Forgetting as the friend of learning: implications for teaching and self-regulated learning.

One of the "important peculiarities" of human learning (Bjork RA and Bjork EL. From Learning Processes to Cognitive Processes: Essays in Honor of William K. Estes, 1992, p. 35-67) is that certain conditions that produce forgetting-that is, impair access to some to-be-learned information studied earlier-also enhance the learning of that information when it is restudied. Such conditions include changing the environmental context from when some to-be-learned material is studied to when that material is restudied; increasing the delay from when something is studied to when it is tested or restudied; and interleaving, rather than blocking, the study or practice of the components of to-be-learned knowledge or skills. In this paper, we provide some conjectures as to why conditions that produce forgetting can also enable learning, and why a misunderstanding of this peculiarity of how humans learn can result in nonoptimal teaching and self-regulated learning.

The Critical Importance of Retrieval--and Spacing--for Learning.

We examined the impact of repeated testing and repeated studying on long-term learning. In Experiment 1, we replicated Karpicke and Roediger's (2008) influential results showing that once information can be recalled, repeated testing on that information enhances learning, whereas restudying that information does not. We then examined whether the apparent ineffectiveness of restudying might be attributable to the spacing differences between items that were inherent in the between-subjects design employed by Karpicke and Roediger. When we controlled for these spacing differences by manipulating the various learning conditions within subjects in Experiment 2, we found that both repeated testing and restudying improved learning, and that learners' awareness of the relative mnemonic benefits of these strategies was enhanced. These findings contribute to understanding how two important factors in learning-test-induced retrieval processes and spacing-can interact, and they illustrate that such interactions can play out differently in between-subjects and within-subjects experimental designs.

Do learners predict a shift from recency to primacy with delay?

The shift from recency to primacy with delay reflects a fundamental observation in the study of memory. As time passes, the accessibility of earlier-learned representations tends to increase relative to the accessibility of later-learned representations. In three experiments involving participants' memory for text materials, we examined whether participants understood that there might be such a shift with retention interval. In marked contrast to their actual performance, participants predicted recency effects at both shorter and longer retention intervals. Our findings add to the evidence that the storage and retrieval dynamics of the human memory system, though adaptive overall from a statistics-of-use standpoint, are both complex and poorly understood by users of the system.

Multiple-choice tests stabilize access to marginal knowledge.

Marginal knowledge refers to knowledge that is stored in memory, but is not accessible at a given moment. For example, one might struggle to remember who wrote The Call of the Wild, even if that knowledge is stored in memory. Knowing how best to stabilize access to marginal knowledge is important, given that new learning often requires accessing and building on prior knowledge. While even a single opportunity to restudy marginal knowledge boosts its later accessibility (Berger, Hall, & Bahrick, 1999), in many situations explicit relearning opportunities are not available. Our question is whether multiple-choice tests (which by definition expose the learner to the correct answers) can also serve this function and, if so, how testing compares to restudying given that tests can be particularly powerful learning devices (Roediger & Karpicke, 2006). In four experiments, we found that multiple-choice testing had the power to stabilize access to marginal knowledge, and to do so for at least up to a week. Importantly, such tests did not need to be paired with feedback, although testing was no more powerful than studying. Overall, the results support the idea that one's knowledge base is unstable, with individual pieces of information coming in and out of reach. The present findings have implications for a key educational challenge: ensuring that students have continuing access to information they have learned.

Desirable Difficulties in Vocabulary Learning.

In this article we discuss the role of desirable difficulties in vocabulary learning from two perspectives, one having to do with identifying conditions of learning that impose initial challenges to the learner but then benefit later retention and transfer, and the other having to do with the role of certain difficulties that are intrinsic to language processes, are engaged during word learning, and reflect how language is understood and produced. From each perspective we discuss evidence that supports the notion that difficulties in learning and imposed costs to language processing may produce benefits because they are likely to increase conceptual understanding. We then consider the consequences of these processes for actual second-language learning and suggest that some of the domain-general cognitive advantages that have been reported for proficient bilinguals may reflect difficulties imposed by the learning process, and by the requirement to negotiate cross-language competition, that are broadly desirable. As Alice Healy and her collaborators were perhaps the first to demonstrate, research on desirable difficulties in vocabulary and language learning holds the promise of bringing together research traditions on memory and language that have much to offer each other.

Multilevel induction of categories: venomous snakes hijack the learning of lower category levels.

