A topological data analysis method for revealing dynamic changes in psychotherapy microprocesses.

Area of Science:

• Psychology
• Data Science
• Complexity Science

Background:

• Understanding moment-to-moment therapeutic change is crucial for improving psychological interventions.
• Existing methods often fail to capture the complex dynamics of therapeutic processes.
• Dynamical systems theory provides a framework for modeling these shifts, but quantitative methods are limited.

Purpose of the Study:

• Introduce Temporal Mapper, a novel topological data analysis (TDA) method.
• Apply Temporal Mapper to psychotherapy microprocess data to analyze interpersonal behaviors and alliance ruptures.
• Demonstrate Temporal Mapper's ability to quantitatively link system dynamics (attractors, transitions) with empirical data.

Main Methods:

• Utilized topological data analysis (TDA) to develop Temporal Mapper.
• Applied Temporal Mapper to high-dimensional psychotherapy microprocess data.
• Represented detected dynamic features as attractor transition networks.

Main Results:

• Therapist warmth was found to stabilize dyadic interpersonal states both within and across therapy sessions.
• Confrontation ruptures stabilized dyadic interpersonal states within sessions.
• Confrontation ruptures diversified and destabilized dyadic interpersonal states across sessions.

Conclusions:

• Temporal Mapper offers a generalizable approach for uncovering fine-grained dynamic patterns in complex systems.
• The method can analyze multimodal psychotherapy process data to identify mechanisms of change.
• Findings inform the development of more effective, system-level psychological interventions.