Modelo computacional para la formación de clases de equivalencia.

A computational model of neuronal net closely related with the formation of equivalence classes is developed. First the formal pattern of the neuronal net is presented and then its operation and its direct relationship with the phenomenon of the formation of the equivalence classes and with the derived relationships are explained. Later on, the validation of the pattern is described carrying out several simulations allowing verification of the pattern so it is able to generate relationships not explicitly trained, these results being adjusted to the basic results of this investigation line. These simulations were carried out using a training of classic conditioning and a test phase by means of conditional discriminations. [ABSTRACT FROM AUTHOR]