Despite considerable scientific effort in the last two decades, no medication has proved to be an effective treatment for cocaine dependence. Cocaine's rewarding effects are mediated primarily through mesolimbic dopamine (DA) pathways. Yet, experiments that have focused on developing medications that either inhibit midbrain DA release or produce post-synaptic DA receptor blockade have shown that these are not effective treatments for cocaine dependence. Therefore, we propose to test a different strategy to developing a medication for treating cocaine dependence: Is opposition of midbrain DA release alone sufficient, or can we more reliably control DA effects by contemporaneously modulating DA functional expression? Expression of midbrain DA-associated reward may be mediated through inhibition of gamma-aminobutyric acid (GABA), via efferents that project from the nucleus accumbens to the cortex and that are themselves under the tonic control of excitatory amino acid (EAA) pathways. Thus, it is reasonable to hypothesize that a pharmacological agent that facilitates mesocortical GABAergic function and inhibits the action of EAAs should more reliably diminish cocaine's rewarding effects by inhibiting the expression of midbrain DA function in addition to decreasing midbrain DA release. The promise of this novel approach is exemplified by our recent proof-of-concept demonstration that topiramate (a fructo-pyranose derivative that diminishes DA functional expression) is effective at reducing craving and alcohol consumption among alcohol-dependent individuals. Next, we will expand the test of our hypothesis by determining in a randomized clinical trial (RCT) whether topiramate significantly reduces cocaine's rewarding effects associated with abuse liability. In this RCT, male and female cocaine dependent individuals will be randomized to receive either topiramate (up to 300 mg/day) or placebo (N = 90 subjects/group x 2 groups = 180) as an adjunct to standardized weekly Cognitive Behavioral Therapy for 12 weeks. Participants also will be followed up at 2 weeks and at 1,2, and 3 months after the end of the 12-week treatment period. The primary specific aims of this study are to test two predictions of our hypothesis: 1) Topiramate will be superior to placebo at increasing the maximum number of cocaine-free (abstinent) days (assessed by self-report of use and urine assays for benzoylecgonine, the major metabolite of cocaine), and 2) Topiramate will be superior to placebo at decreasing cocaine craving, and these reductions in cocaine craving will be associated with decreased cocaine intake. We also will test the additional secondary predictions that: 3) Topiramate, compared with placebo, will be associated with an improvement in psychosocial functioning as exemplified by: a) Improved general well-being; b) Social Functioning, and c) Enhanced Quality of Life, and d) that these improvements in psychosocial functioning will be correlated with decreased cocaine intake. This study supports NIDA's mission to develop effective medications for the treatment of cocaine dependence. It also will provide evidence for or against the above hypothesis, and thus will improve our understanding of the fundamental neurobiological mechanisms that control and modulate cocaine dependence.