Judgment of learning is a predictive judgment of an individual about the possibility of remembering a learned item in a subsequent memory test. People can use judgment of learning to guide the allocation of resources and the choice of learning strategies. Given the importance of judgment of learning to autonomous learning, understanding how learners monitored their learning was the focus of metamemory research. The cue utilization model believed that individuals can make judgment of learning based on different types of cues. Previous studies had found that font size as an external cue had an important impact on the immediate judgment of learning. People's judgment of learning on large-font items were significantly higher than those on small-font items. This phenomenon had been verified in many studies, which researchers called the font size effect of judgment of learning. As a stabilizing factor, why did the font size affect judgment of learning? This was the main focus of this study. In this regard, the processing fluency hypothesis believed that items presented in large font had higher processing fluency than items presented in small font. The ease of processing had direct and unconscious effects on judgment of learning. This hypothesis was supported by partial research. However, some studies had failed to reveal the role of processing fluency in the cue effect of judgment of learning. They believed that the validity of the processing fluency hypothesis might need to meet certain conditions. Taking into account the controversial conclusions of previous studies, in order to further evaluate the role of processing fluency in font size effect of judgment of learning, and to explore the effective conditions of the processing fluency hypothesis, we designed two experiments to examine the effect of processing fluency on the font size of judgment of learning. We intended to use medium-correlated pairs of words, with the self-paced study time and the number of acquisition experiments as indicators for measuring processing fluency, and examined the role of processing fluency on the font size effect of judgments of learning. The hypothesis of this study was that the font size affected the self-paced study time and the number of acquisition experiments; the effect of font size on judgment of learning was based on processing fluency measured by the self-paced study time and the number of acquisition experiments. The results of experiment 1 showed that the font size affected the self-paced study time and judgment of learning, but had no significant effect on the recall performance; and the processing fluency measured by the self-paced study time played a certain role in the font size effect of judgment of learning. The results of experiment 2 showed that the font size affected the number of acquisition experiments and judgment of learning, but had no significant effect on recall performance; and the processing fluency measured by the number of acquisition experiments did not play a role in the font size effect of judgment of learning. In summary, these results further illustrated that the font size effect of judgment of learning was a metacognitive illusion and also provided supporting evidence for the processing fluency hypothesis.