Assessing sustainability literacy is essential for ensuring that sustainability literacy tests accurately measure individuals' understanding of sustainability concepts. However, conventional evaluation methods primarily rely on test scores and response accuracy, providing limited insight into the cognitive processes underlying participants' decision-making. This study investigates the use of eye-tracking technology to analyze participants' visual attention while completing a Sustainability Literacy Test (SLT). Eye movement data, including scanpaths, were analyzed to identify user interactions with test items and determine which visual elements attracted the greatest attention, particularly when participants selected the "I Don't Know" response option. In addition, system usability was evaluated using the Hybrid System Usability Scale (H-SUS). The results indicate that participants required an average response time of more than 60 seconds for items associated with the "I Don't Know" option, suggesting increased visual attention and cognitive load in specific areas of the interface. The website achieved an average H-SUS score of 63.34, indicating an acceptable level of usability, corresponding to Grade C and an "OK" adjective rating. These findings demonstrate that eye-tracking data can reveal hidden patterns of cognitive effort and user uncertainty that cannot be captured through response accuracy alone. The proposed approach provides valuable insights for improving test design, enhancing question readability, and optimizing the user interface of web-based Sustainability Literacy Tests, thereby supporting more effective assessment of sustainability literacy.