Reading the Readers Mind through Eye Tracking: Can AI Generated Texts Match Human Authors?

While Generative AI models like Large Language Models (LLMs)<br/>are capable of generating extensive text, their efficacy in producing<br/>readable content for human participants in experimental settings<br/>remains to be evaluated. Further, eye-tracking technology is increas-<br/>ingly utilized to study cognition and behavior, yet its application to<br/>readers’ cognitive processes when exposed to AI-generated versus<br/>human-authored texts remains unexplored. This study investigates<br/>how text generated by LLMs influences reading by analyzing gaze<br/>patterns.<br/>The study collects gaze data from 13 participants as they read AI-<br/>generated and human-authored passages. A comparative analysis is<br/>conducted within subjects to assess gaze patterns between authors<br/>and between text types based on the robust two-means clustering<br/>(I2MC) algorithm to identify fixations. In addition, pupil dilation<br/>and reading speed were examined.<br/>Our findings reveal significant differences in fixation character-<br/>istics not only between authors but also between AI-generated and<br/>human-authored texts.