Twelve subjects were tested with D-amphetamine, yohimbine, clonidine, and a placebo on a task with two levels of stimulus and two levels of response complexity. The purpose of this study was to test the hypothesis that noradrenergic drugs affect early stimulus processes. D-amphetamine speeded reaction time (RT), clonidine slowed it, and yohimbine had no effect. D-amphetamine and yohimbine decreased N1 latency and clonidine increased it. D-amphetamine and yohimbine decreased P3 latency and clonidine increased it but, in each case, only when latency estimates were based on single trials, not on averages. D-amphetamine's effect on RT, not P3, as measured by the average, is consistent with previous results. Single trial measures appear more sensitive. Speeding of N1 and single-trial P3 data indicate that noradrenergic drugs affect processing of early (visual) information. D-amphetamine's speeding of single-trial P3 estimates was attributed to its noradrenergic actions. Yohimbine's speeding of P3 without changing RT is consistent with neural net (parallel) simulations but not with a serial model. These findings support the assumption that different neurotransmitters modulate specific cognitive processes.