Ketamine is a noncompetitive N-methyl-D-aspartate (NMDA) antagonist. Given the purported role of the NMDA receptor in long-term potentiation, the primary purpose of the present study was to further understand the dose-related effects of ketamine on memory. The study was also designed to provide information about the relative effects of ketamine on memory versus nonmemory effects and to more fully characterize ketamine's overall pattern and time course of effects. Single intramuscular injections of ketamine (0.2 mg/kg, 0.4 mg/kg) were administered to 18 healthy adult volunteers using a double-blind, placebo-controlled, crossover design. Word lists were used to evaluate episodic memory (free recall, recognition memory, source memory) and metamemory. Working memory, time estimation, psychomotor performance, and subjective effects were assessed repeatedly for 5 hours after drug administration. Ketamine selectively impaired encoding (as measured by free recall) while sparing retrieval, working memory while sparing attention, and digit symbol substitution task speed while sparing accuracy. Ketamine did not significantly impair recognition or source memory, metamemory, or time estimation. There were no hallucinations or increases in mystical experiences with ketamine. Memory measures were less sensitive to ketamine effects than subjective or psychomotor measures. Subjective effects lasted longer than memory and most psychomotor impairments. Ketamine produces selective, transient, dose- and time-related effects. In conjunction with previous studies of drugs with different mechanisms of actions, the observed selectivity of effects enhances the understanding of the pharmacological mechanisms underlying memory, attention, psychomotor performance, and subjective experience.