The ICSA tests followed procedures previously described (Ding et al, 2009b (link); Rodd-Henricks et al, 2000 (link)). Briefly, rats were placed into operant conditioning chambers equipped with two levers, one active and one inactive. The active lever was connected to an isolated pulse stimulator (Model 2100) from A-M systems, Inc (Varlsborg, WA) controlled by an operant conditioning control system. The A-M pulse stimulator was connected to two electrodes that were immersed in a solution-filled cylinder container equipped with a 28-gauge injection cannula. Each response on the active lever (FR1 schedule of reinforcement) activated the pulse stimulator that produced a 5-sec infusion current between the electrodes, resulting in an infusion of 100-nl solution into the p-VTA. Each infusion was followed by a 5-sec timeout period. During both the infusion and timeout periods, responses on the active lever were recorded but did not produce further infusions. The responses on the inactive lever were recorded but did not result in any infusions. The assignment of active and inactive lever was counterbalanced among rats. There were a total of seven sessions conducted with each session being 4-hr long and 48–72 hr between sessions.