Getting value out of working memory through strategic prioritisation; implications for storage and control.

Working memory is an active system responsible for "the temporary maintenance and processing of information in the support of cognition and action" (Baddeley et al., 2021). In keeping with this, a growing body of research has explored the close links between working memory and attention, and how these might be harnessed to impact performance and possibly improve working memory efficiency. This is theoretically and practically important, given that working memory is a central hub in complex cognition yet is extremely capacity- and resource-limited. We review work carried out over the last ten years or so looking at how high 'value' items in working memory can be strategically prioritised through selective attention, drawing principally from visual working memory paradigms with young adult participants, while also discussing how the core effects extend to different task domains and populations. A consistent set of core findings emerges, with improved memory for items that are allocated higher 'value' but no change in overall task performance, and a recency advantage regardless of point allocation when items are encountered sequentially. Value-directed prioritisation is effortful, under top-down strategic control, and appears to vary with perceptual distraction and executive load. It is driven by processes operating during encoding, maintenance, and retrieval, though the extent to which these are influenced by different features of the task context remain to be mapped out. We discuss implications for working memory, attention, and strategic control, and note some possible future directions of travel for this promising line of research.

Visual working memory phenomena based on categorical tasks replicate using a continuous measure: A simple interpretation and some methodological considerations.

An increasingly popular method for investigating visuospatial working memory assumes stored features of objects such as color and orientation vary along continua subject to internal noise. It adapts the stimulus adjustment procedure from perceptual psychophysics to assess the precision with which stored features are represented in memory. This contrasts with methods using discrete, categorical measures of feature retention. The current study examined the replicability of some phenomena documented using conventional methodology when assessed using a continuous measure of feature recall. These concern memory for a short series of objects and include effects of recency, prioritizing an individual object, and presenting an irrelevant additional object after the last item (a poststimulus 'suffix'). In two experiments we find broadly similar results using a continuous measure of color-orientation binding to those obtained previously using categorical measures, with small differences we regard as minor. We interpret the convergence between methods in terms of a simple analogy between categorical memory and categorical perception whereby categorical retrieval involves the application of a discrete criterion to an underlying continuum of stored feature information. We conclude by discussing some of the advantages and limitations of continuous and categorical measures of retention.

Translating words into actions in working memory: The role of spatial-motoric coding.

Research from a working memory perspective on the encoding and temporary maintenance of sequential instructions has established a consistent advantage for enacted over verbal recall. This is thought to reflect action planning for anticipated movements at the response phase. We describe five experiments investigating this, comparing verbal and enacted recall of a series of action-object pairings under different potentially disruptive concurrent task conditions, all requiring repetitive movements. A general advantage for enacted recall was observed across experiments, together with a tendency for concurrent action to impair sequence memory performance. The enacted recall advantage was reduced by concurrent action for both fine and gross concurrent movement with the degree of disruption influenced by both the complexity and the familiarity of the movement. The results are discussed in terms of an output buffer store of limited capacity capable of holding motoric plans for anticipated action.

Detecting accelerated long-term forgetting: A problem and some solutions.

While many memory disorders occur with normal rates of forgetting, an accelerated rate of long-term forgetting (ALF) may occur, sometimes in the absence of a learning deficit. Detecting ALF presents a problem as it is desirable that the learned material is re-tested after each of several delays. This may result in earlier retrievals confounding later tests, with evidence suggesting that both positive and negative interaction can occur between successive tests. An earlier study (Baddeley et al., 2019) tested cued recall of a series of four crimes or four visual scenes by probing a different sample of features from all four crimes/scenes at each delay. Even though no question was asked twice, the interpolated tests markedly reduced the rate of forgetting. We suggest that this decelerated forgetting effect may result from the retrieval of probed features activating other associated features within that episode, hence facilitating their recall on subsequent tests. If so, the effect should be removed when only single and separate episodes, or individual items, are tested at each delay. We test this by probing a separate episode at each delay (Experiment 1), or by replacing integrated episodes with recognition memory for isolated words (Experiments 2 and 3) or visual scenes (Experiments 4 and 5). As predicted, we find no reduction in the rate of forgetting, in contrast to our earlier studies. The theoretical and clinical implications of our results are discussed. We conclude that the previously developed Crimes and Four Doors Tests (Baddeley et al., 2019) and the present single item recognition tests are complementary and are both likely to be necessary to ensure the reliable detection of ALF.

