"It sounds like...": A natural language processing approach to detecting counselor reflections in motivational interviewing.

The dissemination and evaluation of evidence-based behavioral treatments for substance abuse problems rely on the evaluation of counselor interventions. In Motivational Interviewing (MI), a treatment that directs the therapist to utilize a particular linguistic style, proficiency is assessed via behavioral coding-a time consuming, nontechnological approach. Natural language processing techniques have the potential to scale up the evaluation of behavioral treatments such as MI. We present a novel computational approach to assessing components of MI, focusing on 1 specific counselor behavior-reflections, which are believed to be a critical MI ingredient. Using 57 sessions from 3 MI clinical trials, we automatically detected counselor reflections in a maximum entropy Markov modeling framework using the raw linguistic data derived from session transcripts. We achieved 93% recall, 90% specificity, and 73% precision. Results provide insight into the linguistic information used by coders to make ratings and demonstrate the feasibility of new computational approaches to scaling up the evaluation of behavioral treatments.

Outcomes of couples with infidelity in a community-based sample of couple therapy.

Infidelity is an often cited problem for couples seeking therapy, but the research literature to date is very limited on couple therapy outcomes when infidelity is a problem. The current study is a secondary analysis of a community-based sample of couple therapy in Germany and Austria. Outcomes for 145 couples who reported infidelity as a problem in their relationship were compared with 385 couples who sought therapy for other reasons. Analyses based on hierarchical linear modeling revealed that infidelity couples were significantly more distressed and reported more depressive symptoms at the start of therapy but continued improving through the end of therapy and to 6 months posttherapy. At the follow-up assessment, infidelity couples were not statistically distinguishable from non-infidelity couples, replicating previous research. Sexual dissatisfaction did not depend on infidelity status. Although there was substantial missing data, sensitivity analyses suggested that the primary findings were not due to missing data. The current findings based on a large community sample replicated previous work from an efficacy trial and show generally optimistic results for couples in which there has been an affair.