There is preliminary evidence that the anticonvulsant medication Zonisamide (ZON) may be an effective, well-tolerated treatment for alcohol use disorder (AUD). However, further evaluation of its efficacy for treating patients with AUD is needed, and much remains unknown about ZON’s therapeutic mechanisms. The present study aimed to evaluate the efficacy and tolerability of ZON in a double-blind, placebo-controlled, randomized trial. Eighty-one adults (ages 21-65) diagnosed with AUD were randomly assigned to receive either ZON (at a target dose of 400 mg/d) or a pill placebo over 12 weeks, followed by a two-week taper. All participants also received a computerized alcohol reduction program, Take Control (TC). Primary drinking outcomes were average daily drinks, percentage drinking days, and percentage heavy drinking days. Further, we evaluated changes in AUD clinical severity and performance on neuropsychological measures. For both groups, drinking outcomes generally decreased, as did AUD clinical severity, though group differences were not statistically significant. Neuropsychological testing performance was similar for both groups at baseline; however, at post-treatment, participants in the ZON group demonstrated poorer working memory and lower performance on verbal fluency tests compared to the placebo group, and these differences were statistically significant with moderate-large effect sizes. One serious adverse event was reported among individuals receiving ZON. Study findings indicate that ZON combined with TC does not demonstrate superior effectiveness for reducing average daily drinks in this clinical sample with principal AUD compared to placebo and TC, and treatment with ZON may be associated with reduced neurocognitive performance over time.