This study examined the effects of pretreatment with phencyclidine (PCP), a selective N-methyl-D-aspartate (NMDA) antagonist, on behavioral and physiologic responses of the rat to experimental traumatic brain injury (TBI). For the behavioral experiments, rats were administered either saline or PCP (1.0, 2.0, or 4.0 mg/kg, intrapentoneally [IP] 15 min before TBI. Rats were ventilated as necessary following injury. The duration of acute suppression of several reflexes (pinna, corneal, righting, and flexion) and responses (escape, head support, and spontaneous locomotion) was recorded for up to 70 min after trauma. Longer-term behavioral assessments (beam walking, beam balance, inclined plane, ambulatory activity, and body weight) were made for up to 10 days after trauma. PCP did not significantly alter the duration of acute behavioral suppression. At a dosage of 1.0 mg/kg, PCP significantly attenuated all long-term deficits except beam walking. Maximal protection against beam walking deficits was provided by the 4.0 mg/kg dosage of PCP. Sixty-three percent of saline-treated animals died within 10 days after injury. For rats pretreated with 1.0, 2.0, and 4.0 mg/kg of PCP, 40%, 23%, and 33% died, respectively. In physiologic experiments, pretreatment with 4.0 mg/kg of PCP (IP) 15 min before injury did not significantly affect systemic cardiovascular responses, plasma glucose levels, or blood gas levels observed within 30 min after injury. While the possibility of effects mediated by other neurotransmitter systems cannot be excluded, these data suggest that NMDA agonist-receptor interactions contribute to the pathophysiology of brain injury. In addition, neural mechanisms that mediate transient unconsciousness following moderate levels of head injury may differ from mechanisms that mediate more persistent neurologic deficits.