Jacolien Van Rij
Generalized Additive Mixed Modeling (GAMMs; Lin & Zhang, 1999) is a nonlinear regression method that can estimate nonlinear trends and nonlinear interaction surfaces. As such, this method is particularly suited for analyzing time course data, such as gaze data, pupil dilation, and ERP.
This one-day workshop will provide a hands-on introduction to GAMMs, as implemented in the R package mgcv (Wood, S.N., 2006; 2011). In the workshop we will analyze the gaze data collected in the STEP course “Exploring the Visual World Eye-Tracking Paradigm” as an example. Different topics will be addressed: how to setup GAMM models (nonlinear trends, nonlinear interactions, and nonlinear random effects), how to visualize nonlinear estimates and nonlinear interactions, how to identify the best model given the data, and how to evaluate GAMM models.
The workshop combines short lectures with practical hands-on sessions, so participants are adviced to bring a laptop with R installed (or share a laptop with someone else during the practical sessions). Some basic R experience is definitely an advantage, but not required: most of the R code for the practical sessions will be provided.