Abstract Judgments of learning (JOLs) are characterized as metacognitive evaluations about the likelihood that studied items can be successfully retrieved in a future memory test. Previous studies found that people employ different types of cues to inform their online JOLs. Some of these cues can guide JOLs to accurately reflect memory status but others cannot (and are even misleading in many situations). A widely studied cue for JOL formation is the subjective processing experience (e.g., perceptual fluency) during completing a given task, which often confers metacognitive illusions. It has been found that people give higher JOLs to large than to small font size words, despite the fact that font size has minimal effect on retention, a phenomenon termed the font size effect on JOLs. A potential mechanism underlying the effect is perceptual fluency: Large words are perceived more fluently than small ones, and fluent processing experience of large words induces a feeling-of-knowing, which drives people to offer higher JOLs. The font size effect is important because it spotlights a processing dissociation between metacognitive judgments and memory itself. The current study aims to explore the influences of encoding strength on the font size effect, and to explore practical techniques/strategies to calibrate metacognitive illusions induced by perceptual features.
Experiment 1 targeted to delineate the role of perceptual fluency in the font size effect. Twenty-six participants first completed a continuous identification (CID) task to measure the difference in perceptual fluency (indexed by response times; RTs) between large (70-pt) and small (9-pt) words, after which they attended a classic learning task. In the learning task, participants studied large and small words one by one, for 2 sec each, and made item-by-item JOLs. Immediately following the learning task, they completed a distractor task, followed by a free recall test. The results showed that, in the CID task, participants responded much faster to large than to small words, indicating the natural difference in perceptual fluency between large and small words. In addition, perceptual fluency (i.e., RTs in the CID task) significantly correlated to JOLs, reflecting the fluency effect on JOLs. More importantly, perceptual fluency significantly mediated the font size effect on JOLs, supporting the claim that perceptual fluency is (at least part of) the source of the font size effect.
Experiment 2 manipulated study durations to investigate the influence of enhancing encoding strength (through prolonging study duration) on the font size effect. Specifically, three groups of participants studied the words for 2 sec, 4 sec, and 8sec, respectively, and made item-by-item JOLs. The results demonstrated that prolonging study duration correspondingly decreased the font size effect on JOLs. It is however worth highlighting that expanding study time cannot fully eliminate the font size effect because the results still showed a significant font size effect even when study time was extended to 8 sec.
Experiment 3 was conducted to further unravel the effectiveness of enhancing encoding strength for calibration of the font size effect. A sentence-making group was instructed to encode the words by generating a sentence with the on-screen word, which was employed to deepen the level of processing (i.e., encoding strength); by contrast, for a control group, there were no explicit requirements of encoding strategies (i.e., participants in the control group could use any strategies as they liked). In the control group, the classic font size effect on JOLs was successfully replicated; of critical interest, the effect disappeared in the sentence-making group. Such results reflect the power of improving encoding strength to calibrate metacognitive illusions induced by perceptual features.
In summary, the current study establishes that perceptual fluency is at least one the mechanisms underlying the font size effect on JOLs; prolonging study duration reduces but fails to eliminate the font size effect; most importantly, directly deepening the level of processing through sentence-making is a valid strategy to calibrate metacognitive illusions induced by perceptual features. Theoretical and practical implications are further discussed in the main text.
Key words: judgments of learning
the font size effect
Just Accepted Date: 12 June 2020
Online First Date: 07 July 2020
Issue Date: 27 May 2020