Generally, compounds discriminated by animals possess psychotropic effects in animals and humans. As with many other drugs of abuse, strength of the ethanol discriminative stimulus is dose related. The majority of studies show that doses close to 1.0 g/kg are close to the minimum at which the discrimination can be learned easily. Substitution studies suggest that anxiolytic, sedative, atactic, and myorelaxant effects of ethanol all play an important role in the formation of its intercoeptive stimulus. Low doses of ethanol produce more excitatory cues, similar to amphetamine-like subjective stimuli, whereas higher doses produce rather sedative/hypnotic stimuli similar to those elicited by barbiturates. Substitution studies have shown that the complete substitution for ethanol may be exerted by certain GABA-mimetic drugs acting through different sites within the GABA(A)-benzodiazepine receptor complex (e.g., diazepam, pentobarbital, certain neurosteroids), gamma-hydroxybutyrate, and antagonists of the glutamate NMDA receptor. Among the NMDA receptor antagonists both noncompetitive (e.g., dizocilpine) and competitive antagonists (e.g., CGP 40116) are capable of substituting for ethanol. Further, some antagonists of strychnine-insensitive glycine modulatory sites among the NMDA receptor complex (e.g., L-701,324) dose-dependently substitute for the ethanol discriminative stimulus. On the other hand, neither GABA-benzodiazepine antagonists nor NMDA receptor agonists produce contradictory effects (i.e., reduce the ethanol discriminative stimulus). There is influence of a particular training dose of ethanol on the substitution pattern of different compounds. For example, 5-HT(1B/2C) agonists substitute for intermediate (1.0 g/kg) but not higher (2.0 g/kg) ethanol training doses. Discrimination studies with ethanol and drugs acting on NMDA and GABA receptors consistently indicate asymmetrical generalization. For example, ethanol is able to generalize to barbiturates and benzodiazepines, but neither the benzodiazepine nor barbiturate response generalizes to ethanol. Only a few drugs are able to antagonize, at least to some extent, the discriminative stimulus of ethanol (e.g., partial inverse GABA-benzodiazepine receptor antagonist Ro 15-4513 and the opioid antagonist naloxone). The ethanol stimulus effect may be increased (i.e., stronger recognition) by N-cholinergic drugs (nicotine), dopaminergic drugs (apomorphine), and 5-HT3 receptor agonists (m-chlorophenylbiguanide). Thus, the ethanol stimulus is composed of the several components, with the NMDA receptor and GABA(A) receptor complex being of particular importance. This suggests that a drug mixture may be more capable of substituting for ethanol (or block its stimulus) than a single compound. The ability of drugs to substitute for the ethanol discriminative stimulus is frequently, although not preclusively, associated with the reduction of voluntary ethanol consumption. The examples of positive correlation are gamma-hydroxybutyrate, possibly memantine and certain serotonergic drugs such as fluoxetine. However, it remains uncertain to what extent the discriminative stimulus of ethanol can be seen as relevant in the understanding of the complex mechanisms of dependence.