Abstract
Conceptual problems derivated from the classical formulation of the operant conditioning in statistical terms. The classical definition of the operant conditioning is based upon the so called by Skinner basic data of pychology: the frequency of responses. It is from this statistical concept as all its components and processes are also defined in a statistical form. The objective of this article is to analyze the problem that implies the operant conditioning estocastic definition as an obstacle to understand the fenomenical nature of behavior, because, instead of being correlationated with mutually independent elements, the discriminative stimulus and the reinforcer stimulus or reinforcement contingency, any operant behavior is itself a "discriminated contingency", that is, a co-presence in which contingencies are undisasociable of discriminations. As a result, it is finally concluded that it is not possible to produce a truly radical behaviorist definition of behavior unless the operant conditioning traditional definition based in the frequency of responses is surpassed and its characteristics as fenomenical-operatory process are respected.
