Cocaine for example, while actually causing a decrease in DA neuron firing rate, is also able to induce AMPAR redistribution. One possibility for this result is that the increased dopamine concentration is responsible for the induction of this plasticity. Our data suggest that DA signaling within the VTA is driving AMPAR redistribution. First, other reports have used fast scan voltammetry to show that similar optogenetic stimulation protocols produce large DA transients in VTA target regions. Second, a previous study has shown that a change in the AMPA/NMDA ratio, induced following administration of addictive drugs, was blocked by application of a D1-like receptor antagonist. Our data with local VTA infusion of the same antagonist confirms this ASP1517 requirement for D1 signaling. This observation is also of interest in the context of recent evidence that some DA neurons co-release glutamate. Further experiments will have to establish the necessity of DA neurons activation by inhibiting the DA neurons while giving a drug or testing for occlusion if the effect of the stimulation after drug exposure. Since previous pharmacological and genetic manipulations also demonstrated the need for NMDARs on DA neurons, intrinsic glutamatergic transmission may also be required and future studies will have to identify the locus and hierarchy of the convergence of DA- and NMDA-signaling. As drug-triggered AMPAR redistribution has also been induced in a VTA slice preparation, this implies a mechanism restricted to the circuitry within the VTA. Indeed, bursting of DA neurons is also particularly efficient at driving DA release within the VTA. However, whether or not reciprocal connections between glutamatergic or GABAergic nuclei and DA VTA neurons were potentiated with this protocol cannot be ruled out. Indeed it is possible that adaptations in the NAc may have an indirect effect on the VTA via the strong back-projection of this nucleus to the midbrain. A previous report has shown that stimulation of DA neurons, albeit with a different protocol, leads to behavioral conditioning, such as conditioned place preference in the NMDAR-mutant mice where AMPAR redistribution was absent. However these mice did show reduced reinstatement and cue-induced cocaine seeking. Our finding that selective stimulation of DA VTA neurons leads to AMPAR redistribution therefore provides strong evidence that increased DA neuron activity is capable of modifying the network at the synaptic level. Given that addictive drugs are chemically very diverse and each has a distinct molecular target, it is surprising that they induce symptoms that are indistinguishable. Our study provides proof of principle for an early point of convergence in the function of the DA neurons of the VTA. The release of mesolimbic DA seems critical for the induction of a form of synaptic plasticity that predicts long-term adaptations in the neural reward circuits.