What is it about?
Clinical supervision helps therapists make decisions about client care, but little is known about what happens in these conversations. We used machine learning to identify and classify topics and decisions discussed between therapists and supervisors.
Featured Image
Why is it important?
A machine-learning model that considers conversational context was able to classify supervision discussions effectively. This matters because having people code conversations by hand requires substantial time and effort. Machine learning could make it possible to examine many more supervision conversations, helping researchers identify patterns and, eventually, support supervision training and quality improvement.
Perspectives
I was especially excited to learn from our natural language processing colleagues throughout this project. Together, we began to see new possibilities for studying clinical supervision—a central but often difficult-to-observe part of therapist training and support. I hope this work encourages others to consider how machine learning can complement, rather than replace, the expertise of supervisors and clinicians as we work to strengthen care for young people and families.
Kimberly Becker
University of South Carolina
Read the Original
This page is a summary of: Toward scalable detection of clinical reasoning in supervision: A machine learning feasibility study., Psychotherapy, September 2026, American Psychological Association (APA),
DOI: 10.1037/pst0000638.
You can read the full text:
Contributors
The following have contributed to this page







