Trial-level pupil dynamics reflect attention-related variability and medication-associated modulation in ADHD

Children with Attention-Deficit/Hyperactivity Disorder (ADHD) often exhibit rapid fluctuations in attention that are difficult to capture using standard behavioural assessments. We investigated whether trial-level pupil dynamics provide a physiological signal associated with task performance during a working-memory task and whether these responses differ with stimulant medication condition. Eye-tracking data were analysed from 50 children (28 ADHD, 22 Ctrl; mean age 10.6 years) performing a visuospatial Sternberg task.

Trial-level pupil signals were characterised using inter-trial variance, within-trial variability (SD), and task-locked mean pupil responses. A hierarchical location–scale model estimated approximately 23% greater residual inter-trial variance in Off-ADHD than in Ctrl after adjustment for memory load, age, and IQ. The corresponding participant-level difference was statistically significant before correction but did not remain significant after FDR correction \((q=.051)\).

Within-trial variability was lower, rather than higher, in Off-ADHD than in Ctrl. Inter-trial variance was negatively associated with task accuracy in the Off-ADHD condition, although the strength of this association did not differ significantly from Ctrl. The On-ADHD condition was associated with an increased relative probe-window pupil response and reduced inter-trial variance within participants.

However, these differences cannot be interpreted exclusively as cognitive medication effects because direct pharmacological and session-related contributions could not be fully excluded. Trajectory-based analyses further showed that the On-ADHD group-mean pupil trajectory was closer to Ctrl than the Off-ADHD trajectory under the DTW analysis. These findings suggest that trial-level pupil dynamics provide complementary physiological information about variability and task performance in ADHD, while remaining an indirect and non-specific measure of attention-related processing.

This research work is supported by the Infosys Center for AI and the Center for Design and New Media (a TCS Foundation Initiative supported by Tata Consultancy Services) at IIIT-Delhi, India. Human-Machine Interaction Lab, IIIT-Delhi, New Delhi, India The authors declare no competing interests. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material.