Do people use category-learning judgments to regulate their learning of natural categories?

Abstract

Although research has established that people can accurately judge how well they have learned categories, no research has examined whether people use their category-learning judgments (CLJs) to regulate their restudy of natural categories. Thus, in five experiments we investigated the relationship between people's CLJs and selections of categories for restudy. Participants first attempted to learn natural categories (bird families; e.g., finches, grosbeaks, and warblers) so that they could categorize new exemplars on a final test. After this initial study phase, participants made a CLJ for each category and then selected a subset of the categories for restudy. Across experiments, we also manipulated several variables (e.g., selecting either three or nine categories, or obtaining 30% vs. 80% performance on the final test) that were expected to influence restudy selections. However, the manipulations typically had minimal impact. More important, in all experiments we found an unexpected outcome: Some participants tended to select the categories they judged to be most well learned for restudy, and others tended to select those judged to be least well learned. We discovered these qualitative differences in the use of CLJs to make restudy selections by using post-hoc analyses in Experiments 1a and 1b, and hence we sought to (a) replicate them in Experiments 2, 3, and 4 and (b) provide preliminary evidence regarding factors that can (vs. cannot) account for them. Most important, evidence across all of the experiments supported the conclusion that people do use their CLJs to select categories for restudy.

Metacognitive judgments of prior material facilitate the learning of new material: The forward effect of metacognitive judgments in inductive learning.

“…Since the study by Jacoby et al (2010), many researchers started using CLJs to examine how accurately people judge how well they know categories and how people use their CLJs to adjust their subsequent study in category learning. For example, Morehead et al (2017) examined whether people used their judgments to regulate their learning of natural categories. Five experiments showed that people indeed used their CLJs for restudy selections.…”

Why Does Excellent Monitoring Accuracy Not Always Produce Gains in Memory Performance?

“…Despite the prior evidence, however, at least three recent studies have shown that some college students do the opposite (Morehead & Dunlosky, 2020;Morehead, Dunlosky, & Foster, 2017;Tullis & Benjamin, 2012); that is, they are more likely to choose to select to restudy materials with which they had given higher than lower JOLs. According to the contingent-efficacy hypothesis, selection strategies will moderate restudy efficacy (even when monitoring is highly accurate).…”

Do students make effective decisions when regulating their learning of categories?

“…We then used the honor-dishonor method to evaluate whether participants effectively regulated their learning. We also investigated whether these differences in selection moderated the honor-dishonor advantage, given that prior research has demonstrated individual differences in study selections (Morehead et al, 2017).…”

“…To form the two strategy groups, any participant with a correlation between CLJs and selections at or below Ϫ.5 was classified as selecting less-well-learned categories (honor ϭ 71 participants; dishonor ϭ 72 participants), and any participant with a correlation at or above .5 was classified as selecting more-well-learned categories (honor ϭ 18 participants; dishonor ϭ 22 participants). The conservative cutoffs of .5 and Ϫ.5 were chosen to increase the likelihood that participants within these groups applied either the strategy to select the less-well-learned categories or the morewell-learned ones (as in Morehead et al, 2017). Participants with a correlation between Ϫ.5 and .5 were excluded from further analyses because 1 Correlations could not be computed for 30 participants because these participants gave the same CLJ for every category.…”