Many studies have failed to replicate the disfluency effect (i.e. disfluent font better than fluent font) on Cognitive Reflection Test (CRT) performance shown in Alter et al. (2007). Those studies manipulated perceptual disfluency using color, style, size, and/or typefaces. The new Sans Forgetica (SF) typeface, designed to promote desirable difficulty, creates perceptual disfluency through fragmented letters, which make it difficult for readers to use good continuation and perceptual completion to identify letters. Here, we compare CRT performance and font legibility ratings when CRT problems are presented in SF or in Arial. After solving CRT problems, participants who solved the problems in SF rated Arial to be harder to read than participants who solved the problems in Arial, suggesting that after prolong use of SF, participants viewed Arial as being closer in legibility to SF.

The new Sans Forgetica (SF) typeface creates perceptual disfluency by breaking up parts of letters vertically, horizontally, or diagonally, thereby fragmentizing them. While patterns of fragmentization are consistent for each unique letter, they are not uniform across letters. With Gestalt principles such as good continuation and perceptual completion being more difficult to implement in these settings, viewers may need to depend on context clues to identify words. This may be a desirable difficulty and improve memory for those words. Here, we investigate whether SF improves recognition of studied words. In Experiment 1, participants studied words in Arial and SF and completed old-new recognition tests where words retained their study fonts. In Experiment 2, we investigated the potential for context reinstatement—testing studied words in their studied fonts or the other font. Hit rate and discrimination sensitivities (d’) were analyzed for both experiments. Participants had significantly better recognition (hit rate) in SF than in Arial (Exp 1) and significantly higher discrimination sensitivities (d’) when words were tested in SF than in Arial (Exp 2). However, further examination of these results (e.g., marginally more response bias with SF than with Arial in Exp 1) lead us to hold reservations for the benefit of SF on word memory and conjecture that SF, at best, plays a limited role in improving recognition of studied words.