Rewarding electrical brain stimulation (EBS) can serve as a discriminative stimulus. It has long been suspected that this discriminative property reflects the subjective intensity of the rewarding effect. In turn, the counter model of spatiotemporal integration attributes the subjective reward intensity produced by a stimulation train of fixed duration to the number of firings triggered in the directly activated neural substrate. If the discriminative property of rewarding EBS depends on subjective reward intensity, it should also obey the counter model. To determine whether this is so, we used rewarding EBS of the medial forebrain bundle in rats as a discriminative stimulus to guide responding for a sucrose reward, and we varied the amplitude and frequency of the pulses constituting the stimulation train. High- or low-EBS trains, signaled which of two levers would deliver the sucrose reward. On generalization trials, the train serving as the discriminative stimulus was intermediate in strength to the high- and low-EBS trains. Responding on the two levers varied systematically as a function of stimulation strength. The order (ascending, descending or random) in which the strength of the discriminative stimulus varied was without discernible effect. In most rats, similar curves relate discriminative performance to stimulation strength regardless of whether pulse amplitude or pulse frequency was varied. Thus, the counter model was largely successful in accounting for the effect of varying stimulation strength on discrimination performance. This result is consistent with the hypothesis that the subjective intensity of the rewarding effect is the basis for the discrimination.