Judgments of learning reflect the encoding of contexts, not items: evidence from a test of recognition exclusion.

Two sources of evidence seem to be shared by judgments of past recognition and judgments of future performance: item memory or familiarity (i.e., memory for the item independent of the context in which it was experienced) and context memory or recollection (i.e., memory for the context specific to a particular prior encounter). However, there are few studies investigating the link between these two putative memory processes and judgments of learning (JOLs). We tested memory and metamemory using a continuous exclusion procedure - a modified recognition memory task where study events for two classes of items are interleaved with test trials in which the subject must endorse items from one class and reject items from the other. This procedure allowed us to estimate the influences of memory for context and memory for item on JOLs and licenses conclusions about the relative role of item and context information in supporting JOLs. An analysis of forgetting revealed that JOLs reflect both the initial degree of learning and the rate of forgetting, but only of memory for context and not of memory for items. These findings suggest that JOLs are predictive of memory for context-bound episodes, rather than for the semantic content of those episodes.

Strategic regulation of memory in dsyphoria: a quantity-accuracy profile analysis.

The mechanisms underlying a tendency among individuals with depression to report personal episodic memories with low specificity remain to be understood. We assessed a sample of undergraduate students with dysphoria to determine whether depression relates to a broader dysregulation of balancing accuracy and informativeness during memory reports. Specifically, we investigated metamnemonic processes using a quantity-accuracy profile approach. Recall involved three phases with increasing allowance for more general, or coarse-grained, responses: (a) forced-precise responding, requiring high precision; (b) free-choice report with high and low penalty incentives on accuracy; (c) a lexical description phase. Individuals with and without dysphoria were largely indistinguishable across indices of retrieval, monitoring, and control aspects of metamemory. The results indicate intact metacognitive processing in young individuals with dysphoria and provide no support for the view that impaired metacognitive control underlies either memory deficits or bias in memory reports that accompany dysphoria.

Metamnemonic predictions of lineup identification.

After a crime is committed, investigators may query witnesses about whether they believe they will be to identify the perpetrator. However, we know little about how such metacognitive judgments are related to performance on a subsequent lineup identification task. The extant research has found the strength of this relationship to be small or nonexistent, which conflicts with the large body of literature indicating a moderate relationship between predictions and performance on memory tasks. In Studies 1-3, we induce variation in encoding quality by having participants watch a mock crime video with either low, medium, or high exposure quality, and then assess their future lineup performance. Calibration analysis revealed that assessments of future lineup performance were predictive of identification accuracy. This relationship was driven primarily by poor performance following low assessments. Studies 4 and 5 showed that these predictions are not based on a witness's evaluation of their encoding experience, nor on a contemporaneous assessment of memory strength. These results reinforce the argument that variation in memory quality is needed to obtain reliable relationships between predictions and performance. An unexpected finding is that witnesses who made a prediction shortly after encoding evinced superior memory compared to those who made a prediction later.

Memory Fidelity Reveals Qualitative Changes in Interactions Between Items in Visual Working Memory.

Memory for objects in a display sometimes reveals attraction-the objects are remembered as more similar to one another than they actually were-and sometimes reveals repulsion-the objects are remembered as more different from one another. The conditions that lead to these opposing memory biases are poorly understood; there is no theoretical framework that explains these contrasting dynamics. In three experiments (each N = 30 adults), we demonstrate that memory fidelity provides a unifying dimension that accommodates the existence of both types of visual working memory interactions. We show that either attraction or repulsion can arise simply as a function of manipulations of memory fidelity. We also demonstrate that subjective ratings of fidelity predict the presence of attraction or repulsion on a trial-by-trial basis. We discuss how these results bear on computational models of visual working memory and contextualize these results within the literature of attraction and repulsion effects in long-term memory and perception.

Is there a K in capacity? Assessing the structure of visual short-term memory.

