Sabrina Mai^1,2,3, Farran Briggs^1,4; ¹Thalamocortical Visual Processing Section, National Eye Institute, National Institutes of Health, Bethesda, MD, USA, ²Graduate Partnership Program, National Institutes of Health, Bethesda, MD, USA, ³Neuroscience Graduate Program, University of Rochester, Rochester, NY, USA, ⁴Department of Neuroscience, University of Rochester School of Medicine, Rochester, NY, USA

Electrophysiology studies in vision science require precise measurements of subjects’ eye position over time. To ensure measurement accuracy, the standard approach with behaving nonhuman primates (NHPs) has been head stabilization via surgically implanted head posts. In recent years, there has been an increase in efforts to utilize less invasive methods, and develop paradigms that encourage animals to voluntarily engage in behavioral tasks. Here, we quantify eye position stability during fixation across two different covert attention discrimination tasks, comparing between NHPs whose heads were stabilized via a head post versus a guide for voluntary head positioning. The guide was a 3D-printed mask of each individual NHP’s face with a large central cutout to allow for eye-tracking and reward delivery. The head-fixed system was the Precision Positioning System by Thomas RECORDING GmbH. Importantly, identical eye-tracking and reward delivery systems were utilized with both head-fixed and mask-stabilized animals. The two NHPs whose data were analyzed here, were trained to perform the same covert attention tasks, requiring discrimination of changes in attended stimulus contrast, or orientation. We analyzed average eye position before stimulus changes to assess fixation precision for each head-stabilization method. We also assessed whether NHPs displayed any systematic biases in eye position as a function of behavioral requirements (covertly attended location) or outcomes (answer by saccading up or down). We found that neither NHP displayed systematic biases in eye position dependent upon behavioral requirements or outcomes (p>0.7 for all comparisons across both NHPs). Interestingly, the head-fixed NHP displayed slightly larger positional variance compared to the voluntarily head stabilizing NHP, which we attribute to reduced motivation and overall task performance in this animal. These results contribute to growing evidence that less invasive head stabilization approaches are as effective as conventional head-fixing in enabling accurate measurements of eye position.