The purpose of this project is to develop mathematical and statistical models for examining the dynamics of dyadic interactions. Although most theories of social relations describe interactions as dynamic processes, many of their theoretical hypotheses remain untested because of a lack of statistical models that are capable of capturing such processes over time. The proposed research addresses this shortcoming by developing and comparing alternative state-of-the-art mathematical and statistical models that can capture the dynamics of interactions in dyads, as these interactions evolve over time. These models will be applied to data collected from couples at multiple time points to study affective processes and emotion regulation, to test hypotheses derived from psychological and sociological theories regarding relationship quality and instability over time. Finally, the developed models will be applied to other types of dyads (e.g., infant-caregiver), other systems of two elements (e.g., cognition-emotion), and more complex social systems (e.g., infant-mother-father triad). The use of the developed models to evaluate hypotheses derived from both psychological and sociological theories should enhance the understanding of dyadic interactions and will broaden the development of theory in this area. In addition to the advancement in models for dyadic interactions, the proposed work presents several broader impacts. These include research training of graduate and undergraduate students (including underrepresented minorities), dissemination of research through publications in scientific journals, integration of methodological and substantive results into the academic curriculum, and the possibility of applying some of the developed models to clinical settings, for example in the evaluation and prediction of relationship quality and stability over time.