Visual short-term memory (VSTM) is a cognitive structure that temporarily maintains a limited amount of visual information in the service of current cognitive goals. There is active theoretical debate regarding how limits in VSTM should be construed. According to discrete-slot models of capacity, these limits are set in terms of a discrete number of slots that store individual objects in an all-or-none fashion. According to alternative continuous resource models, the limits of VSTM are set in terms of a resource that can be distributed to bolster some representations over others in a graded fashion. Hybrid models have also been proposed. We tackled the classic question of how to construe VSTM structure in a novel way, by examining how contending models explain data within traditional VSTM tasks and also how they generalize across different VSTM tasks. Specifically, we fit theoretical ROCs derived from a suite of models to two popular VSTM tasks: a change detection task in which participants had to remember simple features and a rapid serial visual presentation task in which participants had to remember real-world objects. In 3 experiments we assessed the fit and predictive ability of each model and found consistent support for pure resource models of VSTM. To gain a fuller understanding of the nature of limits in VSTM, we also evaluated the ability of these models to jointly model the two tasks. These joint modeling analyses revealed additional support for pure continuous-resource models, but also evidence that performance across the two tasks cannot be captured by a common set of parameters. We provide an interpretation of these signal detection models that align with the idea that differences among memoranda and across encoding conditions alter the memory signal of representations in VSTM.

Long-term inference and memory following retrieval practice.

One exceptional characteristic of the testing effect is its generalizability over time and circumstance. The benefits of testing over rote restudy appear to grow with time, as forgetting occurs, and also have been documented to extend to tasks of inference on previously unstudied stimuli. In the two experiments reported here, we evaluated inference and memory for members of natural categories over time. Rote memory and generalization were tested shortly after the study phase and again after varying delays. Results from both experiments indicate that retrieval practice does indeed enhance inference for novel members of previously learned categories, and that the benefits are maintained over the duration of our experiments-up to 25 days. An analysis of forgetting rates indicates that retrieval practice does not, however, decelerate forgetting when compared with restudy. Rates of forgetting were not discernibly different, for either rote memory or conceptual knowledge, between the two conditions. These results indicate that although testing does not appear to reduce forgetting, it is a potent means of enhancing inference, and the benefits to memory and inference are long lasting.

The role of retrieval during study: Evidence of reminding from self-paced study time.

The reminding effect (Tullis, Benjamin, & Ross, Journal of Experimental Psychology: General, 143[4], 1526-1540, 2014) describes the increase in recall of a studied word when a related word is presented later in the study list. However, because the process of reminding is thought to occur during study, measures of test performance are indirect indicators of the process of reminding and are subject to influences that arise during testing. The present research seeks evidence of reminding during encoding. In two experiments, self-paced study times were used to index the online process of reminding. In Experiment 1, pairs of repeated words, related words, and unrelated words were included in a study list. Study times were shorter for words related to prior words in the list, but only when the lag between those two words was short. Relatedness affected study time by inspiring a reduction in the threshold for termination of study for related words under massed conditions. Experiment 2 replicated the reduction in study time for related words and further showed that the study time allotted to an associate of an earlier item predicted better memory for that earlier word on a cued-recall test. In this experiment, an advantage in memory was observed for related words, and self-paced study time of one word during encoding was predictive of later memory for a related word. These results suggest a link between the action of reminding at study, as indexed by changes in the distribution of study time, and later benefits to remembering, as revealed by the reminding effect.

Monitoring the ebb and flow of attention: Does controlling the onset of stimuli during encoding enhance memory?

Central to the operation of the Atkinson and Shiffrin's (Psychology of learning and motivation, 2, 89-195, 1968) model of human memory are a variety of control processes that manage information flow. Research on metacognition reveals that provision of control in laboratory learning tasks is generally beneficial to memory. In this paper, we investigate the novel domain of attentional fluctuations during study. If learners are able to monitor attention, then control over the onset of stimuli should also improve performance. Across four experiments, we found no evidence that control over the onset of stimuli enhances learning. This result stands in notable contrast to the fact that control over stimulus offset does enhance memory (Experiment 1; Tullis & Benjamin, Journal of memory and language, 64 (2), 109-118, 2011). This null finding was replicated across laboratory and online samples of subjects, and with both words and faces as study material. Taken together, the evidence suggests that people either cannot monitor fluctuations in attention effectively or cannot precisely time their study to those fluctuations.

