The present study assessed the acute effects of p.o. administered ethanol (0, 22, 45, and 67 g) and diazepam (0, 10, 20 and 40 mg) in normal adult humans responding under a multiple schedule of acquisition and performance of 10-response sequences. In each component of the multiple schedule subjects were required to complete a different sequence of 10 responses in a predetermined order using three keys of a numeric keypad. In the acquisition component a new response sequence had to be acquired each session. In the performance component the response sequence remained the same from session to session. The higher doses of ethanol (67 g) and diazepam (20 and 40 mg) increased overall percent errors above placebo levels; the lower doses of these compounds had no significant effect. The increases in percent errors were selective across the two schedule components. Ethanol increased percent errors significantly only in the acquisition component. Diazepam increased percent errors significantly in the acquisition component at a lower dose (20 mg) than was necessary to increase errors in the performance component (40 mg). Ethanol (67 g) and diazepam (40 mg) decreased overall response rates as an orderly function of dose. In contrast to the percent errors measure, the effects on response rates were not significantly different across the performance and acquisition components. Peak effects on percent errors and response rates varied between 30 and 70 min and 45 and 80 min postdrug for ethanol and diazepam, respectively, and the duration of effects of the higher doses was greater than 3 hr.(ABSTRACT TRUNCATED AT 250 WORDS)