Systemically administered cholinomimetics or cholinesterase inhibitors can depress behavior in humans and animals, whereas antimuscarinic agents reverse this effect or even produce euphoria. Although these effects have been well documented, the specific brain regions that mediate them remain largely unknown. In the present experiments, muscarinic agonists and antagonists were locally injected into the nucleus accumbens of female Sprague-Dawley rats to test for their effects on behavioral depression in the Porsolt swim test and locomotor activity. Local, microinjections of the drugs in the accumbens elicited behaviors that were similar to the systemic effects reported in other studies. Injection of the non-specific agonist arecoline (40 and 80 microg) dose-dependently inhibited swimming and escape behavior. This may be mediated in part by accumbens M1 receptors because blocking these receptors with the specific antagonist pirenzepine (17.5 and 35.0 microg) did the opposite by increasing swimming. Gallamine (0.13, 0.44, and 0.88 microg), an antagonist at M2 receptors, dose-dependently decreased swimming. Two-way microdialysis suggested that this was in part due to the release of ACh by blocking M2 autoreceptors. Scopolamine, a mixed M1/M2 receptor antagonist, also released ACh but did not decrease swimming, probably because the M1 receptors were blocked; the drug (1.0 microg) increased swimming time, much like pirenzepine. With the exception of arecoline, none of the drugs significantly affected locomotor activity in a photocell cage. Arecoline (40 microg), which had decreased swimming, reduced activity. The present study suggests that muscarinic receptors in the nucleus accumbens can control immobility in the Porsolt swim test. The onset of immobility may depend on the activation of post-synaptic M1 receptors.