What’s in a font?

How likely are you to remember this post tomorrow?
The question above asks you to make a metacognitive judgment—that is, it asks you to evaluate your own thoughts. People use metacognitive judgments every day, whether it is to believe you know a route well enough to leave your GPS at home, decide that you know a certain chapter well enough to stop studying, or have enough confidence in your ideas before presenting them to others. We use a variety of cues to make metacognitive judgments: past experience, level of importance, and, as a wealth of recent literature has demonstrated, fluency.

Fluency is a broad term that refers to the subjective ease with which we process information. Just a few examples are how easily information comes to mind, how familiar something is, or how quickly we can read a given word. Unfortunately, fluency can be very misleading. To demonstrate the misleading effects of fluency on metacognitive judgments, Rhodes and Castel (2008) presented subjects with words in either 18-point font or 48-point font and asked them to rate on a scale of 1-100 how likely they were to recall the word later. Subjects gave significantly higher predictions for larger words, but recall did not differ between font sizes. Even when subjects were explicitly told prior to the experiment that font size would not affect their recall, they still gave significantly higher predictions for larger, more fluent items.

A more surprising finding is that lower fluency can sometimes actually improve recall for disfluent items. A recent study by Diemand-Yauman, Oppenheimer, and Vaughan (2010) manipulated the typeface of study material. Subjects were presented with a series of imaginary alien species and their various characteristics to memorize. Participants later remembered more information about aliens presented in disfluent typefaces than in fluent (i.e., more common) ones. But Diemand-Yauman et al. didn’t stop there—they took their experiment into a local high school and manipulated the typeface in which teachers presented classroom materials (e.g., handouts, PowerPoint presentations) for a single unit of English, History, Chemistry, or Physics courses. The teachers obviously agreed to the manipulation but were blind to the hypotheses of the study so they would not influence the results. A control condition kept the typeface of the materials unaltered (i.e., in the style that the teacher created, which was most likely fluent and familiar), whereas disfluent conditions contained text materials changed to Haettenschweiler, Monotype Corsiva, Comic Sans, or simply jiggled up and down in the copy machine to diminish the clarity of the text. After normalizing all the classroom assessments to compare them on a common metric, the experimenters found that students in the disfluent conditions performed significantly better than students in the control condition. There were no differences among the individual disfluent conditions. Despite the performance differences between fluent and disfluent conditions, students’ metacognitive judgments indicated no differences in their feelings about the difficulty level or content of the course.

Even though students weren’t aware of any differences, the disfluent font may have triggered additional processing strategies that led to higher performance in those conditions (Alter, Oppenheimer, Epley, & Eyre, 2007). Think about it—when you read something in really small or difficult to read handwriting, you have to slow down and decipher every word. This type of purposeful, methodical processing could be the mechanism through which disfluency acts to improve memory. Connecting it to some earlier posts, disfluency may be a desirable difficulty if you are trying to learn something.

A word of caution: lower fluency does not always improve memory (remember Rhodes & Castel?) The mnemonic benefits can depend on the type of test, among other things (Hirshman, Trembath, & Mulligan, 1994). Still, if you’re looking for a quick way to potentiate a study session, it might not hurt to change your lecture notes into a more disfluent font.

For more information on metacognitive judgments and/or fluency, check out the references below and the recent New York Times article citing the Diemand-Yauman et al. (2010) work.


Alter, A.L., Oppenheimer, D.M., Epley, N., & Eyre, R.N. (2007). Overcoming intuition: Metacognitive difficulty activates analytic reasoning. Journal of Experimental Psychology: General, 136(4), 569-576.

Diemand-Yauman, C., Oppenheimer, D.M., & Vaughan, E.B. (2010). Fortune favors the bold (and the italicized): Effects of disfluency on educational outcomes. Cognition, 118,(1), 111-115.

Hirshman, E., Trembath, D., & Mulligan, N. (1994). Theoretical implications of the mnemonic benefits of perceptual interference. Journal of Experimental Psychology: Learning, Memory, & Cognition, 20(3), 608-620.

Rhodes, M.G. & Castel, A.D. (2008). Memory predictions are influenced by perceptual information: Evidence for metacognitive illusions. Journal of Experimental Psychology, 137(4), 615-625.