Assessing Working Memory Capacity Through Time-Constrained Elementary Activities

Working memory (WM) capacity measured through complex span tasks is among the best predictors of fluid intelligence (Gf). These tasks usually involve maintaining memoranda while performing complex cognitive activities that require a rather high level of education (e.g., reading comprehension, arithmetic), restricting their range of applicability. Because individual differences in such complex activities are nothing more than the concatenation of small differences in their elementary constituents, complex span tasks involving elementary processes should be as good of predictors of Gf as traditional tasks. The present study showed that two latent variables issued from either traditional or new span tasks involving time-constrained elementary activities were similarly correlated with Gf. Moreover, a model with a single unitary WM factor had a similar fit as a model with two distinct WM factors. Thus, time-constrained elementary activities can be integrated in WM tasks, permitting the assessment of WM in a wider range of populations.     

Exploring the forgetting mechanisms in working memory: Evidence from a reasoning span test

In working memory (WM) span tests participants have to maintain to-be-remembered information while processing other, potentially distracting, information. Previous studies have shown that WM span scores are greater when span lists start with a long processing task and end with a short processing task than when these processing tasks are presented in the reverse order (e.g., Towse, Hitch, & Hutton, 2000). In Experiment 1, we obtained a similar stimulus order effect in a reasoning span test, using reasoning sentences that were equated for length in terms of the number of constituent words, but which differed in processing complexity; span scores were greater when lists began with a complex sentence and ended with a simple sentence than when this stimulus order was reversed. In Experiment 2, the stimulus order effect was not found when processing duration was held constant while sentence complexity was varied using a computer-paced moving window presentation paradigm. These results suggest that duration-based constraints can affect degree of forgetting independently of the load generated during processing phases in WM span performance and therefore imply that time-related forgetting can occur in WM span tests, particularly when the difficulty of the processing component blocks active maintenance of to-be-remembered material.     

Relational processing and working memory capacity in comprehension of relative clause sentences

Previous research has indicated that the cognitive load imposed by tasks in various content domains increases with the complexity of the relational information processed. Sentence comprehension entails processing noun-verb relations to determine who did what to whom. The difficulty of object-extracted relative clause sentences might stem from the complex noun-verb relations they entail. Across three studies, participants read 16 types of object- and subject-extracted relative clause sentences at their own pace and then responded to a comprehension question for each sentence. Relational processing was assessed using a premise integration task or a Latin square task. These tasks predicted comprehension of object-relatives before and after controlling for subject-relatives. Working memory (WM) capacity was assessed using reading span or forward and backward digit span tests. WM tasks predicted comprehension of object-relatives before but not after controlling for subject-relatives. Comprehension of object-relatives relied more heavily on a domain-general capacity to process complex relations than on WM capacity.     

Individual Differences in Working Memory Capacity: More Evidence for a General Capacity Theory

The causes of the positive relationship between comprehension and measures of working memory capacity remain unclear. This study tests three hypotheses for the relationship by equating the difficulty, for 48 individual subjects, of processing demands in complex working memory tasks. Even with difficulty of processing equated, the relationship between number of words recalled in the working memory measure and comprehension remained high and significant. The results favour a general capacity view. We suggest that high working memory span subjects have more limited-capacity attentional resources available to them than low span subjects and that individual differences in working memory capacity will have implications for any task that requires controlled effortful processing.     

The real relationship between short-term memory and working memory

Storage-oriented memory span tasks with no explicit concurrent processing are usually referred as short-term memory (STM) tasks, whereas tasks involving storage plus concurrent processing requirements are designated as working memory (WM) tasks. The present study explores a question that remains unsolved: Do STM and WM tasks clearly tap distinguishable theoretical constructs? For that purpose, a large sample of 403 participants was tested through 12 diverse memory span tasks. Half of those tasks are widely accepted as measures of STM, whereas the other half measure WM. The results show that STM and WM share largely overlapping underlying capacity limitations, suggesting that all memory span tasks tap essentially the same construct. Some implications are discussed.     

The real relationship between short-term memory and working memory

Storage-oriented memory span tasks with no explicit concurrent processing are usually referred as short-term memory (STM) tasks, whereas tasks involving storage plus concurrent processing requirements are designated as working memory (WM) tasks. The present study explores a question that remains unsolved: Do STM and WM tasks clearly tap distinguishable theoretical constructs? For that purpose, a large sample of 403 participants was tested through 12 diverse memory span tasks. Half of those tasks are widely accepted as measures of STM, whereas the other half measure WM. The results show that STM and WM share largely overlapping underlying capacity limitations, suggesting that all memory span tasks tap essentially the same construct. Some implications are discussed.     

