Bitonic dose-response functions for reinforcing and self-reported effects of nitrous oxide in humans

The reinforcing and self-reported effects of nitrous oxide (10%, 30%, and 50% N(2)O in O(2)) were examined in 13 humans. During each of three sessions, subjects sampled one dose of N(2)O and 100% O(2) (placebo) for 10 min each, separated by 30-min recovery periods. The agents were identified by letter code, and later in the session, subjects chose nine times, once every 5 min, among N(2)O (e.g., "Agent A"), placebo (e.g., "Agent B"), or "neither" (also 100% O(2), identified as "drug-free air"). Self-reported and psychomotor effects were measured at various times. Dose-response functions varied across subjects and included bitonic, monotonic increasing, monotonic decreasing, U-shaped, and flat dose-response functions for reinforcing and/or self-reported effects. For subjects who showed bitonic reinforcing effects, the descending limb of the dose-response function could not be attributed to behavioral impairment. This study replicates previous studies showing dose-dependent effects of N(2)O, as well as between-subject variability in N(2)O effects. Bitonic dose-response functions for some subjects extend the generality of the phenomenon of bitonicity of drug effects to N(2)O effects in humans.