Complexity, Fractal Patterns, and Interpersonal Dynamics: An Empirical Test of the 5-R Model in Group Therapy
Date of Award
Dissertation - Restricted
Doctor of Philosophy (PhD)
Since the 1930's and 40's, substantial gains have been made in understanding small group processes. Yet, this field has undergone a substantial degree of fragmentation over time, particularly among the investigations of group therapy, family systems theory and small group theories within the field of social psychology. At the same time, methodological limitations have restricted empirical investigations of group dynamics within each of these domains. The current study was designed to test an integrated model of small group processes (the 5-R model) developed from theoretical concepts from family systems, small groups and nonlinear dynamical systems. The conversation of one group therapy session was analyzed using orbital decomposition. An optimal string length of four was found along with evidence of coherent complexity (chaos), with Lyapunov dimensionality equal to 2.12, Shannon's entropy equal to 6.44 and fractal dimension equal to 1.64. Furthermore, the frequency distribution for recurrences corresponded to a 1/fb distribution (R2 = .95). Finally, the degree of patterning in strings (log-frequency of recurrence) was tested for correlations with the relationship constructs of control, closeness and conflict among the members. Significant correlations were found between patterning and: observed control (r = .58), observed closeness (r = .36) and a combined index (self-report and observed) for conflict (r = .51). Correlations between patterning and self-reported control and self-reported closeness were not significant. Theoretical and practical implications of the results are discussed including the possibility of dynamics-based assessments and interventions in small groups from various contexts.