The influence of levels of processing on recall from working memory and delayed recall tasks

Recent research in working memory has highlighted the similarities involved in retrieval from complex span tasks and episodic memory tasks, suggesting that these tasks are influenced by similar memory processes. In the present article, the authors manipulated the level of processing engaged when studying to-be-remembered words during a reading span task (Experiment 1) and an operation span task (Experiment 2) in order to assess the role of retrieval from secondary memory during complex span tasks. Immediate recall from both span tasks was greater for items studied under deep processing instructions compared with items studied under shallow processing instructions regardless of trial length. Recall was better for deep than for shallow levels of processing on delayed recall tests as well. These data are consistent with the primary-secondary memory framework, which suggests that to-be-remembered items are displaced from primary memory (i.e., the focus of attention) during the processing phases of complex span tasks and therefore must be retrieved from secondary memory.     

The scope and control of attention as separate aspects of working memory

The present study examines two varieties of working memory (WM) capacity task: visual arrays (i.e., a measure of the amount of information that can be maintained in working memory) and complex span (i.e., a task that taps WM-related attentional control). Using previously collected data sets we employ confirmatory factor analysis to demonstrate that visual arrays and complex span tasks load on separate, but correlated, factors. A subsequent series of structural equation models and regression analyses demonstrate that these factors contribute both common and unique variance to the prediction of general fluid intelligence (Gf). However, while visual arrays does contribute uniquely to higher cognition, its overall correlation to Gf is largely mediated by variance associated with the complex span factor. Thus we argue that visual arrays performance is not strictly driven by a limited-capacity storage system (e.g., the focus of attention; Cowan, 2001), but may also rely on control processes such as selective attention and controlled memory search.     

Bilingual working memory span is affected by language skill

A complex span procedure was used to study whether there is a measurable extra load on general working memory (WM) when a not fully automatised language has to be comprehended. Participants verified heard sentences in relation to simultaneously shown pictures and memorised the last word of each sentence. The sentences were presented in growing set sizes and recall of all the last words was required after each set. The largest number of sentences that could be processed in combination with successful recall of their last words determined the participant's WM span. Sentences were either in the participants' native language or a well-mastered second language, English. Participants were psychology and English students. WM span and decision accuracy were better for native-language than foreign-language sentences, especially for the psychology students. A second experiment ascertained that the difference between participant groups could not be explained by variables related to phonological shortterm memory or word processing. A third experiment controlled for systematic trade-off differences between languages and groups. We conclude that aspects of sentence comprehension in foreign language develop with practice so that they require less WM resources.     

Working memory and language comprehension: A meta-analysis

This paper presents a meta-analysis of the data from 6,179 participants in 77 studies that investigated the association between working-memory capacity and language comprehension ability. A primary goal of the meta-analysis was to compare the predictive power of the measures of working memory developed by Daneman and Carpenter (1980) with the predictive power of other measures of working memory. The results of the meta-analysis support Daneman and Carpenter’s (1980) claim that measures that tap the combined processing and storage capacity of working memory (e.g., reading span, listening span) are better predictors of comprehension than are measures that tap only the storage capacity (e.g., word span, digit span). The meta-analysis also showed that math process plus storage measures of working memory are good predictors of comprehension. Thus, the superior predictive power of the process plus storage measures is not limited to measures that involve the manipulation of words and sentences.     

The role of working memory in inferential sentence comprehension

Existing literature on inference making is large and varied. Trabasso and Magliano (Discourse Process 21(3):255–287, 1996) proposed the existence of three types of inferences: explicative, associative and predictive. In addition, the authors suggested that these inferences were related to working memory (WM). In the present experiment, we investigated whether WM capacity plays a role in our ability to answer comprehension sentences that require text information based on these types of inferences. Participants with high and low WM span read two narratives with four paragraphs each. After each paragraph was read, they were presented with four true/false comprehension sentences. One required verbatim information and the other three implied explicative, associative and predictive inferential information. Results demonstrated that only the explicative and predictive comprehension sentences required WM: participants with high verbal WM were more accurate in giving explanations and also faster at making predictions relative to participants with low verbal WM span; in contrast, no WM differences were found in the associative comprehension sentences. These results are interpreted in terms of the causal nature underlying these types of inferences.     

Individual differences in working memory: introduction to the special section.

