NEURON (Formal)
| Collection | International Encyclopedia of Systems and Cybernetics |
|---|---|
| Year | 2004 |
| Vol. (num.) | 2(2) |
| ID | ◀ 2285 ▶ |
| Object type | Methodology or model |
Model of the biological neuron made up by a binary element whose state is +1 (active) or -1 (inactive). (L. PERSONNAZ et al., 1988, p.1364).
These authors explain: “It was conceived by W. McCULLOCH and W. PITTS, from Chicago University in 1943… It updates its state periodically in the following way: It computes the sum of its inputs (which are the outputs of the formal neurons to which it is connected). The value of every input being modulated by the corresponding synaptic efficiency, it takes a decision by comparing this sum with a threshold of its own. If the sum is superior to the threshold, the neuron adopt its active state (+1); in the opposite case, it remains in its inactive state (-1)” (Ibid).
The formal neuron is thus a totalizer of impulses, combined with a simple binary decider, or an “all or nothing” model, as stated by H.von FOERSTER (1981, p.33).
(For more precise and detailed description and comments of this important model, see H.von FOERSTER, 1981, p.33-42)
See also
Parallel Distributed Processing