The joint contribution of participation and performance to learning functions: Exploring the effects of age in large-scale data sets.

Large-scale data sets from online training and game platforms offer the opportunity for more extensive and more precise investigations of human learning than is typically achievable in the laboratory. However, because people make their own choices about participation, any investigation into learning using these data sets must simultaneously model performance-that is, the learning function-and participation. Using a data set of 54 million gameplays from the online brain training site Lumosity, we show that learning functions of participants are systematically biased by participation policies that vary with age. Older adults who are poorer performers are more likely to drop out than older adults who perform well. Younger adults show no such effect. Using this knowledge, we can extrapolate group learning functions that correct for these age-related differences in dropout.

Updating metacognitive control in response to expected retention intervals.

In five experiments, we investigated whether expected retention intervals affect subjects' encoding strategies. In the first four experiments, our subjects studied paired associates consisting of words from the Graduate Record Exam and a synonym. They were told to expect a test on a word pair after either a short or a longer interval. Subjects were tested on most pairs after the expected retention interval. For some pairs, however, subjects were tested after the other retention interval, allowing for a comparison of performance at a given retention interval conditional upon the expected retention interval. No effect of the expected retention interval was found for 1 min versus 4 min (Exp. 1), 30 s versus 3 min (Exp. 2), and 30 s versus 10 min (Exps. 3 and 4), even when subjects were given complete control over the pacing of study items (Exp. 4). However, when the difference between the expected retention intervals was increased massively (10 min vs. 24 h; Exp. 5), subjects remembered more items that they expected to be tested sooner, an effect consistent with the idea that they traded off efforts to remember items for the later test versus items that were about to be tested. Overall, this set of results accords with much of the test-expectancy literature, revealing that subjects are often reluctant to adjust encoding strategies on an item-by-item basis, and when they do, they usually make quantitative, rather than qualitative, adjustments.

Memory for conversation and the development of common ground.

Efficient conversation is guided by the mutual knowledge, or common ground, that interlocutors form as a conversation progresses. Characterized from the perspective of commonly used measures of memory, efficient conversation should be closely associated with item memory-what was said-and context memory-who said what to whom. However, few studies have explicitly probed memory to evaluate what type of information is maintained following a communicative exchange. The current study examined how item and context memory relate to the development of common ground over the course of a conversation, and how these forms of memory vary as a function of one's role in a conversation as speaker or listener. The process of developing common ground was positively related to both item and context memory. In addition, content that was spoken was remembered better than content that was heard. Our findings illustrate how memory assessments can complement language measures by revealing the impact that basic conversational processes have on memory for what has been discussed. By taking this approach, we show that not only does the process of forming common ground facilitate communication in the present, but it also promotes an enduring record of that event, facilitating conversation into the future.

The influences of valence and arousal on judgments of learning and on recall.

Much is known about how the emotional content of words affects memory for those words, but only recently have researchers begun to investigate whether emotional content influences metamemory-that is, learners' assessments of what is or is not memorable. The present study replicated recent work demonstrating that judgments of learning (JOLs) do indeed reflect the superior memorability of words with emotional content. We further contrasted two hypotheses regarding this effect: a physiological account in which emotional words are judged to be more memorable because of their arousing properties, versus a cognitive account in which emotional words are judged to be more memorable because of their cognitive distinctiveness. Two results supported the latter account. First, both normed arousal (Exp. 1) and normed valence (Exp. 2) independently influenced JOLs, even though only an effect of arousal would be expected under a physiological account. Second, emotional content no longer influenced JOLs in a design (Exp. 3) that reduced the primary distinctiveness of emotional words by using a single list of words in which normed valence and arousal were varied continuously. These results suggest that the metamnemonic benefit of emotional words likely stems from cognitive factors.