The induction of categories and concepts from examples-which plays an important role in how people come to organize and understand the world-can happen at multiple levels, but how do competing values at these different levels affect learning? Using perceptually rich images of snakes, we asked participants to attend to either the snakes' specific genus or a broader categorization and then tested induction at both levels. We also varied the intrinsic value of the broader categorization (high value: whether the snake was venomous; low value: whether it was tropical). We found an interaction between study instruction and intrinsic value: Participants in the low-value condition were better able to induce the level they were instructed to attend to (i.e., genus or broader category) than to induce the level they were not instructed to attend to, whereas participants in the high-value condition, regardless of the level of categorization they were instructed to attend to, were significantly better at learning the broad categorization (for them, whether the snake was venomous) than were participants in the low-value condition. Our results suggest that intrinsically valuable features can disrupt the intentional learning of other, task-relevant information, but enhance the incidental learning of the same information.

Self-regulated learning: beliefs, techniques, and illusions.

Knowing how to manage one's own learning has become increasingly important in recent years, as both the need and the opportunities for individuals to learn on their own outside of formal classroom settings have grown. During that same period, however, research on learning, memory, and metacognitive processes has provided evidence that people often have a faulty mental model of how they learn and remember, making them prone to both misassessing and mismanaging their own learning. After a discussion of what learners need to understand in order to become effective stewards of their own learning, we first review research on what people believe about how they learn and then review research on how people's ongoing assessments of their own learning are influenced by current performance and the subjective sense of fluency. We conclude with a discussion of societal assumptions and attitudes that can be counterproductive in terms of individuals becoming maximally effective learners.

Equal spacing and expanding schedules in children's categorization and generalization.

To understand how generalization develops across the lifespan, researchers have examined the factors of the learning environment that promote the acquisition and generalization of categories. One such factor is the timing of learning events, which recent findings suggest may play a particularly important role in children's generalization. In the current study, we build on these findings by examining the impact of equally spaced versus expanding learning schedules on children's ability to generalize from studied exemplars of a given category to new exemplars presented on a later test. We found no significant effects of learning schedule when the generalization test was administered immediately after the learning phase, but there was a clear difference when the generalization test was delayed by 24h, with children in the expanding condition significantly outperforming children in the equally spaced learning condition. These results suggest that forgetting and retrieval dynamics may be lower level cognitive mechanisms promoting generalization and have several implications for broad theories of learning, cognition, and development.

Why does guessing incorrectly enhance, rather than impair, retention?

The finding that trying, and failing, to predict the upcoming to-be-remembered response to a given cue can enhance later recall of that response, relative to studying the intact cue-response pair, is surprising, especially given that the standard paradigm (e.g., Kornell, Hays, & Bjork, 2009) involves allocating what would otherwise be study time to generating an error. In three experiments, we sought to eliminate two potential heuristics that participants might use to aid recall of correct responses on the final test and to explore the effects of interference both at an immediate and at a delayed test. In Experiment 1, by intermixing strongly associated to-be-remembered pairs with weakly associated pairs, we eliminated a potential heuristic participants can use on the final test in the standard version of the paradigm-namely, that really strong associates are incorrect responses. In Experiment 2, by rigging half of the participants' responses to be correct, we eliminated another potential heuristic-namely, that one's initial guesses are virtually always wrong. In Experiment 3, we examined whether participants' ability to remember-and discriminate between-their incorrect guesses and correct responses would be lost after a 48-h delay, when source memory should be reduced. Across all experiments, we continued to find a robust benefit of trying to guess to-be-learned responses, even when incorrect, versus studying intact cue-response pairs. The benefits of making incorrect guesses are not an artifact of the paradigm, nor are they limited to short retention intervals.

Learning from errors versus explicit instruction in preparation for a test that counts.

BACKGROUND: Although the generation of errors has been thought, traditionally, to impair learning, recent studies indicate that, under particular feedback conditions, the commission of errors may have a beneficial effect. AIMS: This study investigates the teaching strategies that facilitate learning from errors. MATERIALS AND METHODS: This 2-year study, involving two cohorts of ~88 students each, contrasted a learning-from-errors (LFE) with an explicit instruction (EI) teaching strategy in a multi-session implementation directed at improving student performance on the high-stakes New York State Algebra 1 Regents examination. In the LFE condition, instead of receiving instruction on 4 sessions, students took mini-tests. Their errors were isolated to become the focus of 4 teacher-guided feedback sessions. In the EI condition, teachers explicitly taught the mathematical material for all 8 sessions. RESULTS: Teacher time-on in the LFE condition produced a higher rate of learning than did teacher time-on in the EI condition. The learning benefit in the LFE condition was, however, inconsistent across teachers. Second-by-second analyses of classroom activities, directed at isolating learning-relevant differences in teaching style revealed that a highly interactive mode of engaging the students in understanding their errors was more conducive to learning than was teaching directed at getting to the correct solution, either by lecturing about corrections or by interaction focused on corrections. CONCLUSION: These results indicate that engaging the students interactively to focus on errors, and the reasons for them, facilitates productive failure and learning from errors.