Can valuable information be prioritized in verbal working memory?

Though there is substantial evidence that individuals can prioritize more valuable information in visual working memory (WM), little research has examined this in the verbal domain. Four experiments were conducted to investigate this and the conditions under which effects emerge. In each experiment, participants listened to digit sequences and then attempted to recall them in the correct order. At the start of each block, participants were either told that all items were of equal value, or that an item at a particular serial position was worth more points. Recall was enhanced for these higher value items (Experiment 1a), a finding that was replicated while rejecting an alternative account based on distinctiveness (Experiment 1b). Thus, valuable information can be prioritized in verbal WM. Two further experiments investigated whether these boosts remained when participants completed a simple concurrent task disrupting verbal rehearsal (Experiment 2), or a complex concurrent task disrupting verbal rehearsal and executive resources (Experiment 3). Under simple concurrent task conditions, prioritization boosts were observed, but with increased costs to the less valuable items. Prioritization effects were also observed under complex concurrent task conditions, although this was accompanied by chance-level performance at most of the less valuable positions. A substantial recency advantage was also observed for the final item in each sequence, across all conditions. Taken together, this indicates that individuals can prioritize valuable information in verbal WM even when rehearsal and executive resources are disrupted, though they do so by neglecting or abandoning other items in the sequence. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Attention and binding in visual working memory: Two forms of attention and two kinds of buffer storage.

We review our research on the episodic buffer in the multicomponent model of working memory (Baddeley, 2000), making explicit the influence of Anne Treisman's work on the way our research has developed. The crucial linking theme concerns binding, whereby the individual features of an episode are combined as integrated representations. We summarize a series of experiments on visual working memory that investigated the retention of feature bindings and individual features. The effects of cognitive load, perceptual distraction, prioritization, serial position, and their interactions form a coherent pattern. We interpret our findings as demonstrating contrasting roles of externally driven and internally driven attentional processes, as well as a distinction between visual buffer storage and the focus of attention. Our account has strong links with Treisman's concept of focused attention and aligns with a number of contemporary approaches to visual working memory.

Charting the trajectory of forgetting: Insights from a working memory period paradigm.

Working memory capacity is commonly measured in terms of its item span, and much less often in terms of its time span, or "period." The former measures how many items can be stored in working memory when carrying out episodes of concurrent processing. The latter complements this by determining the duration of processing episodes that can be tolerated while successfully storing a fixed number of items. We investigated the generality of previous evidence that working memory period varies with the distribution of longer and shorter processing episodes within a trial, and that notwithstanding such differences, a global measure of period is a reliable predictor of children's educational attainment. We describe data from 184 children, between 7 and 11 years of age, who completed variants of an operation period task with different distributions of processing episodes together with measures of scholastic attainment. Individual differences in period scores were consistent over two test sessions, and were predictive of reading and number skills. We replicated previous effects of the order of longer and shorter processing episodes, but found that they did not generalize fully to other manipulations of order. The results point to the contribution of subtle within-trial sequence configurations for working memory. We make the case for a broader view of what constrains working memory than exists in current models.

The phonological loop as a buffer store: An update.

We regard our multicomponent model of working memory as reflecting a hierarchy of buffer stores with buffer storage providing an effective way of combining information from two or more streams that may differ in either the speed of input or in the features coded. We illustrate this through the case of the phonological loop component of the model. We discuss its gradual development through a combination of evidence from mainstream cognition and neuropsychology with the need for more detailed modelling of issues such as the representation of serial order. A brief account follows of the application, beyond the laboratory and clinic, of the concept of a phonological loop and the methods designed to study it. We then discuss some criticisms of the overall multicomponent model, concluding with a discussion of the major contribution made by neuropsychological evidence to its development together with some suggestions as to comparative lack of influence from more recent studies based on neuro-imaging.