This special section includes a set of 5 articles that examine the nature of inter- and intraindividual differences in working memory, using working memory span tasks as the main research tools. These span tasks are different from traditional short-term memory spans (e.g., digit or word span) in that they require participants to maintain some target memory items (e.g., words) while simultaneously performing some other tasks (e.g., reading sentences). In this introduction, a brief discussion of these working memory span tasks and their characteristics is provided first. This is followed by an overview of 2 major theoretical issues that are addressed by the subsequent articles--(a) the factors influencing the inter- and intraindividual differences in working memory performance and (b) the domain generality versus domain specificity of working memory--and also of some important issues that must be kept in mind when readers try to evaluate the claims regarding these 2 theoretical issues.     

Working memory demands modulate cognitive control in the Stroop paradigm

One important task of cognitive control is to regulate behavior by resolving information processing conflicts. In the Stroop task, e.g., incongruent trials lead to conflict-related enhancements of cognitive control and to improved behavioral performance in subsequent trials. Several studies suggested that these cognitive control processes are functionally and anatomically related to working memory (WM) functions. The present study investigated this suggestion and tested whether these control processes are modulated by concurrent WM demands. For this purpose, we performed three experiments in which we combined different WM tasks with the Stroop paradigm and measured their effects on cognitive control. We found that high WM demands led to a suppression of conflict-triggered cognitive control, whereas our findings suggest that this suppression effect is rather due to WM updating than to maintenance demands. We explain our findings by assuming that WM processes interfere with conflict-triggered cognitive control, harming the efficiency of these control processes.     

Stroke damages attentional maintenance in working memory

Stroke is the main cause of acquired disability in adults, and specific deficits in working memory (WM) are among the most common cognitive consequences. In neuropsychological routine, WM is most of the time investigated in the framework of the multicomponent model (Baddeley & Hitch, 1974, The psychology of learning and motivation, 47). Using a more recent theoretical WM model, the time-based resource-sharing (TBRS) model (Barrouillet et al., 2011, Psychol. Rev., 118, 175), the aim of the present study was to investigate in young post-stroke patients to which extent attentional maintenance is impaired in WM. To address this question, we discarded other factors known to directly influence WM performance, that is processing speed and short-term memory span. We proposed to 53 post-stroke patients and to 63 healthy controls a complex span paradigm in which participants were asked to alternate between the memorization of a series of images and a concurrent parity judgement task of a series of digits. To investigate the attentional maintenance processes, we manipulated the cognitive load (CL) of the concurrent task. CL effect is typically interpreted as the involvement of attentional maintenance processes. The task was adapted to each participant according to their processing speed and memory span. As expected, the results showed higher recall performance in healthy controls compared with post-stroke patients. Consistent with the literature, we also observed higher performance when the CL was low compared with high. However, the improvement in recall at low CL was smaller for post-stroke patients compared with controls, suggesting that post-stroke WM deficit could be in part due to a deficit of the attentional maintenance processes.     

The role of working memory in the metaphor interference effect

Participants took longer to judge that metaphors (e.g., an insult is a razor, memory is a warehouse) were literally false than to judge that scrambled sentences (e.g., an insult is a warehouse) were false. This result is the metaphor interference effect (MIE). It demonstrates that metaphor processing is automatic. In this experiment, we found that the magnitude of the MIE is predicted by working memory (WM) capacity, with higher WM yielding a smaller MIE. This suggests that although metaphor comprehension is automatic, the early processing of metaphors is controllable by executive mechanisms. We relate our results to Kintsch’s (2000, 2001) predication model. Specifically, we suggest that mechanisms of WM influence metaphor processing by affecting the effectiveness of the construction-integration process that identifies common properties between topics and vehicles. WM also influences the speed with which meanings are identified as literal or figurative.     

Working memory load and distraction: dissociable effects of visual maintenance and cognitive control

We establish a new dissociation between the roles of working memory (WM) cognitive control and visual maintenance in selective attention as measured by the efficiency of distractor rejection. The extent to which focused selective attention can prevent distraction has been shown to critically depend on the level and type of load involved in the task. High perceptual load that consumes perceptual capacity leads to reduced distractor processing, whereas high WM load that reduces WM ability to exert priority-based executive cognitive control over the task results in increased distractor processing (e.g., Lavie, Trends in Cognitive Sciences, 9(2), 75–82, 2005). WM also serves to maintain task-relevant visual representations, and such visual maintenance is known to recruit the same sensory cortices as those involved in perception (e.g., Pasternak & Greenlee, Nature Reviews Neuroscience, 6(2), 97–107, 2005). These findings led us to hypothesize that loading WM with visual maintenance would reduce visual capacity involved in perception, thus resulting in reduced distractor processing—similar to perceptual load and opposite to WM cognitive control load. Distractor processing was assessed in a response competition task, presented during the memory interval (or during encoding; Experiment 1a) of a WM task. Loading visual maintenance or encoding by increased set size for a memory sample of shapes, colors, and locations led to reduced distractor response competition effects. In contrast, loading WM cognitive control with verbal rehearsal of a random letter set led to increased distractor effects. These findings confirm load theory predictions and provide a novel functional distinction between the roles of WM maintenance and cognitive control in selective attention.     

