In the classical conditioning, the truly random control (TRC) procedure, in which conditioned stimuli (CS) and unconditioned stimuli (US) are presented independently, results in an excitatory conditioning in early part of training. The purpose of the present research was to examine an effect of CS intensity upon this initial excitatory conditioning effect. Both in Experiments 1 and 2, the TRC procedure, in which a tone was used as CS and an electric shock as US, was conducted using a technique of conditioned suppression of licking in rats. Suppression to the CS was assessed four times every fifth session of the TRC training. In Experiment 1, using relatively low CS-density, the group trained with a strong CS showed more suppression than the group trained with a moderate CS in the first test session. In Experiment 2, using relatively high CS-density, the group trained with a moderate CS and the group trained with a weak CS showed equivalent suppression in the first test session, but the suppression of the former group was long-lived. Although these results were consistent with the Rescorla-Wagner model as a whole, the interaction between the amount and the duration of the initial excitatory conditioning effect remained to be investigated in future research.