Removing irrelevant information from working memory: a cognitive aging study with the modified Sternberg task

Two experiments with a modified Sternberg recognition task explored the ability of young and old adults to remove irrelevant information from working memory. The task involved 2 memory sets, 1 of which was later cued as irrelevant. The recognition probe was presented at a variable time after the cue. Two indicators of inhibition, the setsize effect of the irrelevant set and the reaction time cost of intrusion probes (i.e., negative probes present in the irrelevant list), were dissociated. Irrelevant setsize effects lasted less than 1 s after the cue and did not differ between old and young adults. Intrusion costs lasted up to 5 s and were disproportionally large for old adults. With the additional requirement to remember both lists until after the probe, young adults' intrusion costs in Experiment 2 were equivalent to those of old adults in Experiment 1, but the setsize effects of the irrelevant set was larger. The results are compatible with a dual-process model of recognition in combination with a working-memory model distinguishing the focus of attention from the activated portion of long-term memory.     

Analogous mechanisms of selection and updating in declarative and procedural working memory: Experiments and a computational model

The article investigates the mechanisms of selecting and updating representations in declarative and procedural working memory (WM). Declarative WM holds the objects of thought available, whereas procedural WM holds representations of what to do with these objects. Both systems consist of three embedded components: activated long-term memory, a central capacity-limited component for building structures through temporary bindings, and a single-element focus of attention. Five experiments test the hypothesis of analogous mechanisms in declarative and procedural WM, investigating repetition effects across trials for individual representations (objects and responses) and for sets (memory sets and task sets), as well as set-congruency effects. Evidence for analogous processes was obtained from three phenomena: (1) Costs of task switching and of list switching are reduced with longer preparation interval. (2) The effects of task congruency and of list congruency are undiminished with longer preparation interval. (3) Response repetition interacts with task repetition in procedural WM; here we show an analogous interaction of list repetition with item repetition in declarative WM. All three patterns were reproduced by a connectionist model implementing the assumed selection and updating mechanisms. The model consists of two modules, an item-selection module selecting individual items from a memory set, or responses from a task set, and a set-selection module for selecting memory sets or task sets. The model codes the matrix of binding weights in the item-selection module as a pattern of activation in the set-selection module, thereby providing a mechanism for building chunks in LTM, and for unpacking them as structures into working memory.     

无关长时记忆表征能否引导视觉注意选择?

与任务相关的长时记忆表征在引导视觉注意选择的过程中扮演着重要的角色,它可以使人们在熟悉的视觉情境中快速搜索目标刺激,并偏离干扰刺激。但当长时记忆表征与任务无关时,还能否引导视觉注意选择?目前还不清楚。实验1采用眼动追踪技术直接比较无关工作记忆表征与无关长时记忆表征在视觉搜索阶段对视觉注意的捕获效应,行为反应时与首次注视点百分率的结果都发现,当无关工作记忆表征在视觉搜索中再次出现时能引导视觉注意偏向到与之匹配的干扰刺激,但无关长时记忆表征并没有表现出类似的注意引导效应;实验2探讨记忆表征由工作记忆系统转移到长时记忆系统的过程中对视觉注意的引导效应,结果发现,随着记忆表征的转移,注意引导效应消失了,实验3排除工作记忆表征的干扰后,依然没有发现无关长时记忆表征对注意的引导效应。以上结果表明,无关长时记忆表征并不能像工作记忆表征一样引导视觉注意选择,工作记忆表征和长时记忆表征对视觉注意的引导属于两个不同的认知过程。     

Selective interference with image retention and generation: Evidence for the workspace model

We address three types of model of the relationship between working memory (WM) and long-term memory (LTM): (a) the gateway model, in which WM acts as a gateway between perceptual input and LTM; (b) the unitary model, in which WM is seen as the currently activated areas of LTM; and (c) the workspace model, in which perceptual input activates LTM, and WM acts as a separate workspace for processing and temporary retention of these activated traces. Predictions of these models were tested, focusing on visuospatial working memory and using dual-task methodology to combine two main tasks (visual short-term retention and image generation) with two interference tasks (irrelevant pictures and spatial tapping). The pictures selectively disrupted performance on the generation task, whereas the tapping selectively interfered with the retention task. Results are consistent with the predictions of the workspace model.     

