Idiographic psychological networks based on intensive longitudinal data are increasingly employed in clinical practice. However, these models mainly focus on the associations among psychological variables and changes in these associations, whereas the underlying factors for those changes are not taken into account. The factors contributing to change can be studied with moderation analyses, but although such analyses are standard in

Abstract

Idiographic psychological networks based on intensive longitudinal data are increasingly employed in clinical practice. However, these models mainly focus on the associations among psychological variables and changes in these associations, whereas the underlying factors for those changes are not taken into account. The factors contributing to change can be studied with moderation analyses, but although such analyses are standard in clinical research, they are hardly applied in the domain of idiographic networks. Therefore, we implement the fixed moderated time series model to study how networks change depending on context factors. Fixed moderated time series analysis is a vector autoregressive based model, in which all parameters of the model can be moderated, including the innovation structure. As the model is based on the state space framework, it can also directly estimate changes in the mean levels of the variables in the network. With two empirical examples, we demonstrate how the fixed moderated time series model can reveal changing network structures. We show that this idiographic moderation approach not only provides a new way to look at what parameters in a network change over time, but also offers tools to see which factors are associated with the change.