Examining what prestudy and immediate judgments of learning reveal about the bases of metamemory judgments Jodi Price ⇑ , Alan Harrison 1 The University of Alabama in Huntsville, United States article info Article history: Received 1 March 2016 revision received 25 August 2016 Keywords: JOLs Font size effect Relatedness Associative learning Fluency Memory beliefs abstract Across three experiments we examined the basis for participants’ judgments of learning (JOLs) – memory beliefs or fluency – by manipulating item relatedness and font size and collecting prestudy JOLs, immediate JOLs, or both types (combination) of JOLs. Experiment 3 also measured self-paced study time as an indirect measure of participants’ perceived fluency of items. All three experiments revealed higher prestudy, immediate, and combination JOLs for related, and large font items than for unrelated, or small font items. However, combination conditions suggested that prestudy and immediate JOLs are not identical. Prestudy and immediate JOLs differed both in terms of magnitude and accuracy, with higher magnitude and relative accuracy for immediate JOLs than prestudy JOLs in all three experiments, suggesting memory beliefs alone are not sufficient to accurately track memory performance. These results combine with the study time data in Experiment 3, which did not differ significantly as a function of font size, to suggest that memory beliefs influence both types of JOLs. However, as participants gained exposure to stimuli charac- teristics, immediate JOLs shifted more than prestudy JOLs, suggesting that the timing of immediate JOLs facilitates participants’ analysis of fluency and other item qualities that affect recall, which are then factored into participants’ immediate JOLs more so than pre- study JOLs. Ó 2016 Elsevier Inc. All rights reserved. Introduction Learners engaged in self-regulated learning must assess the likelihood of being able to learn and recall the material they are studying (Metcalfe & Kornell, 2005). It is critical that these judgments are accurate given that these assess- ments likely affect future study decisions (Metcalfe & Finn, 2008; Nelson & Narens, 1990). Researchers frequently col- lect metamemory judgments to evaluate how accurately people are able to gauge their learning. Judgments of learn- ing (JOLs) are one type of item-level judgment that mea- sure participants’ confidence they will be able to recall a studied item (0 = no confidence and 100 = complete confidence). JOLs are typically collected after participants study an item. If the JOLs are collected immediately after studying the item they are called immediate JOLs, whereas JOLs col- lected after all items have been studied are referred to as delayed JOLs (Dunlosky & Nelson, 1994; Nelson & Dunlosky, 1991). Recently, Castel (2008) introduced another type of JOL, a prestudy JOL, in which participants predict the likelihood of being able to learn and recall an item before studying it. That is, participants are provided information about the to-be-learned material, such as whether the material is semantically related or unrelated http://dx.doi.org/10.1016/j.jml.2016.12.003 0749-596X/Ó 2016 Elsevier Inc. All rights reserved. ⇑ Corresponding author at: Department of Psychology, MOR 335, University of Alabama in Huntsville, Huntsville, AL 35899, United States. E-mail addresses: jodi.price@uah.edu (J. Price), ath0010@uah.edu (A. Harrison). 1 Address: Department of Psychology, University of Alabama in Huntsville, Huntsville, AL 35899, United States. Journal of Memory and Language 94 (2017) 177–194 Contents lists available at ScienceDirect Journal of Memory and Language journal homepage: www.elsevier.com/locate/jml