Pupil diameter is a non-invasive biomarker of brain state, correlating with arousal, attention, cognitive processing, and consciousness. Existing pupillometry software often lacks scalability and robustness across diverse experimental conditions and species. Pupil-DLC is an open-source, offline, DeepLabCut-based pipeline for scalable, marker-less pupil tracking, primarily designed for mice. Trained on 21,909 manually annotated frames from over 140 videos of head-fixed mice spanning wakefulness and drug-induced states, including psychedelics and anesthesia, the dataset was deliberately selected to maximize pupil size variability and model generalization.
A continuous 14-day infusion of the NMDA receptor antagonist phencyclidine (PCP) via osmotic mini-pumps in Lister hooded rats produced a selective deficit at the extra-dimensional shift stage of the attentional set-shifting task, mirroring the executive function impairment seen in schizophrenia. An acute injection of modafinil reversed this deficit. The impairment and its reversal were quantitatively and qualitatively similar to those observed after sub-chronic intraperitoneal PCP administration, suggesting that continuous mini-pump infusion is a viable alternative to systemic injection protocols for modeling cognitive symptoms of schizophrenia.