From short-term store to multicomponent working memory: The role of the modal model.

The term "modal model" reflects the importance of Atkinson and Shiffrin's paper in capturing the major developments in the cognitive psychology of memory that were achieved over the previous decade, providing an integrated framework that has formed the basis for many future developments. The fact that it is still the most cited model from that period some 50 years later has, we suggest, implications for the model itself and for theorising in psychology more generally. We review the essential foundations of the model before going on to discuss briefly the way in which one of its components, the short-term store, had influenced our own concept of a multicomponent working memory. This is followed by a discussion of recent claims that the concept of a short-term store be replaced by an interpretation in terms of activated long-term memory. We present several reasons to question these proposals. We conclude with a brief discussion of the implications of the longevity of the modal model for styles of theorising in cognitive psychology.

Benchmarks for models of short-term and working memory.

Any mature field of research in psychology-such as short-term/working memory-is characterized by a wealth of empirical findings. It is currently unrealistic to expect a theory to explain them all; theorists must satisfice with explaining a subset of findings. The aim of the present article is to make the choice of that subset less arbitrary and idiosyncratic than is current practice. We propose criteria for identifying benchmark findings that every theory in a field should be able to explain: Benchmarks should be reproducible, generalize across materials and methodological variations, and be theoretically informative. We propose a set of benchmarks for theories and computational models of short-term and working memory. The benchmarks are described in as theory-neutral a way as possible, so that they can serve as empirical common ground for competing theoretical approaches. Benchmarks are rated on three levels according to their priority for explanation. Selection and ratings of the benchmarks is based on consensus among the authors, who jointly represent a broad range of theoretical perspectives on working memory, and they are supported by a survey among other experts on working memory. The article is accompanied by a web page providing an open forum for discussion and for submitting proposals for new benchmarks; and a repository for reference data sets for each benchmark. (PsycINFO Database Record

Benchmarks provide common ground for model development: Reply to Logie (2018) and Vandierendonck (2018).

We respond to the comments of Logie and Vandierendonck to our article proposing benchmark findings for evaluating theories and models of short-term and working memory. The response focuses on the two main points of criticism: (a) Logie and Vandierendonck argue that the scope of the set of benchmarks is too narrow. We explain why findings on how working memory is used in complex cognition, findings on executive functions, and findings from neuropsychological case studies are currently not included in the benchmarks, and why findings with visual and spatial materials are less prevalent among them. (b) The critics question the usefulness of the benchmarks and their ratings for advancing theory development. We explain why selecting and rating benchmarks is important and justifiable, and acknowledge that the present selection and rating decisions are in need of continuous updating. The usefulness of the benchmarks of all ratings is also enhanced by our concomitant online posting of data for many of these benchmarks. (PsycINFO Database Record

Are there multiple ways to direct attention in working memory?

In visual working memory tasks, memory for an item is enhanced if participants are told that the item is relatively more valuable than others presented within the same trial. Experiment 1 explored whether these probe value boosts (termed prioritization effects in previous literature) are affected by probe frequency (i.e., how often the more valuable item is tested). Participants were presented with four colored shapes sequentially and asked to recall the color of one probed item following a delay. They were informed that the first item was more valuable (differential probe value) or as valuable as the other items (equal probe value), and that this item would be tested more frequently (differential probe frequency) or as frequently (equal probe frequency) as the other items. Probe value and probe frequency boosts were observed at the first position, though both were accompanied by costs to other items. Probe value and probe frequency boosts were additive, suggesting the manipulations yield independent effects. Further supporting this, experiment 2 revealed that probe frequency boosts are not reliant on executive resources, directly contrasting with previous findings regarding probe value. Taken together, these outcomes suggest there may be several ways in which attention can be directed in working memory.

Is the phonological similarity effect in working memory due to proactive interference?

