Prepulse inhibition (PPI) refers to the attenuation of startle when a weak prestimulus precedes the startling stimulus. PPI is deficient in several psychiatric illnesses involving poor sensorimotor gating. Previous studies indicate that a1 adrenergic receptors regulate PPI, yet the extent to which these effects are mediated by central vs peripheral receptors is unclear. The present studies compared the effects of intracerebroventricular (ICV) vs intraperitoneal (IP) delivery of several a1 receptor agonists on PPI. Male Sprague-Dawley rats received either cirazoline (0, 10, 25, 50 mg/5 ml), methoxamine (0, 30, 100 mg/5 ml), or phenylephrine (0, 3, 10, 30 mg/5 ml) ICV immediately before testing. Separate groups received either cirazoline (0, 0.25, 0.50, 0.75 mg/kg), methoxamine (0, 2, 5, 10 mg/kg), or phenylephrine (0, 0.1, 2.0 mg/kg) IP 5 min before testing. PPI, baseline startle responses, and piloerection, an index of autonomic arousal, were measured. Cirazoline disrupted PPI; effective ICV doses were approximately six times lower than effective IP doses. Methoxamine disrupted PPI after ICV infusion but failed to affect PPI with IP doses that were up to 30-fold higher than the effective ICV dose. Phenylephrine disrupted PPI with ICV administration, but did not alter PPI after IP injection of even a 20-fold higher dose. None of the ICV treatments altered baseline startle magnitude, but phenylephrine and methoxamine lowered startle after administration of high systemic doses. Piloerection was induced by cirazoline via either route of administration, and by IP methoxamine and phenylephrine, but not by ICV infusion of methoxamine or phenylephrine. These findings indicate that a1 receptor-mediated PPI disruption occurs exclusively through stimulation of central receptors and is dissociable from alterations in baseline startle or autonomic effects.