Analyzing the pupil response due to increased cognitive demand: an independent component analysis study

Pupillometry is used to indicate the relative extent of processing demands within or between tasks; however, this analysis is complicated by the fact that the pupil also responds to low-level aspects of visual input. First, we attempted to identify "principal" components that contribute to the pupil response by computing a principal component analysis (PCA) and second, to reveal "hidden" sources within the pupil response by calculating an independent component analysis (ICA). Pupil response data were collected while subjects read, added or multiplied numbers. A set of 3 factors/components were identified as resembling the individual pupil responses, but only one ICA component changed in concordance to the cognitive demand. This component alone accounted for about 50% of the variance of the pupil response during the most demanding task, i.e. the multiplication task. The highest impact of this factor was observed for 2000 to 300ms after task onset. Even though we did not attempt to answer the question of the functional background of the components 1 and 3, we speculated that component 2 might reflect the effort a subject engages to perform a task with greater difficulty.