Why is there such a gap between the evidence base for effective drug treatment and its dissemination in the community? Among behavioral interventions, Motivational Interviewing (MI) should be well positioned for high-impact dissemination: There is an international organization devoted to the training and dissemination of MI (The Motivational Interviewing Network of Trainers [MINT]), and MI developers have been at the forefront of research on dissemination. In addition, MI mechanisms are well specified and many are considered basic building blocks of behavioral interventions (e.g., empathy, use of reflections and open questions).1 Empathy, a central MI component, has a strong research tradition in basic science and is among the most consistent predictors of positive outcome in psychotherapy across treatments and disorders.2 At present, quality control in MI relies on ongoing supervision and feedback to therapists in the community. However, measuring therapist behavior is synonymous with behavioral coding by humans, which is not feasible in community settings3. Accordingly, there is no feasible mechanism for evaluating the practice of evidenced based treatments in clinical settings. Advances in linguistic processing have brought a technology for behavioral coding within reach. Specifically, in the basic science and engineering fields, there are tools that combine acoustic and semantic information into predictive models of human-defined knowledge (e.g., empathy codes).4-8 The current proposal brings together a highly interdisciplinary team to produce a viable technology for generating feedback on therapist empathy without human intervention. This technology may eventually be packaged and disseminated in the community. As an initial step, we will apply a sequence of linguistic processing approaches to develop a process for predicting therapist empathy from audio files obtained from MI sessions (AIM 1).9 Next, we will run a feasibility study to determine the performance of the coding technology with new data obtained from MI sessions conducted by novice and experienced therapists (AIM 2). During this evaluation we will troubleshoot issues related to audio quality and speech recognition as well as the process of feedback. In addition, we will evaluate the agreement of computer-based codes with human coding of new sessions. Drug abuse and dependence represents an incredible societal burden. The current research will develop an innovative process for enhancing the ability disseminate and maintain quality control in the behavioral treatment of addictions.

Public Health Relevance

Currently, there is no objective/scalable method for evaluating the practice of evidence- based treatments like Motivational Interviewing (MI) in clinical settings Computer-based assessments of MI fidelity have the potential to scale-up in such a manner that the provision of performance-based feedback to therapists in the community is feasible. The current proposal will apply linguistic processing approaches to develop and test the feasibility of an initial technology for coding a central component of MI - provider empathy and its specific behavioral indicators. !

National Institute of Health (NIH)
National Institute on Drug Abuse (NIDA)
Planning Grant (R34)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Aklin, Will
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Utah
Schools of Education
Salt Lake City
United States
Zip Code
Hirsch, Tad; Merced, Kritzia; Narayanan, Shrikanth et al. (2017) Designing Contestability: Interaction Design, Machine Learning, and Mental Health. DIS (Des Interact Syst Conf) 2017:95-99
Gaut, Garren; Steyvers, Mark; Imel, Zac E et al. (2017) Content Coding of Psychotherapy Transcripts Using Labeled Topic Models. IEEE J Biomed Health Inform 21:476-487
Imel, Zac E; Caperton, Derek D; Tanana, Michael et al. (2017) Technology-enhanced human interaction in psychotherapy. J Couns Psychol 64:385-393
Xiao, Bo; Huang, Chewei; Imel, Zac E et al. (2016) A technology prototype system for rating therapist empathy from audio recordings in addiction counseling. PeerJ Comput Sci 2:
Xiao, Bo; Imel, Zac E; Georgiou, Panayiotis et al. (2016) Computational Analysis and Simulation of Empathic Behaviors: a Survey of Empathy Modeling with Behavioral Signal Processing Framework. Curr Psychiatry Rep 18:49
Tanana, Michael; Hallgren, Kevin A; Imel, Zac E et al. (2016) A Comparison of Natural Language Processing Methods for Automated Coding of Motivational Interviewing. J Subst Abuse Treat 65:43-50
Lord, Sarah Peregrine; Sheng, Elisa; Imel, Zac E et al. (2015) More than reflections: empathy in motivational interviewing includes language style synchrony between therapist and client. Behav Ther 46:296-303
Imel, Zac E; Sheng, Elisa; Baldwin, Scott A et al. (2015) Removing very low-performing therapists: A simulation of performance-based retention in psychotherapy. Psychotherapy (Chic) 52:329-36
Imel, Zac E; Steyvers, Mark; Atkins, David C (2015) Computational psychotherapy research: scaling up the evaluation of patient-provider interactions. Psychotherapy (Chic) 52:19-30
Xiao, Bo; Imel, Zac E; Georgiou, Panayiotis G et al. (2015) ""Rate My Therapist"": Automated Detection of Empathy in Drug and Alcohol Counseling via Speech and Language Processing. PLoS One 10:e0143055

Showing the most recent 10 out of 13 publications