Experiments with animals have shown that D2 dopamine receptors are involved in regulating prepulse inhibition (PPI) of the acoustic startle reflex (suppression of the reflex response evoked by a loud sound by prior presentation of a low-intensity stimulus). Recently we found that PPI of the human eyeblink startle response could be suppressed by a D2 receptor agonist, bromocriptine, and that this suppression could be reversed by a D2 receptor-blocking neuroleptic, haloperidol. The present work attempted to replicate this finding and to extend it to PPI of the N1/P2 component of the auditory-evoked potential. Eleven healthy males (18-30 years) participated in four sessions in which they received oral doses of placebo, bromocriptine 1.25 mg, haloperidol 3mg and combined treatment with bromocriptine 1.25 mg + haloperidol 3 mg, according to a balanced double-blind protocol. Thirty-minute simultaneous recordings of the electromyographic (EMG) responses of the orbicularis oculi muscle of the right eye and the vertex auditory-evoked potentials took place 120 min after ingestion of haloperidol and/or 90 min after ingestion of bromocriptine. Sound stimuli (1-kHz) were presented in 60 trials separated by variable intervals (mean 25 sec): (i) 40 msec 115 dB ('pulse alone': 20 trials); (ii) 40 msec 85 dB (20 trials); (iii) 40 msec 85 dB, followed after 120 msec by 40 msec 115 dB ('prepulse/ pulse': 20 trials). The amplitudes of the EMG and N1/P2 responses were not altered significantly by any of the treatments. Bromocriptine attenuated PPI of the EMG response significantly, this attenuation being absent following combined haloperidol/bromocriptine treatment. Neither bromocriptine nor haloperidol significantly altered PPI of the N1/P2 complex. Bromocriptine suppressed and haloperidol elevated serum prolactin levels, these changes being absent when the two drugs were given in combination. The results suggest that different mechanisms may be involved in regulating PPI of the eyeblink and the N1/P2 component of the auditory-evoked potential, and that D2 receptors may be involved in the former case, but not the latter.