Binding and inhibition in working memory: individual and age differences in short-term recognition

Two studies investigated the relationship between working memory capacity (WMC), adult age, and the resolution of conflict between familiarity and recollection in short-term recognition tasks. Experiment 1 showed a specific deficit of young adults with low WMC in rejecting intrusion probes (i.e., highly familiar probes) in a modified Sternberg task, which was similar to the deficit found in old adults in a parallel experiment (K. Oberauer, 2001). Experiment 2 generalized these results to 3 recognition paradigms (modified Sternberg, local recognition, and n back tasks). Old adults showed disproportional performance deficits on intrusion probes only in terms of reaction times, whereas young adults with low WMC showed them only in terms of errors. The generality of the effect across paradigms is more compatible with a deficit in content-context bindings subserving recollection than with a deficit in inhibition of irrelevant information in working memory. Structural equation models showed that WMC is related to the efficiency of recollection but not of familiarity.     

Working memory retrieval differences between medial temporal lobe epilepsy patients and controls: A three memory layer approach

Multi-store models of working memory (WM) have given way to more dynamic approaches that conceive WM as an activated subset of long-term memory (LTM). The resulting framework considers that memory representations are governed by a hierarchy of accessibility. The activated part of LTM holds representations in a heightened state of activation, some of which can reach a state of immediate accessibility according to task demands. Recent neuroimaging studies have studied the neural basis of retrieval information with different states of accessibility. It was found that the medial temporal lobe (MTL) was involved in retrieving information within immediate access store and outside this privileged zone. In the current study we further explored the contribution of MTL to WM retrieval by analyzing the consequences of MTL damage to this process considering the state of accessibility of memory representations. The performance of a group of epilepsy patients with left hippocampal sclerosis in a 12-item recognition task was compared with that of a healthy control group. We adopted an embedded model of WM that distinguishes three components: the activated LTM, the region of direct access, and a single-item focus of attention. Groups did not differ when retrieving information from single-item focus, but patients were less accurate retrieving information outside focal attention, either items from LTM or items expected to be in the WM range. Analyses focused on items held in the direct access buffer showed that consequences of MTL damage were modulated by the level of accessibility of memory representations, producing a reduced capacity.     

Why do we move our eyes while trying to remember? The relationship between non-visual gaze patterns and memory

Non-visual gaze patterns (NVGPs) involve saccades and fixations that spontaneously occur in cognitive activities that are not ostensibly visual. While reasons for their appearance remain obscure, convergent empirical evidence suggests that NVGPs change according to processing requirements of tasks. We examined NVPGs in tasks with long-term memory (LTM) and working memory (WM) requirements. Experiment 1 yielded significantly higher eye movement rate (EMR) in tasks requiring LTM search than in a WM task requiring maintenance of information. Experiment 2 manipulated accessibility of items in study-test episodic tasks using the levels of processing paradigm. EMR was high in episodic recall irrespective of item accessibility. Experiment 3 examined functional significance of saccades in LTM tasks. Voluntary saccadic suppression produced no evidence that saccades contribute to task performance. We discuss the apparent epiphenomenal nature of spontaneous saccades from an evolutionary perspective and outline a neuroanatomical model of the link between the saccadic and memory system.     

The reaction-time task-rule congruency effect is not affected by working memory load: further support for the activated long-term memory hypothesis

Previous studies claimed that task representation is carried out by the activated long-term memory portion of working memory (WM; Meiran and Kessler in J Exp Psychol Human Percept Perform 34:137–157, 2008). The present study provides a more direct support for this hypothesis. We used the reaction-time task-rule congruency effect (RT-TRCE) in a task-switching setup, and tested the effects of loading WM with irrelevant task rules on RT-TRCE. Experiment 1 manipulated WM load in a between-subject design. WM participants performed a color/shape task switching, while having 0, 1 or 3 numerical task rules as WM load. Experiment 2 used a similar load manipulation (1 or 3 rules to load WM) in a within-subject design. Experiment 3 extended these results by loading WM with perceptual tasks that were more similar to the shape/color tasks. The results show that RT-TRCE was not affected by WM load supporting the activated long-term memory hypothesis.     

