The purpose of this study was to characterize the reinforcing, subjective, and psychomotor effects of nitrous oxide (N2O) in healthy volunteers who were given different amounts of information regarding the drugs they were being administered in the experiment. A choice procedure was used in which subjects first sampled a placebo and a given concentration of N2O, and then chose between the two. N2O concentration varied across the four-session experiment from 10-40%. Besides choice, subjective and psychomotor effects served as dependent measures. In the INFORMED group (n = 11), subjects were told at the beginning of each sampling trial what concentration of N2O they were inhaling or whether they were inhaling 100% oxygen (placebo). They were also informed about the prototypic effects of N2O (e.g. tingling or numbing, euphoria, dysphoria) and oxygen (e.g. no discernible effects). In the NON-INFORMED group (n = 11), subjects were only told at the beginning of each sampling trial that the drugs they would be inhaling came from one of six classes of drugs. Thirty percent N2O was chosen by a significantly higher proportion of subjects than expected by chance in the INFORMED group, but not in the NON-INFORMED group. Further, the probability of choosing 20-40% N2O was higher in the INFORMED group than in the NON-INFORMED group. Subjective effects of N2O were not affected by the information manipulation. Psychomotor performance at the highest N2O concentration tested (40%) was impaired to a greater extent in the NON-INFORMED than in the INFORMED group. We conclude that the reinforcing effects of N2O, and perhaps the impairing effects, can be modulated by telling subjects beforehand that they are inhaling N2O and what effects they might be expected to experience from the drug. (C) 1997 Elsevier Science Ireland Ltd.