Epistemic and Bayesian Updating

Rare when an undergraduate honor's thesis turns into a poster at Psychonomics, but Sungjoon Park (now a graduate student at Texas A&M University) managed to pull it off with a lot of work and effort. In addition to all the research and analysis required of any project he also had to teach himself how to use a new eye tracker I had purchased for the lab (which works great by the way). As the distance/virtual nature of the meeting may have made it hard for people to find the work I am posting it here. If you don't have time to watch the tl;dr is that we used the change in pupil size to explore when a "surprising" event was merely surprising in the statistical - low probability sense - or when it was surprising in a way that gave you new information about the scenario (epistemic surprise). We adapted this task from some fMRI work with the idea that the pupil might be a more accessible way to get at some of the same ideas. In short, it wasn't all that clear whether there is or is not a big difference between these two types of surprise in these data, but as a first step we gained a lot of familiarity with the technique, and both pupilometrics and this simple sort of game/knowledge manipulation seems like it deserves more detailed and systematic experimentation. Let us know what you think.

Sungjoon