The Effect of Word Frequency on Judgments of Learning: Contributions of Beliefs and Processing Fluency

Previous research has shown that word frequency affects judgments of learning (JOLs). Specifically, people give higher JOLs for high-frequency (HF) words than for low-frequency (LF) words. However, the exact mechanism underlying this effect is largely unknown. The present study replicated and extended previous work by exploring the contributions of processing fluency and beliefs to the word frequency effect. In Experiment 1, participants studied HF and LF words and made immediate JOLs. The findings showed that participants gave higher JOLs for HF words than for LF ones, reflecting the word frequency effect. In Experiment 2a (measuring the encoding fluency by using self-paced study time) and Experiment 2b (disrupting perceptual fluency by presenting words in an easy or difficult font style), we evaluated the contribution of processing fluency. The findings of Experiment 2a revealed no significant difference in self-paced study time between HF and LF words. The findings of Experiment 2b showed that the size of word frequency effect did not decrease or disappear even when presenting words in a difficult font style. In Experiment 3a (a questionnaire-based study) and Experiment 3b (making pre-study JOLs), we evaluated the role of beliefs in this word frequency effect. The results of Experiment 3a showed that participants gave higher estimates for HF as compared to LF words. That is, they estimated that hypothetical participants would better remember the HF words. The results of Experiment 3b showed that participants gave higher pre-study JOLs for HF than for LF words. These results across experiments suggested that people's beliefs, not processing fluency, contribute substantially to the word frequency effect on JOLs. However, considering the validation of the indexes reflecting the processing fluency in the current study, we cannot entirely rule out the possible contribution of processing fluency. The relative contribution of processing fluency and beliefs to word frequency effect and the theoretical implications were discussed.