This paper investigates the application of novel eye-tracking metrics to assess cognitive load in visual decision-making processes. In particular, it aims to contribute to understanding the cognitive process and developing complementary measures for evaluating visual behavior in neonatal pain assessment. We have carried out an eye-tracking study using the relative Explore-Exploit Ratio proposed and the application of the Task-Evoked Pupillary Response to evaluate pediatrician experts, non-experts, and parents while analyzing frontal facial faces of distinct newborns, before and after painful procedures, from a benchmark dataset considering areas of interest clinically relevant. The Tobii TX300 eye-tracking system recorded this data in a closed room with controlled lighting. Our results disclose that the visual attention described by the traditional metrics may not correspond directly to the respective fixation patterns and pupillary changes quantified for all the sample groups of participants investigated, highlighting statistically significant differences in the visual behavior between experts and non-experts in such vital decision-making tasks

Facial expression has been widely used in clinical practice to assess pain in newborns. However, the inherent visual attention required to make such vital inference is poorly understood. It is also unknown whether this inference occurs differently when comparing health professionals with other adults. To investigate these issues, we have recorded and monitored the pupil size signal of 102 subjects (44 experts, 29 parents, and 29 non-experts) while visually analyzing 20 frontal face images of 10 distinct newborns after a painful procedure and at painless rest. Our experimental results have showed that neonatal pain assessment is more cognitively demanding when analyzing the presence of pain rather than its absence. Moreover, our results disclose that a 2-second exposure to a facial expression is sufficient to make this assessment, regardless whether done by health professionals or non-health ones, suggesting that this highly specific visual task is not driven by clinical experience.