Michael McPhee¹ (), Morgan McCarty¹, Seth Schallies¹, Michael Young^2,3, Jorge Morales¹; ¹Northeastern University, ²Massachusetts General Hospital, ³Harvard Medical School

The pupillary response is driven by endogenous signals of brightness in addition to direct light. The pupil’s diameter adjusts in response to the brightness of imagined objects, the semantic brightness evoked by words, and the illusory exaggerated brightness of images depicting stereotypically bright objects like the sun. While these effects have been individually demonstrated in isolated experiments, their reliability remains uncertain and the relationship between them is unknown. Across four within-subjects experiments, we systematically investigated these effects using a head-mounted display equipped with an eye tracker, which provided a tightly controlled lighting environment for pupil size recording. Subjects imagined light or dark shapes (Exp. 1), read (Exp. 2) and listened (Exp. 3) to words evoking lightness (e.g. “flame”) or darkness (e.g. “abyss”), and examined a curated set of landscape photographs containing a clearly visible sun (Exp. 4). Words were categorized as light or dark based on ratings we collected beforehand, and the words’ valence, arousal, and length were matched across categories. Each sun image was presented upright and upside down across trials, as perspective inversion disrupts construal of the sun as exceptionally bright. Critically, our design allows us to compare the effects of these different endogenous brightness cues within each subject. While we replicated previously reported effects of illusory (Exp. 4) and semantic (Exps. 2 & 3) brightness on pupil size, we did not find a significant pupillary response to imaginary light (Exp. 1), in contrast with previous findings. Within individuals, however, we found that the magnitudes of semantic and imaginary brightness effects on pupil size were correlated – but that each was unrelated to the magnitude of the illusory brightness effect. Together, our findings suggest that pupillary reactivity to endogenous brightness signals may be governed by multiple distinct mechanisms rather than a single system sensitive to high-order sensory representations.