Using the Internet to access information inflates future use of the Internet to access other information.

The ways in which people learn, remember, and solve problems have all been impacted by the Internet. The present research explored how people become primed to use the Internet as a form of cognitive offloading. In three experiments, we show that using the Internet to retrieve information alters a person's propensity to use the Internet to retrieve other information. Specifically, participants who used Google to answer an initial set of difficult trivia questions were more likely to decide to use Google when answering a new set of relatively easy trivia questions than were participants who answered the initial questions from memory. These results suggest that relying on the Internet to access information makes one more likely to rely on the Internet to access other information.

Conflict and metacognitive control: the mismatch-monitoring hypothesis of how others' knowledge states affect recall.

Information about others' success in remembering is frequently available. For example, students taking an exam may assess its difficulty by monitoring when others turn in their exams. In two experiments, we investigated how rememberers use this information to guide recall. Participants studied paired associates, some semantically related (and thus easier to retrieve) and some unrelated (and thus harder). During a subsequent cued recall test, participants viewed fictive information about an opponent's accuracy on each item. In Experiment 1, participants responded to each cue once before seeing the opponent's performance and once afterwards. Participants reconsidered their responses least often when the opponent's accuracy matched the item difficulty (easy items the opponent recalled, hard items the opponent forgot) and most often when the opponent's accuracy and the item difficulty mismatched. When participants responded only after seeing the opponent's performance (Experiment 2), the same mismatch conditions that led to reconsideration even produced superior recall. These results suggest that rememberers monitor whether others' knowledge states accord or conflict with their own experience, and that this information shifts how they interrogate their memory and what they recall.

Cueing others' memories.

Many situations require us to generate external cues to support later retrieval from memory. For instance, we create file names in order to cue our memory to a file's contents, and instructors create lecture slides to remember what points to make during classes. We even generate cues for others when we remind friends of shared experiences or send colleagues a computer file that is named in such a way so as to remind them of its contents. Here we explore how and how well learners tailor retrieval cues for different intended recipients. Across three experiments, subjects generated verbal cues for a list of target words for themselves or for others. Learners generated cues for others by increasing the normative cue-to-target associative strength but also by increasing the number of other words their cues point to, relative to cues that they generated for themselves. This strategy was effective: such cues supported higher levels of recall for others than cues generated for oneself. Generating cues for others also required more time than generating cues for oneself. Learners responded to the differential demands of cue generation for others by effortfully excluding personal, episodic knowledge and including knowledge that they estimate to be broadly shared.

Cue generation: How learners flexibly support future retrieval.

The successful use of memory requires us to be sensitive to the cues that will be present during retrieval. In many situations, we have some control over the external cues that we will encounter. For instance, learners create shopping lists at home to help remember what items to later buy at the grocery store, and they generate computer file names to help remember the contents of those files. Generating cues in the service of later cognitive goals is a complex task that lies at the intersection of metacognition, communication, and memory. In this series of experiments, we investigated how and how well learners generate external mnemonic cues. Across 5 experiments, learners generated a cue for each target word in a to-be-remembered list and received these cues during a later cued recall test. Learners flexibly generated cues in response to different instructional demands and study list compositions. When generating mnemonic cues, as compared to descriptions of target items, learners produced cues that were more distinct than mere descriptions and consequently elicited greater cued recall performance than those descriptions. When learners were aware of competing targets in the study list, they generated mnemonic cues with smaller cue-to-target associative strength but that were even more distinct. These adaptations led to fewer confusions among competing targets and enhanced cued recall performance. These results provide another example of the metacognitively sophisticated tactics that learners use to effectively support future retrieval.

Retrieval speeds context fluctuation: why semantic generation enhances later learning but hinders prior learning.