When disfluency is--and is not--a desirable difficulty: the influence of typeface clarity on metacognitive judgments and memory.

There are many instances in which perceptual disfluency leads to improved memory performance, a phenomenon often referred to as the perceptual-interference effect (e.g., Diemand-Yauman, Oppenheimer, & Vaughn (Cognition 118:111-115, 2010); Nairne (Journal of Experimental Psychology: Learning, Memory, and Cognition 14:248-255, 1988)). In some situations, however, perceptual disfluency does not affect memory (Rhodes & Castel (Journal of Experimental Psychology: General 137:615-625, 2008)), or even impairs memory (Glass, (Psychology and Aging 22:233-238, 2007)). Because of the uncertain effects of perceptual disfluency, it is important to establish when disfluency is a "desirable difficulty" (Bjork, 1994) and when it is not, and the degree to which people's judgments of learning (JOLs) reflect the consequences of processing disfluent information. In five experiments, our participants saw multiple lists of blurred and clear words and gave JOLs after each word. The JOLs were consistently higher for the perceptually fluent items in within-subjects designs, which accurately predicted the pattern of recall performance when the presentation time was short (Exps. 1a and 2a). When the final test was recognition or when the presentation time was long, however, we found no difference in recall for clear and blurred words, although JOLs continued to be higher for clear words (Exps. 2b and 3). When fluency was manipulated between subjects, neither JOLs nor recall varied between formats (Exp. 1b). This study suggests a boundary condition for the desirable difficulty of perceptual disfluency and indicates that a visual distortion, such as blurring a word, may not always induce the deeper processing necessary to create a perceptual-interference effect.

Why interleaving enhances inductive learning: the roles of discrimination and retrieval.

Kornell and Bjork (Psychological science 19:585-592, 2008) found that interleaving exemplars of different categories enhanced inductive learning of the concepts based on those exemplars. They hypothesized that the benefit of mixing exemplars from different categories is that doing so highlights differences between the categories. Kang and Pashler (Applied cognitive psychology 26:97-103, 2012) obtained results consistent with this discriminative-contrast hypothesis: Interleaving enhanced inductive learning, but temporal spacing, which does not highlight category differences, did not. We further tested the discriminative-contrast hypothesis by examining the effects of interleaving and spacing, as well as their combined effects. In three experiments, using photographs of butterflies and birds as the stimuli, temporal spacing was harmful when it interrupted the juxtaposition of interleaved categories, even when total spacing was held constant, supporting the discriminative-contrast hypothesis. Temporal spacing also had value, however, when it did not interrupt discrimination processing.

Going beyond the spacing effect: Does it matter how time on a task is distributed?

We assessed the effects of removing some constraints that characterise traditional experiments on the effects of spaced, rather than massed, study opportunities. In five experiments-using lists of to-be-remembered words-we examined the effects of how total study time was distributed across multiple repetitions of a given to-be-remembered word. Overall, within a given list, recall profited from study time being distributed (e.g., four 1-s presentations or two 2-s presentations vs one 4-s presentation). Among the implications of these findings is that if students choose to engage in massed studying (by virtue of constraints on their study time or a failure to appreciate the benefits of spaced study sessions), then studying the information twice but for half the time may produce memory benefits in a single study session.

Multiple-choice tests exonerated, at least of some charges: fostering test-induced learning and avoiding test-induced forgetting.

Among the criticisms of multiple-choice tests is that-by exposing the correct answer as one of the alternatives-such tests engage recognition processes rather than the productive retrieval processes known to enhance later recall. We tested whether multiple-choice tests could trigger productive retrieval processes-provided the alternatives were made plausible enough to enable test takers to retrieve both why the correct alternatives were correct and why the incorrect alternatives were incorrect. In two experiments, we found not only that properly constructed multiple-choice tests can indeed trigger productive retrieval processes, but also that they had one potentially important advantage over cued-recall tests. Both testing formats fostered retention of previously tested information, but multiple-choice tests also facilitated recall of information pertaining to incorrect alternatives, whereas cued-recall tests did not. Thus, multiple-choice tests can be constructed so that they exercise the very retrieval processes they have been accused of bypassing.

