Is metacognitive ability a predictor of creative performance? Previous studies have produced conflicting findings. To clarify whether this relationship exists, the current study used eye tracking techniques and vocal thinking reports to explore creativity differences in individuals with different levels of metacognitive ability. One hundred and twelve participants completed the Metacognitive Ability scale, and were divided into two groups (with thirty participants in each group) based on their metacognition scores (the highest and lowest 27% of metacognitive ability scores). Then, participants in both groups completed two creative thinking tasks (AUT and CCRAT) while their eye behaviors were recorded by eye tracking. The results showed that participants with high metacognitive ability were better at divergent thinking, as evidenced by greater fixation and saccade counts, as well as smaller saccade amplitudes in the AUT task. In addition, Bayesian analyses provide anecdotal evidence that participants with high metacognitive ability tended to be better at convergent thinking. Furthermore, eye tracking results demonstrated that they exhibited longer fixation duration and more fixation count on the materials in the CCRAT task. These findings reflect an important role of metacognition in creative thinking, especially in divergent thinking.

This study aimed to revise the updated Chinese version of the Runco Ideation Behavior Scale (RIBS) and evaluate its reliability and validity among college students in Chinese Mainland. A total of 689 students completed the RIBS along with the College Students' Metacognitive Ability Scale and Creative Self-efficacy Scale as criterion measures. 488 of these participants were retested after an 8-week interval for test-retest reliability analysis while data from 825 primary and secondary school students were used for robustness analysis. EFA and CFA results indicated that 7 items with poor psychometric properties were eliminated, resulting in a 16-item revised RIBS with a stable two-factor structure (Openness-divergence, Premature closure) that was consistent with the original scale; the bifactor model of this structure showed excellent model fit (SRMR = 0.042, CFI = 0.937, TLI = 0.913, RMSEA = 0.074). Multi-group CFA confirmed strict measurement invariance (configural, metric, scalar) of the 16-item RIBS across gender, ethnicity, and only-child status. The revised scale exhibited high internal consistency (Cronbach's α: total scale = 0.910, subscales = 0.840-0.901) and acceptable 8-week test-retest reliability (ICC: 0.651-0.685). Significant positive correlations were observed between the revised RIBS's total and subscale scores and metacognitive ability, as well as creative self-efficacy and TTCT task, supporting good criterion-related validity. A supplementary sensitivity analysis with an expanded sample of 825 students (7-17 years) further verified the structural stability of the scale across different age and educational stages. The updated Chinese version of the RIBS demonstrates satisfactory psychometric properties among Chinese students and can be used to assess their everyday creativity.