Olessia Jouravlev¹, Veronica Chiarelli¹, Dalia Serik¹, Tamaya Murray¹; ¹Carleton University

Introduction: The semantic Pupillary Light Response (sPLR)—pupil constriction to semantically bright words and dilation to dark words—has been proposed as evidence for embodied language processing, reflecting activation of visual sensory representations during word comprehension. Prior work shows robust sPLR effects (Mathôt et al., 2017), yet the contribution of individual differences in visual mental imagery and language background remains unclear. We investigated whether (1) semantic processing of visually bright and dark words elicits sPLR, (2) imagery vividness modulates this effect, and (3) native versus non-native English status influences sPLR strength during English word reading task. Methods: Seventy participants completed a visual go/no-go categorization task while pupil diameter was recorded with an EyeLink 1000. Stimuli included semantically bright, dark, matched control words, and animal names (go-targets). Brightness/darkness stimuli were matched to controls on lexical frequency, arousal, valence, and visual intensity. Pupil dilation ratio was computed as proportional change from a pre-stimulus baseline. After the task, participants completed the Vividness of Visual Imagery Questionnaire (VVIQ-2). Results: There was a significant main effect of word condition. Pupil constriction was greater for semantically bright words (M=99.23) relative to dark (M=100.06) and control words (M=100.21), which did not differ. No interaction was observed between condition and language status or VVIQ-2. However, a follow-up analysis including only bright versus control conditions showed a significant interaction between condition and imagery scores, suggesting imagery ability selectively modulates constriction to bright words. Conclusion: These findings partially replicate the sPLR for semantic brightness but not darkness, providing qualified support for embodied accounts. Results further indicate that individual differences in visual imagery may influence sPLR, whereas language status did not modulate the effect in this sample.