Immediate serial recall of verbal material is highly sensitive to impairment attributable to phonological similarity. Although this has traditionally been interpreted as a within-sequence similarity effect, Engle (2007) proposed an interpretation based on interference from prior sequences, a phenomenon analogous to that found in the Peterson short-term memory (STM) task. We use the method of serial reconstruction to test this in an experiment contrasting the standard paradigm in which successive sequences are drawn from the same set of phonologically similar or dissimilar words and one in which the vowel sound on which similarity is based is switched from trial to trial, a manipulation analogous to that producing release from PI in the Peterson task. A substantial similarity effect occurs under both conditions although there is a small advantage from switching across similar sequences. There is, however, no evidence for the suggestion that the similarity effect will be absent from the very first sequence tested. Our results support the within-sequence similarity rather than a between-list PI interpretation. Reasons for the contrast with the classic Peterson short-term forgetting task are briefly discussed. (PsycINFO Database Record

Competition for the focus of attention in visual working memory: perceptual recency versus executive control.

Previous research on memory for a short sequence of visual stimuli indicates that access to the focus of attention (FoA) can be achieved in either of two ways. The first is automatic and is indexed by the recency effect, the enhanced retention of the final item. The second is strategic and based on instructions to prioritize items differentially, a process that draws on executive capacity and boosts retention of information deemed important. In both cases, the increased level of retention can be selectively reduced by presenting a poststimulus distractor (or suffix). We manipulated these variables across three experiments. Experiment 1 generalized previous evidence that prioritizing a single item enhances its retention and increases its vulnerability to interference from a poststimulus suffix. A second experiment showed that the enhancement from prioritizing one or two items comes at a cost to the recency effect. A third experiment showed that prioritizing two items renders memory for both vulnerable to interference from an irrelevant suffix. The results suggest that some but not all items in working memory compete to occupy a narrow FoA and that this competition is determined by a combination of perceptually driven recency and internal executive control.

How is the serial order of a visual sequence represented? Insights from transposition latencies.

A central goal of research on short-term memory (STM) over the past 2 decades has been to identify the mechanisms that underpin the representation of serial order, and to establish whether these mechanisms are the same across different modalities and domains (e.g., verbal, visual, spatial). A fruitful approach to addressing this question has involved comparing the transposition error latency predictions of models built from different candidate mechanisms for representing serial order. Experiments involving the output-timed serial recall of sequences of verbal (Farrell & Lewandowsky, 2004) and spatial (Hurlstone & Hitch, 2015) items have revealed an error latency profile consistent with the prediction of a competitive queuing mechanism within which serial order is represented via a primacy gradient of activations over items, associations between items and position markers, with suppression of items following recall. In this paper, we extend this chronometric analysis of recall errors to the serial recall of sequences of visual, nonspatial, items and find across 3 experiments an error latency profile broadly consistent with the prediction of the same representational mechanism. The findings suggest that common mechanisms and principles contribute to the representation of serial order across the verbal, visual, and spatial STM domains. The implications of these findings for theories of short-term and working memory are considered. (PsycINFO Database Record

The limits of visual working memory in children: Exploring prioritization and recency effects with sequential presentation.

Recent research has demonstrated that, when instructed to prioritize a serial position in visual working memory (WM), adults are able to boost performance for this selected item, at a cost to nonprioritized items (e.g., Hu, Hitch, Baddeley, Zhang, & Allen, 2014). While executive control appears to play an important role in this ability, the increased likelihood of recalling the most recently presented item (i.e., the recency effect) is relatively automatic, possibly driven by perceptual mechanisms. In 3 Experiments 7 to 10 year-old's ability to prioritize items in WM was investigated using a sequential visual task (total N = 208). The relationship between individual differences in WM and performance on the experimental task was also explored. Participants were unable to prioritize the first (Experiments 1 and 2) or final (Experiment 3) item in a 3-item sequence, while large recency effects for the final item were consistently observed across all experiments. The absence of a priority boost across 3 experiments indicates that children may not have the necessary executive resources to prioritize an item within a visual sequence, when directed to do so. In contrast, the consistent recency boosts for the final item indicate that children show automatic memory benefits for the most recently encountered stimulus. Finally, for the baseline condition in which children were instructed to remember all 3 items equally, additional WM measures predicted performance at the first and second but not the third serial position, further supporting the proposed automaticity of the recency effect in visual WM. (PsycINFO Database Record

Executive and perceptual distraction in visual working memory.

