A Primer on Design and Data Analysis for Cognitive Pupillometry

The human pupil dilates almost imperceptibly in tandem with rising cognitive demands. These perturbations in pupillary surface area can yield a sensitive index of physiological arousal when time-locked to a controlled stimulus (e.g., listening for a target word). Cognitive pupillometry involves measurement of these response functions modulated by task (e.g., attention) rather than luminance. The popularity of cognitive pupillometry is growing exponentially due to the availability of low cost eyetracking systems and a variety of open source software packages. Researchers face numerous challenges in conducting a valid, replicable, and interpretable cognitive pupillometry study. This chapter provides an accessible introduction to some of the major considerations for design, execution, and analysis in cognitive pupillometry. Topics include artifact detection and correction procedures (e.g., blink detection, pupil foreshortening), mathematical approaches to baseline response scaling, and options for statistical analysis (e.g., growth curve modeling vs. ANOVA).