Effects of Chinese proficiency and illustration types on Chinese reading comprehension among international students: evidence from eye-tracking

Purpose This research investigates how Chinese proficiency and different types of illustrations affect international students reading comprehension of Chines science and Technology text. Eye tracking technology will be used to clarify the cognitive process involved. Methods We conducted a mixed-factorial eye-tracking experiment with a 2 (Chinese proficiency: high vs. low) × 4 (illustration type: representational, explanatory, organizational, transformational) design. We used a Tobii Pro Glasses 2 eye-tracker to record eye movement data, specifically total text-image fixation time, illustration area fixation duration, and fixation count. To analyze the effect of proficiency, illustration type, and their interaction, we applied a Linear Mixed Model (LMM), treating proficiency and illustration type as fixed factors and individual differences as random effects. Results The results revealed a significant interaction between Chinese proficiency and illustration type on comprehension scores [ F (3,82) = 3.25, p = 0.035]. Specifically, high-proficiency learners demonstrated better comprehension with explanatory and organizational illustrations. They appeared to leverage the structured features of these semantic integration as evidenced by higher fixation counts on organizational illustrations (28.07). Conversely, low-proficiency learners relied more on representational illustrations to reduce cognitive load. They also exhibited significantly increased fixation counts and durations when processing transformational illustrations (total fixation duration: 325.62 s), suggesting that their limited language capacity led to cognitive resource competition and integration difficulties. Conclusion This study confirms a proficiency-illustration matching mechanism in Science and Technology Chinese reading. Key findings indicate that: (1) high-proficiency learners benefit most from explanatory and organizational illustrations, while low-proficiency learners rely on representational illustrations to reduce cognitive load; (2) transformational illustrations pose higher processing demands for low-proficiency learners, reflected in prolonged fixation durations; and (3) eye-movement patterns reveal distinct, proficiency-dependent processing strategies. The results underscore the need for aligning illustration types with learners’ Chinese proficiency to optimize comprehension and cognitive efficiency in academic second-language contexts.