Working memory binding of visual object features in older adults

Accurate mental representation of visual stimuli requires retaining not only the individual features but also the correct relationship between them. This associative process of binding is mediated by working memory (WM) mechanisms. The present study re-examined reports of WM-related binding deficits with aging. In Experiment 1, 31 older and 31 younger adults completed a visual change detection task with feature–location relations presented either simultaneously or sequentially; the paradigm was also designed specifically to minimize the impact of lengthy retention intervals, elaborative rehearsal, and processing demands of multi-stimulus probes. In Experiment 2, 38 older and 42 younger adults completed a modified task containing both feature–location relations and feature–feature conjunctions. In Experiment 1 although feature–location binding was more difficult with sequential compared with simultaneous presentation, the effect was independent of age. In Experiment 2 while older adults were overall slower and less accurate than young adults, there were no age-specific deficits in WM binding. Overall, after controlling for methodological factors, there was no evidence of an age-related visual WM binding deficit for surface or location features. However, unlike younger adults, older adults appeared less able to restrict processing of irrelevant features, consistent with reported declines with age in strategic capacities of WM.     

A processing approach to the working memory/long-term memory distinction: evidence from the levels-of-processing span task

Recent theories suggest that performance on working memory (WM) tasks involves retrieval from long-term memory (LTM). To examine whether WM and LTM tests have common principles, Craik and Tulving's (1975) levels-of-processing paradigm, which is known to affect LTM, was administered as a WM task: Participants made uppercase, rhyme, or category-membership judgments about words, and immediate recall of the words was required after every 3 or 8 processing judgments. In Experiment 1, immediate recall did not demonstrate a levels-of-processing effect, but a subsequent LTM test (delayed recognition) of the same words did show a benefit of deeper processing. Experiment 2 showed that surprise immediate recall of 8-item lists did demonstrate a levels-of-processing effect, however. A processing account of the conditions in which levels-of-processing effects are and are not found in WM tasks was advanced, suggesting that the extent to which levels-of-processing effects are similar between WM and LTM tests largely depends on the amount of disruption to active maintenance processes.     

Modulation of working, short- and long-term memory by nicotinic receptors in the basolateral amygdala in rats

Male Wistar rats were exposed to one-trial step-down inhibitory avoidance training using a 0.5 mA footshock. Through bilaterally implanted indwelling cannulae, they received bilateral 0.5 microL infusions of saline, mecamylamine (1.0 or 10.0 microg/side), or nicotine (0.6 or 3.0 microg/side) into the basolateral complex of the amygdaloid nucleus (BLA). Infusions were either 10 min before training (Experiment 1) or 4 min after training (Experiment 2). In Experiment 1, the animals were tested three times: first for working memory (WM) 5 s after training, then for short-term memory (STM) 90 min later, and finally for long-term memory (LTM) 24 h later. Mecamylamine depressed and nicotine enhanced WM, STM, and LTM. In Experiment 2, the treatments were given after WM was presumably over. Again, mecamylamine inhibited and nicotine enhanced STM and LTM. The results indicate that nAChRs in BLA participate in the regulation of WM formation and STM and LTM acquisition and consolidation.     

What happens to information to be suppressed in working-memory tasks? Short and long term effects

The study explored, from an individual differences point of view, what happens to information to be suppressed in a working-memory task at short and long term. In particular, it was examined whether control mechanisms of irrelevant information in working memory imply their complete elimination from working memory or just the modulation of their activation. To this end, we compared the fate of irrelevant information in groups of subjects with high and low reading comprehension (Experiments 1 and 2) and subjects with high and low working memory (Experiments 1, 2, 3, and 4). All the experiments presented a working-memory task devised by De Beni, Palladino, Pazzaglia, and Cornoldi (1998), which required participants to process lists of words, to tap when a word from a particular category was presented, and then to recall only the last items in each list. Results confirmed that participants with high reading comprehension also have higher working memory and make less intrusion errors due to irrelevant items that have to be processed but then discarded. Furthermore, it was found that participants with low working memory have slightly better implicit (Experiment 1) and explicit memory (Experiments 3 and 4) of highly activated irrelevant information. Nevertheless, in a long-term recognition test, participants with high and low reading comprehension/working memory presented a similar pattern of memory for different types of irrelevant information (Experiment 2), whereas in a short-term memory recognition test, low-span participants presented a facilitation effect in the time required for the recognition of highly activated irrelevant information (Experiment 4). It was concluded that efficient working-memory performance is related to the temporary reduction of activation of irrelevant information but does not imply its elimination from memory.     

