DESCRIPTION (provided by applicant): Cognitive Behavioral Therapy (CBT) is an established approach for alcohol use disorders, yet it is hindered by a number of obstacles to dissemination and effective implementation in clinical practice, which include a limited understanding regarding how it exerts its effects (i.e., mechanisms of action). This proposal aims to evaluate a novel computer-based cognitive behavioral therapy program for alcohol use disorders (CBT4CBT) designed to address these obstacles. The proposed R01 project (which follows the completion of R21 AA 021405; Carroll, PI) will evaluate the efficacy of CBT4CBT or clinician-delivered CBT in comparison to standard treatment as usual (TAU) at reducing alcohol use among 180 outpatient treatment seekers. Treatments will be delivered over the course of 8-weeks and will include a 6-month follow-up period to assess durability of treatment effects. Multiple behavioral, experiential, and cognitive measures will be used to develop a more complete understanding of the mechanistic processes by which CBT works to reduce alcohol use. The specific aims of the project are as follows: 1. Evaluate the efficacy of CBT4CBT or clinician-delivered CBT relative to TAU at reducing alcohol use through an 8-week randomized trial, with 6-month follow-up. We expect either form of CBT will be more effective than TAU at increasing the percentage of days abstinent (PDA) during treatment and follow-up. 2. Evaluate the extent to which CBT's putative mechanisms act as a mediator of treatment effect on reducing alcohol use in both forms of CBT relative to TAU. We expect indices reflecting the acquisition and implementation of skills will mediate the effect of CBT on alcohol use (i.e., PDA). Exploratory aims will include the evaluation of levels of cognitive function as moderators of CBT's direct and indirect effects, as well as exploration of differences across CBT formats (clinician-delivered vs. computer- delivered) with respect to the mechanisms of action. This will be the first adequately powered randomized controlled trial to evaluate the effectiveness of a computer-delivered version of CBT as a virtual `stand-alone' intervention for alcohol use disorders, which could have clear implications for making this cost effective form of CBT more broadly accessible. Furthermore, the inclusion of traditional clinician-delivered CBT within the same trial will allow for novel examination of CBT's mechanisms, as the active ingredients can be evaluated with and without `relational' factors (e.g., therapist empathy, warmth, therapeutic alliance) that typically contribte to outcomes.