes a class of elements tends to go back to its critical state, which is thus “self-organized”. It is in this case that “the network… has been broken down to the level where the noise cannot be communicated through infinite distances” (Ibid).
Obviously the percolation threshold of the composite system is closely related to its critical state “in which minor events can cause chain reactions of many sizes” (P. BAK & K. CHEN, 1991, p.32). A percolation event is probably a runaway fluctuation iniciated in favorable conditions by a small one.