Aging and the intrusion superiority effect in visuo-spatial working memory

This study investigated the active component of visuo-spatial working memory (VSWM) in younger and older adults testing the hypotheses that elderly individuals have a poorer performance than younger ones and that errors in active VSWM tasks depend, at least partially, on difficulties in avoiding intrusions (i.e., avoiding already activated information). In two experiments, participants were presented with sequences of matrices on which three positions were pointed out sequentially: their task was to process all the positions but indicate only the final position of each sequence. Results showed a poorer performance in the elderly compared to the younger group and a higher number of intrusion (errors due to activated but irrelevant positions) rather than invention (errors consisting of pointing out a position never indicated by the experiementer) errors. The number of errors increased when a concurrent task was introduced (Experiment 1) and it was affected by different patterns of matrices (Experiment 2). In general, results show that elderly people have an impaired VSWM and produce a large number of errors due to inhibition failures. However, both the younger and the older adults' visuo-spatial working memory was affected by the presence of activated irrelevant information, the reduction of the available resources, and task constraints.     

Activation and binding in verbal working memory: a dual-process model for the recognition of nonwords

The article presents a mathematical model of short-term recognition based on dual-process models and the three-component theory of working memory [Oberauer, K. (2002). Access to information in working memory: Exploring the focus of attention. Journal of Experimental Psychology: Learning, Memory, and Cognition, 28, 411-421]. Familiarity arises from activated representations in long-term memory, ignoring their relations; recollection retrieves bindings in the capacity-limited component of working memory. In three experiments participants encoded two short lists of nonwords for immediate recognition, one of which was then cued as irrelevant. Probes from the irrelevant list were rejected more slowly than new probes; this was also found with probes recombining letters of irrelevant nonwords, suggesting that familiarity arises from individual letters independent of their relations. When asked to accept probes whose letters were all in the relevant list, regardless of their conjunction, participants accepted probes preserving the original conjunctions faster than recombinations, showing that recollection accessed feature bindings automatically. The model fit the data best when familiarity depended only on matching letters, whereas recollection used binding information.     

Dissociating retention and access in working memory: an age-comparative study of mental arithmetic

In two experiments, young and older adults solved arithmetic chain tasks with single-digit operands, with or without a concurrent memory load of three or six digits. Variables in the arithmetic tasks had to be replaced by digits from the screen or from the memory set. A task-irrelevant concurrent load impaired neither speed nor accuracy of arithmetic in younger adults. In Experiment 2, this was also true for older adults. A large decrease in arithmetic performance was observed, however, when variables in the arithmetic task had to be substituted by digits from the memory list. Older adults had specific problems with this condition in Experiment 1, where the substitution involved two successive steps, but not in Experiment 2, where the substitution from memory could be done in a single step. The results are difficult to reconcile with models assuming a common resource for storage and processing. Rather, they are compatible with the hypothesis that a concurrent memory load interferes with a processing task only during the points of access to working memory. Further, even though access to working memory was found to be the critical source of concurrent-load interference, it was found to be insensitive to the effects of adult aging.     

Creation of visual long-term memory

Currently, it is not known by which mechanisms novel visual representations are stored in long-term memory (LTM). Here we report evidence that visual working memory (WM) plays an important role in the formation of visual LTM. By varying exposure times and perceptual difficulty of the stimuli, we find that the rate-limiting factor constraining storage in LTM is the amount of information that can be simultaneously kept in WM, whereas the time needed to store this information into LTM is constant irrespective of the size of the WM content. These results support the hypothesis that visual WM serves as a gate for the storage of information into LTM.     

Neural substrates of successful working memory and long-term memory formation in a relational spatial memory task

