There is a growing body of literature correlating changes in spine density and morphology with changes in brain function and behavior. Altered spine density has been reported in many disease states including schizophrenia and depression. Therefore, spine density could be a useful index to measure the therapeutic profile of psychoactive drugs. In the same way, the signal transduction systems that induce synaptogenesis and maintain spine density represent promising targets for the development of new treatments for mental illness. Our laboratory has studied the role of the NMDA-type glutamate receptor in the regulation of spine density, with a focus on the mechanisms by which altered spine density could translate into psychiatric disease states. NMDA receptors have been implicated in a number of psychiatric diseases as a causal agent or a target for therapeutic intervention. For example, NMDA receptor hypofunction is thought to contribute to the symptoms of schizophrenia, a disease in which reductions in spine density have been reported. Using a genetic mouse model, we have discovered that sustained NMDA receptor dysfunction causes an age-dependent reduction in spine density. Our findings show that NMDA receptor signaling is more important for synapse maintenance than synapse formation. Furthermore we found a temporal developmental correlation between spine loss and several behavioral abnormalities. We sought to determine the molecular mechanisms by which NMDA receptor dysfunction leads to spine loss. Our investigation revealed reductions in the levels of the Rho GTPase Rac1 and its downstream effectors. Therefore, RhoGTPase signaling represents a novel pathway to modulate spine density and improve psychiatric conditions.