In recent work, retrieval has been shown to enhance memory for events following that retrieval. In this set of experiments, we examined the effects of interleaved semantic retrieval on both previous and future learning within a multilist learning paradigm. Interleaved retrieval led to enhanced memory for lists learned following retrieval. In contrast, memory was impaired for lists learned prior to retrieval (Experiment 1). These results are consistent with recent work in multilist learning, directed forgetting, and list-before-last retrieval, all of which indicate a crucial role for retrieval in enhancing mental list segregation. This pattern of results follows clearly from a theoretical perspective in which retrieval drives internal contextual change and in which contextual overlap between study and test promotes better memory. Consistent with that perspective, a 15-min delay before the final test eliminated both effects (Experiment 2). Experiment 2 replicated the results of Experiment 1 with materials and assessments more appropriate for educational settings: Interleaved semantic retrieval led learners to be more able to answer questions correctly about texts studied after a retrieval event but less able to do so for texts studied earlier.

Self-pacing study of faces of different races: metacognitive control over study does not eliminate the cross-race recognition effect.

People often recognize same-race faces better than other-race faces. This cross-race effect (CRE) has been proposed to arise in part because learners devote fewer cognitive resources to encode faces of social out-groups. In three experiments, we evaluated whether learners' other-race mnemonic deficits are due to "cognitive disregard" during study and whether this disregard is under metacognitive control. Learners studied each face either for as long as they wanted (the self-paced condition) or for the average time taken by a self-paced learner (the fixed-rate condition). Self-paced learners allocated equal amounts of study time to same-race and other-race faces, and having control over study time did not change the size of the CRE. In the second and third experiments, both self-paced and fixed-rate learners were given instructions to "individuate" other-race faces. Individuation instructions caused self-paced learners to allocate more study time to other-race faces, but this did not significantly reduce the size of the CRE, even for learners who reported extensive contact with other races. We propose that the differential processing that people apply to faces of different races and the subsequent other-race mnemonic deficit are not due to learners' strategic cognitive disregard of other-race faces.

State-based metacognition: how time of day affects the accuracy of metamemory.

Although there is an abundance of research on how stimulus characteristics and encoding conditions affect metamemory, and how those effects either do or do not mirror effects on memory, there is little research on whether and how characteristics of participants' states-like mood, fatigue, or hunger-affect metamemory. The present study examined whether metamemory ability fluctuates with time of day. Specifically, we evaluated whether learners can successfully account for the effects of time of day on their memory, and whether metacognitive monitoring is more accurate at an individual's optimal time of day. Young adults studied and recalled lists of words in both the morning and the afternoon, providing various metamemory judgements during each test session. We replicated the finding that young participants recalled more words in the afternoon than in the morning. Prior to study, participants did not predict superior recall in the afternoon, but they did after they had an opportunity to study the list (but before the test on that material). We also found that item-by-item predictions were more accurate in the afternoon, suggesting that self-regulated learning might benefit from being scheduled during times of day that accord with individuals' peak arousal.

Metacognition of the testing effect: guiding learners to predict the benefits of retrieval.

If the mnemonic benefits of testing are to be widely realized in real-world learning circumstances, people must appreciate the value of testing and choose to utilize testing during self-guided learning. Yet metacognitive judgments do not appear to reflect the enhancement provided by testing Karpicke & Roediger (Science 319:966-968, 2008). In this article, we show that under judicious conditions, learners can indeed reveal an understanding of the beneficial effects of testing, as well as the interaction of that effect with delay (experiment 1). In that experiment, subjects made judgments of learning (JOLs) for previously studied or previously tested items in either a cue-only or a cue-target context, and either immediately or after a 1-day delay. When subjects made judgments in a cue-only context, their JOLs accurately reflected the effects of testing, both immediately and at a delay. To evaluate the potential of exposure to such conditions for promoting generalized appreciation of testing effects, three further experiments elicited global predictions about restudied and tested items across two study/test cycles (experiments 2, 3, and 4). The results indicated that learners' global naïve metacognitive beliefs increasingly reflect the beneficial effects of testing when learners experience these benefits with increasing external support. If queried under facilitative circumstances, learners appreciate the mnemonic enhancement that testing provides on both an item-by-item and global basis but generalize that knowledge to future learning only with considerable guidance.