Soichiro Matsuda

Soichiro Matsuda

he/his

Assistant Professor, University of Tsukuba | Behavioral Design Lab

From Tool to Skill: Monitoring Thought-Record Structural Quality as a Proximal Indicator in CBT Training Using a Concurrent Multiple Baseline Single-Case Design

Published in Single Case in the Social Sciences, 2026

Summary

This study examined whether the structural quality of thought-record entries—the clear differentiation among situation, thought, and emotion—improved following a brief, seven-session CBT-based interviewing intervention. Three nonclinical Japanese undergraduates participated in a concurrent multiple-baseline single-case design, with thought-record quality evaluated continuously using a six-item rubric.

Mean thought-record quality was higher during intervention than baseline for all participants, with the largest improvement observed for the participant who began with the lowest level of structural differentiation. State and trait anxiety and negative affect also decreased across participants, while social validity ratings were high. Depressive symptoms were minimal at baseline, and positive affect showed variable changes.

The study shifts the thought record from being treated only as an intervention tool to also being a measurable target of change. Continuous assessment of its structural quality provides a proximal, low-burden indicator of CBT skill acquisition that can complement more distal symptom measures.

From an HCI and digital mental health perspective, the behaviorally anchored scoring rubric provides a transparent target for automated or semi-automated assessment. This creates a foundation for systems that monitor longitudinal changes in thought-record use, provide structured feedback, and evaluate human–AI support without relying solely on symptom-level outcomes.

Research line

  • Measuring: continuous measurement of thought-record structural quality as a proximal indicator of skill acquisition
  • Digital Mental Health: translating a core CBT practice into a longitudinally measurable interaction target
  • Single-Case Experimental Design: evaluating individual-level change using staggered intervention onset across participants
  • Human–AI Support: establishing a transparent behavioral rubric that can serve as a benchmark for automated scoring and semi-automated feedback
  • Skill Acquisition: distinguishing improvement in how a CBT tool is used from broader changes in psychological outcomes

Matsuda, S., Karasawa, H., & Sugawara, D. (2026). From tool to skill: Monitoring thought-record structural quality as a proximal indicator in CBT training using a concurrent multiple baseline single-case design. Single Case in the Social Sciences, 3(1), 56–67. https://doi.org/10.15763/issn.1936-9298.2026.3.1.56-67

Recommended citation: Matsuda, S., Karasawa, H., & Sugawara, D. (2026). From tool to skill: Monitoring thought-record structural quality as a proximal indicator in CBT training using a concurrent multiple baseline single-case design. *Single Case in the Social Sciences, 3*(1), 56–67. https://doi.org/10.15763/issn.1936-9298.2026.3.1.56-67
Download Paper