Towards the Computational Modeling and Analysis of Arts-Based Therapy

David Harel (Weizmann), Billie Sandak

The mechanisms by which arts-based therapy exerts its therapeutic and psychosocial effects are only partially understood. A typical clinical setting consists of the therapist, the patient and the creation work itself, e.g., the artwork or the music played. The dynamics of this rich environment is typically interpreted subjectively and described verbally. We introduce a computational method for the rigorous tracking, analysis and documentation of such dynamics. The method utilizes the language of Statecharts and tools for its execution, and enables one to carry out exploratory, hypotheses-generating and knowledge discovery investigations. We have tested the method in a small non-therapeutic artwork setup, in a larger music improvisation one, and in an initial music therapy setup. In all cases, interesting emergent properties were discovered.