Jacolien van Rij (Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence, Gronigen University)
This course introduces Generalized Additive Mixed Modeling (GAMMs) for analyzing time course data such as EEG, pupil dilation, and gaze data. The course consists of short lectures and hands-on lab sessions in which we will use R (www.r-project.org) for visualizing and analyzing time course data.
In the course, we will start with the basics of nonlinear regression, and discuss how to setup a GAMM model, different smooth functions, nonlinear random effects, and visualizing the model estimates. At the end of the course we will also cover more advanced topics such as model criticism, and correcting for autocorrelation.
If you would like to explore whether and how to apply GAMMs to your own data, please bring a very short description of your data (dependent variable(s), experimental design, and predictors and covariates in the data).
Requirements
The course requires basic experience with R and linear mixed effects models, although most R code will be provided. Please bring your own laptop with R or R Studio installed. (If you cannot bring your own laptop, please inform ccpling@ualberta.ca and we will try to arrange a laptop to use during the course.)
Further, please make sure to read the following papers/tutorials in nonlinear regression before the start of the course:
– Jacolien van Rij, Petra Hendriks, Hedderik van Rijn, R. Harald Baayen, and Simon N. Wood: Analyzing the time course of pupillometric data. Accepted for Trends in Hearing Science. (Will be available online soon.)
– Wieling, M. (2018): Analyzing dynamic phonetic data using generalized additive mixed modeling: a tutorial focusing on articulatory differences between L1 and L2 speakers of English. Journal of Phonetics, 70, 86-116.