| Cognitive load refers to the amount of effort required by an individual to process information. Nearly every task such as memorizing, searching, reading, writing, or playing a game induces Cognitive load on an individual. The cognitive psychology community has conducted experiments for over fifty years demonstrating that cognitive load can be measured using sub-millimeter fluctuations in ocular pupil size, assessed by using medical-grade infrared devices known as pupillometers. Recent work has shown that the same fluctuations in pupillary response can be measured using high-resolution infrared gaze-trackers. However, the cost and availability of these gaze-trackers limit most of the research on pupil response measurement to laboratory settings. We argue that ubiquitously measuring pupillary response, also referred to as pupillometry, could transform the next generation of context-aware computing applications---enabling computational devices to understand a user's current ability to process information. To this end, we present PupilWare, a system that analyzes pupil size changes through commodity cameras like those commonly installed in laptops, tablets, or smartphones. We evaluate PupilWare's ability to measure changes in pupil dilation using classic cognitive psychology experiments and validate its performance compared to infrared gaze-trackers and medical-grade pupillometers. We conclude that, in controlled conditions, PupilWare is as accurate as infrared gaze-tracking in assessing task evoked cognitive load. Furthermore, we employ personalized classification algorithms for assessing a user's cognitive load in real-time. We validate our procedures in both online and offline experiments using human subjects. We also employ classical cognitive load tests from the psychology community, demonstrating that PupilWare classifies cognitive load correctly in the majority of the scenarios. |