Does Controlling for Temporal Parameters Change the Levels-of-Processing Effect in Working Memory?

The distinguishability between working memory (WM) and long-term memory has been a frequent and long-lasting source of debate in the literature. One recent method of identifying the relationship between the two systems has been to consider the influence of long-term memory effects, such as the levels-of-processing (LoP) effect, in WM. However, the few studies that have examined the LoP effect in WM have shown divergent results. This study examined the LoP effect in WM by considering a theoretically meaningful methodological aspect of the LoP span task. Specifically, we fixed the presentation duration of the processing component a priori because such fixed complex span tasks have shown differences when compared to unfixed tasks in terms of recall from WM as well as the latent structure of WM. After establishing a fixed presentation rate from a pilot study, the LoP span task presented memoranda in red or blue font that were immediately followed by two processing words that matched the memoranda in terms of font color or semantic relatedness. On presentation of the processing words, participants made deep or shallow processing decisions for each of the memoranda before a cue to recall them from WM. Participants also completed delayed recall of the memoranda. Results indicated that LoP affected delayed recall, but not immediate recall from WM. These results suggest that fixing temporal parameters of the LoP span task does not moderate the null LoP effect in WM, and further indicate that WM and long-term episodic memory are dissociable on the basis of LoP effects.     

Individual differences in working memory strategies for reading expository text

This study investigated whether individual differences in working memory (WM) span are associated with different WM management strategies during the reading of expository text. In Experiment 1, probe questions were presented on line during reading to determine whether thematic information was maintained in WM throughout comprehension. The data indicated that readers across the range of WM span maintained thematic information in WM throughout the reading of a given passage. In Experiment 2, sentence reading times and accuracy for both topic and detail questions were measured in two conditions: when topic sentences were present and when topic sentences were absent. Subjects performed similarly across the range of WM span in the topic-present condition, but lower span subjects performed more poorly on detail questions in the topic-absent condition. In Experiment 3, the topic-present condition of the second experiment was replicated, except that subjects expected to receive questions about details only. Thematic processing and retention of topic and detail information all increased with span. Taken together, these results suggest that, for more difficult text processing tasks, high- and low-span subjects adopt different WM management strategies and these strategies influence what is learned from reading the text.     

Why does working memory span predict complex cognition? Testing the strategy affordance hypothesis

We introduce and empirically evaluate the strategy affordance hypothesis, which holds that individual differences in strategy use will mediate the relationship between performances on a working memory (WM) span task and another cognitive task only when the same strategies are afforded by both tasks. One hundred forty-eight participants completed basic memory tasks and verbal span tasks that afford the same strategies, such as imagery and sentence generation, and completed reading comprehension tasks that afford different ones, such as self-questioning and summarization. Effective strategy use on WM span tasks accounted for variance in the span-memory relationship, but not for the span-comprehension relationship, supporting the strategy affordance hypothesis. Strategy use mediated the span-cognition relationship only when both tasks afforded the same strategies.     

Temporal–contextual processing in working memory: Evidence from delayed cued recall and delayed free recall tests

Three experiments are reported that addressed the nature of processing in working memory by investigating patterns of delayed cued recall and free recall of items initially studied during complex and simple span tasks. In Experiment 1, items initially studied during a complex span task (i.e., operation span) were more likely to be recalled after a delay in response to temporal–contextual cues, relative to items from subspan and supraspan list lengths in a simple span task (i.e., word span). In Experiment 2, items initially studied during operation span were more likely to be recalled from neighboring serial positions during delayed free recall than were items studied during word span trials. Experiment 3 demonstrated that the number of attentional refreshing opportunities strongly predicts episodic memory performance, regardless of whether the information is presented in a spaced or massed format in a modified operation span task. The results indicate that the content–context bindings created during complex span trials reflect attentional refreshing opportunities that are used to maintain items in working memory.