Using verification feedback to correct errors made on a multiple-choice test.

A key educational challenge is how to correct students' errors and misconceptions so that they do not persist. Simply labelling an answer as correct or incorrect on a short-answer test (verification feedback) does not improve performance on later tests; error correction requires receiving answer feedback. We explored the generality of this conclusion and whether the effectiveness of verification feedback depends on the type of test with which it is paired. We argue that, unlike for short-answer tests, learning whether one's multiple-choice selection is correct or incorrect should help participants narrow down the possible answers and identify specific lures as false. To test this proposition we asked participants to answer a series of general knowledge multiple-choice questions. They received no feedback, answer feedback, or verification feedback, and then took a short-answer test immediately and two days later. Verification feedback was just as effective as answer feedback for maintaining correct answers. Importantly, verification feedback allowed learners to correct more of their errors than did no feedback, although it was not as effective as answer feedback. Overall, verification feedback conveyed information to the learner, which has both practical and theoretical implications.

Improving conceptual learning via pretests.

Although examples can be structured to emphasize diagnostic features of concepts, novice learners tend to focus on irrelevant surface features and struggle to encode deeper structures. Experiment 1 examined whether pretesting-answering questions about content before it is studied-could enhance learners' noticing of diagnostic features, making them easier to process during subsequent study. Participants studied statistical concepts with examples that emphasized surface details or deep structure, and then classified new examples of these concepts. Studying examples that emphasized deep structure increased classification performance compared to examples that emphasized surface details. Moreover, taking pretests prior to studying the examples increased classification performance and eliminated differential benefits of studying structure versus surface examples. Experiment 2 examined whether pretesting serves a role beyond directing attention. After studying different statistical concepts with only surface-emphasizing examples, classification performance was better when participants actually took pretests compared to being given the correct responses. It is the generative aspect of pretesting, beyond attention directing, that improves conceptual learning among novice learners. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Where and how to learn: The interactive benefits of contextual variation, restudying, and retrieval practice for learning.

Students are often advised to do all of their studying in one good place, but restudying to-be-learned material in a new context can enhance subsequent recall. We examined whether there are similar benefits for testing. In Experiment 1 (n = 106), participants studied a 36-word list and 48 hr later-when back in the same or a new context-either restudied or recalled the list without feedback. After another 48 hr, all participants free-recalled the list in a new context. Experiment 2 (n = 203) differed by having the testing-condition participants restudy the list before being tested. Across both experiments, testing in a new context reduced recall, which carried over to the final test, whereas restudying in a new context did not impair (and in Experiment 2, significantly enhanced) recall. These findings reveal critical interactions between contextual-variation and retrieval-practice effects, which we interpret as consistent with a distribution-of-memory-strengths framework.

Optimizing retrieval as a learning event: when and why expanding retrieval practice enhances long-term retention.

Retrieving information from memory makes that information more recallable in the future than it otherwise would have been. Optimizing retrieval practice has been assumed, on the basis of evidence and arguments tracing back to Landauer and Bjork (1978), to require an expanding-interval schedule of successive retrievals, but recent findings suggest that expanding retrieval practice may be inferior to uniform-interval retrieval practice when memory is tested after a long retention interval. We report three experiments in which participants read educational passages and were then repeatedly tested, without feedback, after an expanding or uniform sequence of intervals. On a test 1 week later, recall was enhanced by the expanding schedule, but only when the task between successive retrievals was highly interfering with memory for the passage. These results suggest that the extent to which learners benefit from expanding retrieval practice depends on the degree to which the to-be-learned information is vulnerable to forgetting.

Conceptual and non-conceptual repetition priming in category exemplar generation: Evidence from bilinguals.

One measure of conceptual implicit memory is repetition priming in the generation of exemplars from a semantic category, but does such priming transfer across languages? That is, do the overlapping conceptual representations for translation equivalents provide a sufficient basis for such priming? In Experiment 1 (N=96) participants carried out a deep encoding task, and priming between languages was statistically reliable, but attenuated, relative to within-language priming. Experiment 2 (N=96) replicated the findings of Experiment 1 and assessed the contributions of conceptual and non-conceptual processes using a levels-of-processing manipulation. Words that underwent shallow encoding exhibited within-language, but not between-language, priming. Priming in shallow conditions cannot therefore be explained by incidental activation of the concept. Instead, part of the within-language priming effect, even under deep-encoding conditions, is due to increased availability of language-specific lemmas or phonological word forms.