Many of our cognitive and metacognitive judgments are based on sheer subjective experience. Subjective experience, however, may be contaminated by irrelevant factors, resulting in biased judgments. Under certain conditions people exert a metacognitive correction process to remedy such biased judgments. In this study we examine the proposition that even after a judgment has been corrected to avoid the biasing effects on subjective experience, subjective experience itself remains biased. We asked participants to judge the difficulty of anagrams for others. When they were aware of having been exposed to the solutions of some of the anagrams, they corrected their difficulty judgments for these anagrams. Despite this correction, their speeded choices in a subsequent task disclosed their biased subjective experience that these anagrams were easier to solve. Implications for the study of metacognition and for the educational domain are discussed.