[Variable cycle avoidance program: systematic variation of the probability of a programmed noxious stimulus]

Six albino rats were exposed in three groups to different values of T (a repetitive time cycle at the end of which an electric shock occurs with a specified probability in absence of a first response), (V.g.: P(E-/R) = .75 and .10) on different weeks. When occurs at least one response on each T cycle it has a probability to delete the shock equal to P(E-/R) that changed on different weeks from .50 to .10. In this way, as the Bayesian probability system indicate for exclusive events, Pk = P(E-/R) - [P(E-/R) x P(E-/R)] was the effective probability of shocks when at least one response occur on each T cycle and Pm = P(E-/R) - [P(E-/R) x P(E-/R) x P(1rst Response)] is the modulated probability of shocks when a first response occur on some T cycles. The results indicated that T/Pk (the mean interval between shocks before the experimental session) nor T/Pm (The mean interval between shocks after the experimental session when the subject responds), or its inverse, are reductional continua that explains monotonically ordered changes of the response rate or the mean total number of responses. As a consequence T and P must be dissociated. On the other hand, corrected response rate (Total responses minus Total first responses/minute) is a direct and monotonically increased function of Pk and of 1/T. In this sense is the increased probability with which noxious stimuli impinge on the behavior stream and the time reduction between stimuli, that separated, sustain responding.