CCP | Exploring the Visual World Eye-tracking Paradigm – from data collection to statistical analysis
Exploring the Visual World Eye-tracking Paradigm – from data collection to statistical analysis

Jacolien Van Rij (University of Tübingen, Germany)
Juhani Järvikivi (University of Alberta, Canada)

This course provides an introduction to the Visual World Paradigm (VWP), an eye-tracking method in which participants are presented with pictures (or videos) on the screen, while they listen to sentences or discourse. This combined use of visual information and speech makes VWP particularly well suited to the investigation of spoken language processing, the interaction between visual and linguistic information and the on-­‐line language comprehension in young children and other populations for which we choose not to or cannot rely on reading measures.

The course focuses on practical training, including eye movement recording, statistical analysis, and interpretation of eye tracking data. Following an introduction to the Visual World Paradigm and the relevant eye tracking measures, the first hands-­‐on sessions allow students to learn how to operate an eye tracker and how to collect data using the VWP. In the second part of the course, students work with new statistical methods for analyzing eye movement data, including pupil size and gaze data. Typically, eye movement data is not analyzed directly, but aggregated or simplified for the analyses. As a result, a lot of important information about the time course and the individual trials is lost. To counter this, we will introduce participants to a new approach, Generalized Additive Modeling (GAM;
Wood, 2006), to analyze both pupil size and time course data. This type of mixed-­effects regression analysis allows for modeling non-­‐linear effects and interactions typical for the VWP.

The course combines short lectures with practical hands-­‐on sessions. No previous knowledge of eye tracking is required. However, participants are free to bring their own (VWP) eye tracking data and apply the assignments to their own data. We will use R for the analysis. Basic knowledge of statistics (regression analysis) is assumed.