Validation of webcam-based eye-tracking in human sign- and goal-trackers

The tendency to attribute incentive salience to reward-predictive cues represents a key source of individual differences in human behaviour (Colaizzi et al., 2020; Heck et al., 2025). In animal research, the autoshaping procedure presents a neutral cue (e.g., a light) and a reward (e.g., a food pellet) in distinct spatial locations. After training, animals are classified as sign-trackers, which approach the cue and reflect the attribution of incentive salience, or goal-trackers, which instead approach the food-delivery location (Meyer et al., 2012). However, translating these findings from animal models to humans remains challenging. A particularly promising approach is the use of eye-tracking during a Pavlovian conditioning paradigm, with gaze duration serving as an index of approach behaviour (Garofalo & di Pellegrino, 2015; Heck et al., 2025). In recent years, there has been growing interest in conducting experimental research in online settings, including paradigms traditionally implemented in the laboratory. Webcam-based eye-tracking provides a scalable, cost-effective, and privacy-preserving solution for online experimentation, enabling rapid data collection from large, more diverse samples without specialised hardware, thereby enhancing generalisability (Yang & Krajbich, 2021; Semmelmann & Weigelt, 2018). We aim to validate WebGazer (Papoutsaki et al., 2016) as a reliable tool for measuring gaze allocation in a Pavlovian conditioning paradigm. Participants will perform the same task in two sessions: the first in the laboratory, using both a professional eye-tracker (Tobii Pro Nano) and the webcam with WebGazer, and the second entirely online, using only the webcam. This design will allow us to evaluate the reliability and validity of WebGazer for studying individual differences in incentive salience attribution in remote settings.