We applied continuous, on-line and real-time spectral analysis of electroencephalographic (EEG) signals and microdialysis to evaluate the possible participation of noradrenergic neurotransmission at the medial prefrontal cortex (mPFC) in EEG desynchronization induced by cocaine. Male Sprague-Dawley rats that were under chloral hydrate anesthesia were used. Intravenous administration of cocaine (1.5 or 3.0 mg/kg) dose-dependently induced EEG desynchronization, as represented by a decrease in root mean square (RMS) and an increase in mean power frequency (MPF) value of the EEG signals. Power spectral analysis further revealed that whereas both doses of cocaine promoted a reduction in the alpha (8-13 Hz), theta (4-8 Hz), and delta (1-4 Hz) components, the lower dose of cocaine decreased, and the higher dose increased the beta band (13-32 Hz). Microdialysis data indicated an elevation in extracellular concentration of norepinephrine at the mPFC that paralleled temporally and correlated positively with the maximal effect of cocaine on EEG activity. Bilateral microinjection of the selective noradrenergic neurotoxin, DSP4 (50 micrograms), or equimolar concentration (500 pmol) of the alpha 1-adrenoceptor antagonist, prazosin, or alpha 2-adrenoceptor antagonist, yohimbine, into the mPFC significantly blunted the decrease in delta component (prazosin) or both delta and theta components (DSP4 or yohimbine) of EEG activity by the lower dose of cocaine. On the other hand, the same pretreatments appreciably antagonized the increase in beta band by cocaine at 3.0 mg/kg. The potency of the antagonism by yohimbine, however, was higher than prazosin. These results suggest that cocaine may elicit EEG desynchronization via noradrenergic neurotransmission, and that alpha 2-adrenoceptors, and to a lesser extent, alpha 1-adrenoceptors, at the mPFC may be involved in the subtle dose-dependent changes in individual EEG spectral components.