< Back to previous page


Putting affective dynamics into context: a comprehensive moderated time series approach

Affect research relies on intensive longitudinal data nowadays to understand the dynamics underneath affective functioning. To unravel them, vector autoregressive (VAR) models, in which affective states are regressed upon each other over time, have become especially popular. Theoretical work, however, emphasizes the notion of regulatory flexibility, suggesting that affective dynamics depend on and thus change with context. To empirically study this, VAR models that include contextual moderators are required. This research proposal therefore aims to build a comprehensive modeling framework to put affective dynamics into context. We will base this framework on fixed moderated time series analysis (fmTSA; Adolf et al., 2017), a recently introduced method to incorporate a single time-varying contextual moderator into a VAR model for a single person. We will extend fmTSA in important ways. First, we will study which types of parameter changes are induced by a moderator. Next, we will include multiple moderators, followed by accommodating varying intervals between measurements. Finally, we will model data from multiple individuals simultaneously. To evaluate the models and select among them, we will propose routines based on cross-validation, which will also allow to prospectively optimize study design. Finally, we will disseminate the developed framework to applied researchers by building userfriendly software and providing proof-of-principle applications to empirical data.

Date:1 Jan 2019 →  31 Dec 2021
Keywords:Learning, motivation and emotion
Disciplines:Data models