Two experiments, concerned with signal-produced variation of unconditioned response (UR) amplitude, evaluated the roles of US intensity and response measure in determining when signaling the US results in a conditioned diminution or a conditioned facilitation of responding. In Experiment 1, rabbits received discrimination training, in an eye-blink-conditioning preparation, with a 1-mA, 50-msec shock US. In testing, preceding the US by SC+ facilitated the eye blink response compared with preceding the US by CS- or neither CS. In Experiment 2, subjects received discrimination training with a 5-mA US and were tested with 1-, 2-, and 5-mA US. During testing, subjects' eye blink responses (for which robust CRs were observed) and gross body movement responses (for which no CRs were observed) were simultaneously recorded. for the eye blink response, preceding the US by CS+ facilitated responding to the 1-mA US, produced negligible differences in responding to the 2-mA US, and diminished responding to the 5-mA US, compared with preceding the US by CS- or neither CS. For the movement response, preceding the US by CS+ diminished responding at all three US intensities, with the decremental effects increasing with increases in US intensity. The results of Experiments 1 and 2 are discussed in terms of the dual effects of CS+: the diminution of US processing and the contribution of the conditioned response to the measured response.