Neurodynamic analysis of psychiatric disease mechanisms using computational network models
The focus of this subproject (SP) is the development of data-based mathematical models and methods to elucidate the dynamics of neuronal networks and the manner in which they change under psychiatric and pharmacological conditions, and which can predict the impact of neuronal dynamics on therapy. The approach is based on the conceptualization of neuronal system dynamics as the functional bridge between physiological and structural changes in the nervous system on the one hand, and cognitive abilities and behavior on the other. Neurodynamic models are derived systematically from physiological data and experimental measurements, such as those obtained from functional imaging or electrophysiological methods. They are then analyzed in relation to behavioral characteristics, or to pharmacological or genetic conditions, as based on data collected within other subprojects (SP2, SP3, SP4). The development and application of the methodology are achieved in close collaboration with SP1, and involve the use of various methods from statistical time series analysis, machine learning, and non-linear dynamics. The aim is to characterize and differentiate proband and patient groups in terms of their neurodynamics in order to identify new biomarkers, and to predict disease outcome and the unique response of the individual to pharmacological or other therapy.