Cocaine produces rewarding and locomotor stimulant effects by increasing extracellular dopamine (DA) levels in the terminal areas of the mesolimbic DA system. Our recent in vitro studies have shown that a muscarinic receptor antagonist, trihexyphenidyl (THP) inhibits the binding of a cocaine analogue to the DA transporter at concentrations that are ineffective in inhibiting 3H-DA uptake, suggesting that THP may attenuate the actions of cocaine selectively. The present study examined whether THP could affect conditioned place preference (CPP) for and locomotor stimulant activity of cocaine and methamphetamine (MAP) in mice. Mice were injected with cocaine (10 mg/kg) or MAP (1 mg/kg) in one compartment of the CPP chamber 4 times every second day. On alternate days the animals received saline in the other compartment of the CPP chamber. Pretreatment with THP was made 10 min before cocaine or MAP injection. The CPP score and locomotor activity were assessed using a novel activity monitor, SCANET. Cocaine and MAP produced CPP for the drug-paired compartment. Pretreatment with THP (0.05-5 mg/kg) had no influence on cocaine-induced CPP at any dose tested. In contrast, MAP-induced CPP was completely antagonized by THP at 5 mg/kg, which produced no CPP by itself. Another muscarinic receptor antagonist, scopolamine (SCP, 3 mg/kg) neither caused CPP by itself nor affected the development of cocaine- or MAP-induced CPP. Both THP and SCP enhanced spontaneous, cocaine- or MAP-induced locomotor activity. Though the present conditioning treatments failed to develop locomotor sensitization to cocaine, THP, but not SCP, acted cooperatively with cocaine to develop locomotor sensitization. The development of locomotor sensitization to MAP was retarded by SCP but was not affected by THP. These results suggest that, contrary to our anticipation, THP has a unique characteristic of specifically counter-acting the rewarding properties of MAP via a non-cholinergic (muscarinic) mechanism.