Motivational Interviewing (MI) is a goal-oriented psychotherapy focused on addictions, which helps clients (i.e., patients) explore and resolve their ambivalence about the problem at hand. Measuring the counselor's proficiency with MI has typically been assessed via behavioral coding - a time consuming, low technology approach. This paper examines a computational approach to assessing the quality of MI. Specifically, we focus on a particular aspect of the counselor behavior - reflections - believed to be a critical indicator of MI therapy quality. We automatically tag reflection instances in a maximum entropy Markov modeling framework using several lexical features with rich linguistic and contextual information obtained from the session transcripts.

Index Terms: dialog act tagging, behavioral signal processing, motivational interviewing skills code