Frequency analysis of a task-evoked pupillary response: Luminance-independent measure of mental effort

Abstract

The pupil diameter is a widely studied cognitive load measure, which, despite its convenience for non-intrusive operator state monitoring in complex environments, is still not available for in situ measurements because of numerous methodological limitations. The most important of these limitations is the influence of the pupillary light reflex. Hence, there is the need of providing a pupil-based cognitive load measure independent of light conditions. In this paper, we present a promising technique of pupillary signal analysis resulting in luminance-independent measure of mental effort that could be used in real-time without a priori on luminous conditions. Twenty two participants performed a short-term memory task under different screen luminance conditions. Our results showed that the amplitude of pupil dilation due to load on memory is luminance-dependent with higher amplitude corresponding to low-luminance condition. Furthermore, our experimentation showed that the load on memory and luminance factors express themselves differently according to frequency. Therefore, as our statistical analysis revealed, the ratio between low (0–1.6 Hz) and high frequency (1.6–4 Hz) bands (LF/HF ratio) of power spectral densities of pupillary signal is sensitive to the cognitive load but not to luminance. This result is promising for the measurements of load on memory in ecological settings.