Research Seminar

Understanding the linear Gaussian state space model


Tom Lodewyckx


KU Leuven

Abstract: Recently, emotion research has been focusing on the conceptualization of emotions as multicomponential, dynamical systems. This development created new types of research questions, pertaining to, for instance, autoregressive and crossregressive relations between emotion components. From this point of view, state space modeling is an interesting approach for modeling emotion processes. The goal of this presentation is to familiarize the linear Gaussian state space model and to illustrate its flexibility. We start with a clear description of the general framework (its model structure, parameters and assumptions). We continue with specific applications of the state space model, ranging from a basic measurement context in psychophysiology to a dynamic factor model, an autoregressive moving average model and a typical engineering model. Although the linear Gaussian state space model is already quite flexible, we discuss some extensions of the model which are of particular interest in psychology.
Date: Tue Feb 22, 12:15 pm - 1:15 pm
Place: room 02.60 (Department of Psychology, Tiensestraat 102, 3000 Leuven)