The effect of distinctive visual information on false recognition

Using the false memory paradigm (Deese, 1959), recently revived by Roediger and McDermott (1995), we examined the effect on true and false recognition of presenting study items in unusual looking fonts. In one condition, each font was associated with a single study item. In a second condition, each font was presented 12 times per study list, randomly distributed across several themes. In a third condition, each font was presented 12 times in the study list, and was associated with a particular study theme. False recognition levels were lowest when there was a unique association between each font and a single study item, whereas false recognition levels were highest when all items from a theme were presented in the same font. Further, the effects of font condition on false recognition of lures maintained when font condition was manipulated within participants and lists. These results, taken together, are inconsistent with theories proposing that false recognition reduction is the product of global shifts in response strategies across conditions (e.g., Schacter, Israel, & Racine, 1999). However, perspectives highlighting the effects of memory based processes on true and false recognition provide an adequate account.