Signaled tailshock is perceived as similar to a stronger unsignaled tailshock: implications for a functional analysis of classical conditioning.

Water-deprived rats given fixed-electrode, variable-intensity tailshock at random times rated each trial by pressing either a "high-aversiveness" or "low-aversiveness" lever in order to obtain water. Trials on which a warning signal preceded tailshock resulted in more "high-aversiveness" leverpressing than did otherwise equivalent unsignaled trials. The magnitude of this effect increased and decreased as a function of several parameters including signal-shock interval, signal duration, and range and absolute value of shock intensities but was never reversed despite efforts to achieve such a reversal. Variation in the size of the effect as a function of signal parameters as well as lick suppression scores indicated that the signal had acquired aversive characteristics, which suggests that the effect of the signal on lever choice was due largely to the aversiveness of the signal summating with the aversiveness of the tailshock. Several hypotheses concerning factors that might have either masked or prevented classically conditioned preparatory responses elicited by the signal from reducing tailshock aversiveness were tested and rejected. Despite the greater aversiveness of the signaled condition, when given the choice of receiving or not receiving the signal, the animals displayed a preference for signaled tailshock. Implications for the role of preparatory responding both in the preference-for-signaled-shock phenomenon and in classical conditioning are discussed.