Metacognitive awareness of associative learning: What underlies delayed judgments-of-learning?

Cognitive processes, such as memory, are accompanied by metacognitive states of awareness that allow for evaluation of their function. Across seven experiments we employed the delayed judgment-of-learning (JOL) paradigm with healthy young adults to examine metacognitive monitoring of learning. After studying cue-target word-pairs, participants were presented with the studied cues and predicted their ability to retrieve the target on a subsequent memory test. The key question of interest was the nature of the underlying processes guiding such judgments with a focus on how they relate to memory. The delayed JOL literature has assumed that it is an absolute judgment, based on the ease of access to the target item. Chapters 2 and 3 manipulated target- and cue-related variables and investigated their influence on memory and metamemory. The results showed delayed JOLs are also sensitive to memory for contextual information about the target (Chapter 2) and the level of familiarity with the cue term (Chapter 3). This is strengthened by results from Chapter 4 in which participants provided written justifications of their JOL responses without any experimental manipulations of the learned material. Analysis of these responses confirmed that both cue- and target-related information influences delayed JOLs. Lastly, we showed that delayed JOLs are not sensitive to whether they are predicting recognition or recall (so called theory-based influences) unless participants make a different prediction on each trial (i.e. trial-level design, Chapter 5). Overall, delayed JOLs are shown to vary with variables that fluctuate on a trial level, which can but do not necessarily need to map onto memory. The results suggest that delayed JOLs are primarily comparative judgments, involving the evaluation of the quantity and quality of evidence on any given trial in the context of the task at hand (e.g. by comparison to preceding trials). This is contrary to how it is often treated in the delayed JOL literature but is consistent with other metacognitive paradigms.