The contents of visual working memory are likely to reflect the influence of both executive control resources and information present in the environment. We investigated whether executive attention is critical in the ability to exclude unwanted stimuli by introducing concurrent potentially distracting irrelevant items to a visual working memory paradigm, and manipulating executive load using simple or more demanding secondary verbal tasks. Across 7 experiments varying in presentation format, timing, stimulus set, and distractor number, we observed clear disruptive effects of executive load and visual distraction, but relatively minimal evidence supporting an interactive relationship between these factors. These findings are in line with recent evidence using delay-based interference, and suggest that different forms of attentional selection operate relatively independently in visual working memory. (PsycINFO Database Record

Temporal precision and the capacity of auditory-verbal short-term memory.

The capacity of serially ordered auditory-verbal short-term memory (AVSTM) is sensitive to the timing of the material to be stored, and both temporal processing and AVSTM capacity are implicated in the development of language. We developed a novel "rehearsal-probe" task to investigate the relationship between temporal precision and the capacity to remember serial order. Participants listened to a sub-span sequence of spoken digits and silently rehearsed the items and their timing during an unfilled retention interval. After an unpredictable delay, a tone prompted report of the item being rehearsed at that moment. An initial experiment showed cyclic distributions of item responses over time, with peaks preserving serial order and broad, overlapping tails. The spread of the response distributions increased with additional memory load and correlated negatively with participants' auditory digit spans. A second study replicated the negative correlation and demonstrated its specificity to AVSTM by controlling for differences in visuo-spatial STM and nonverbal IQ. The results are consistent with the idea that a common resource underpins both the temporal precision and capacity of AVSTM. The rehearsal-probe task may provide a valuable tool for investigating links between temporal processing and AVSTM capacity in the context of speech and language abilities.

Effects of rhythm on memory for spoken sequences: A model and tests of its stimulus-driven mechanism.

Immediate memory for spoken sequences depends on their rhythm - different levels of accuracy and patterns of error are seen according to the way in which items are spaced in time. Current models address these phenomena only partially or not at all. We investigate the idea that temporal grouping effects are an emergent property of a general serial ordering mechanism based on a population of oscillators locally-sensitive to amplitude modulations on different temporal scales. Two experiments show that the effects of temporal grouping are independent of the predictability of the grouping pattern, consistent with this model's stimulus-driven mechanism and inconsistent with alternative accounts in terms of top-down processes. The second experiment reports detailed and systematic differences in the recall of irregularly grouped sequences that are broadly consistent with predictions of the new model. We suggest that the bottom-up multi-scale population oscillator (or BUMP) mechanism is a useful starting point for a general account of serial order in language processing more widely.

Executive control of stimulus-driven and goal-directed attention in visual working memory.

We examined the role of executive control in stimulus-driven and goal-directed attention in visual working memory using probed recall of a series of objects, a task that allows study of the dynamics of storage through analysis of serial position data. Experiment 1 examined whether executive control underlies goal-directed prioritization of certain items within the sequence. Instructing participants to prioritize either the first or final item resulted in improved recall for these items, and an increase in concurrent task difficulty reduced or abolished these gains, consistent with their dependence on executive control. Experiment 2 examined whether executive control is also involved in the disruption caused by a post-series visual distractor (suffix). A demanding concurrent task disrupted memory for all items except the most recent, whereas a suffix disrupted only the most recent items. There was no interaction when concurrent load and suffix were combined, suggesting that deploying selective attention to ignore the distractor did not draw upon executive resources. A final experiment replicated the independent interfering effects of suffix and concurrent load while ruling out possible artifacts. We discuss the results in terms of a domain-general episodic buffer in which information is retained in a transient, limited capacity privileged state, influenced by both stimulus-driven and goal-directed processes. The privileged state contains the most recent environmental input together with goal-relevant representations being actively maintained using executive resources.