Working memory (WM) tasks may involve brain activation actually implicated in long-term memory (LTM). In order to disentangle these two memory systems, we employed a combined WM/LTM task, using a spatial relational (object-location) memory paradigm and analyzed which brain areas were associated with successful performance for either task using fMRI. Critically, we corrected for the performance on the respective memory task when analyzing subsequent memory effects. The WM task consisted of a delayed-match-to-sample task assessed in an MRI scanner. Each trial consisted of an indoor or outdoor scene in which the exact configuration of four objects had to be remembered. After a short delay (7–13 s), the scene was presented from a different angle and spatial recognition for two objects was tested. After scanning, participants received an unexpected subsequent recognition memory (LTM) task, where the two previously unprobed objects were tested. Brain activity during encoding, delay phase and probe phase was analyzed based on WM and LTM performance. Results showed that successful WM performance, when corrected for LTM performance, was associated with greater activation in the inferior frontal gyrus and left fusiform gyrus during the early stage of the maintenance phase. A correct decision during the WM probe was accompanied by greater activation in a wide network, including bilateral hippocampus, right superior parietal gyrus and bilateral insula. No voxels exhibited supra-threshold activity during the encoding phase, and we did not find any differential activity for correct versus incorrect trials in the WM task when comparing LTM correct versus LTM incorrect trials.     

工作记忆和长时记忆共享信息表征的ERP证据

工作记忆和长时记忆共享记忆系统的信息表征网络, 关系密切。实验采用事件相关电位技术记录长时记忆信息对工作记忆的长时语义启动和工作记忆对相关长时记忆信息的长时语义启动。结果显示：与新项目相比, 学习项(即长时记忆信息)对工作记忆的目标信息产生了明显的行为启动效应, N2潜伏期、波幅和P3潜伏期也表现出明显的神经启动效应, 即, 学习项比新项目的反应时更短、N2和P3的潜伏期更短、N2更正, 但学习对非记忆项的影响不大; 工作记忆对相关的长时记忆信息也产生了神经启动, 表现在类别比较任务中记忆类信息比其他信息(即, 与工作记忆任务中的信息无关的内容)的N400更正。而且, 这种神经启动也会因刺激重复次数的增多而降低：与新项目相比, 学习项会使与之语义类似信息的N400更负。上述两种长时语义启动的存在及其ERPs的神经启动均为长时记忆和工作记忆使用共同信息表征的观点提供了证据支持。证据还表明, 在工作记忆和长时记忆的相互作用中注意是关键的调节因素。     

A secondary task is not always costly: Context‐based guidance of visual search survives interference from a demanding working memory task

Repeatedly encountering a visual search display with the target located at a fixed position relative to the distractors facilitates target detection, relative to novel displays – which is attributed to search guidance by (acquired) long‐term memory (LTM) of the distractor ‘context’ of the target. Previous research has shown that this ‘contextual cueing’ effect is severely impeded during learning when participants have to perform a demanding spatial working memory (WM) task concurrently with the search task, though it does become manifest when the WM task is removed. This has led to the proposal that search guidance by LT context memories critically depends on spatial WM to become ‘expressed’ in behaviour. On this background, this study, of two experiments, asked: (1) Would contextual cueing eventually emerge under dual‐task learning conditions if the practice on the task(s) is extended beyond the short training implemented in previous studies? and given sufficient practice, (2) Would performing the search under dual‐task conditions actually lead to an increased cueing effect compared to performing the visual search task alone? The answer is affirmative to both questions. In particular, Experiment 1 showed that a robust contextual cueing effect emerges within 360–720 dual‐task trials as compared to some 240 single‐task trials. Further, Experiment 2 showed that when dual‐ and single‐task conditions are performed in alternating trials blocks, the cueing effect for the very same set of repeated displays is significantly larger in dual‐task blocks than in single‐task blocks. This pattern of effects suggests that dual‐task practice eventually leads to direct, or ‘automatic’, guidance of visual search by learnt spatial LTM representations, bypassing WM processes. These processes are normally engaged in single‐task performance might actually interfere with direct LTM‐based search guidance.     

Task-set switching and long-term memory retrieval

The authors tested the hypothesis of a close relationship between the intentional component of task-set switching ("advance reconfiguration;" R. D. Rogers & S. Monsell, 1995) and long-term memory (LTM) retrieval. Consistent with this hypothesis, switch costs are reported to be larger when the switched-to task involves high retrieval demands (i.e., retrieval of episodic information) than when it involves low retrieval demands (i.e., retrieval of semantic information). In contrast, switch costs were not affected by a primary-task difficulty manipulation unrelated to intentional retrieval demands (Experiment 2). Also, the retrieval-demand effect on switch costs was eliminated when time for advanced preparation or task cues explicitly specifying the task rules were provided (Experiment 3). Overall, results were consistent with the hypothesis that the intentional switch-cost component reflects the time demands of retrieving appropriate task rules from LTM.