Investigating the Intersection between Pupillometry and Anticipatory Eye Movements in Measuring Prediction and Prediction Error in 14-month-old Infants and Adults

Abstract

In investigating the mechanisms of prediction, recent developments in pupillometry and anticipatory eye movement (AEM) studies have allowed for key insights to be revealed regarding both adults and infants. However, few studies have examined a combination of these measures to investigate both prediction and prediction error, as the present study has attempted to do. The research question at hand is: how do anticipation and prediction error work together in 14-month-old infants compared to adults, and how can these particular mechanisms be measured using this innovative combination of techniques? The present study confirms that infants and adults engage anticipatory eye movements in a similar manner, as an increase in correct AEMs during the pre-switch trials ensured that learning was taking place, though adults demonstrated a more robust prediction mechanism. Though we did not see larger pupil size during the first post-switch trial compared to the other trials, the high spike in incorrect AEMs for this particular trial was a suggested indicator of prediction error. We also saw in both adults and infants an overall decrease in incorrect AEMs during the post-switch trials, suggesting that both groups are learning from prediction error, though adults more rapidly adjust to the new pattern. Overall, this study supports the theory that similar prediction mechanisms are in place yet are more fine-tuned in adulthood, but also provides avenues to continue refining the experimental approach in future research.