The role of biased competition in visual short-term memory

Despite being extensively studied, many important questions regarding visual short-term memory (VSTM) are yet to be fully answered. The present review seeks to explore the extent to which competitive interactions present during perception of stimuli may also operate during stimulus encoding and/or maintenance in VSTM. Additionally, the paper provides an overview of the methods and approaches that have been used to study the properties of VSTM, and a review of research into the limits of VSTM capacity. We take as our starting point the biased competition model of attention (Desimone & Duncan, 1995) and discuss the influence that such competition may have on the limited capacity of VSTM by a review of the literature and by recent experiments from Shapiro's lab. The present report outlines several experiments that provide results consistent with the idea that low-level competitive interactions may influence VSTM, with the aim of stimulating further research into this new area.     

Orienting attention to objects in visual short-term memory

We measured electroencephalographic activity during visual search of a target object among objects available to perception or among objects held in visual short-term memory (VSTM). For perceptual search, a single shape was shown first (pre-cue) followed by a search-array, and the task was to decide whether the pre-cue was or was not in the search-array. For search of VSTM, a search-array was shown first followed by a single shape (post-cue), and the task was to decide whether the post-cue was or was not in the previously displayed search-array. We focused on early lateralized electrical brain activity over posterior and temporal areas time-locked to search-arrays in pre-cue trials and to post-cues in post-cue trials. In Experiment 1, search-arrays were composed of two lateralized shapes, displayed in the upper/lower two quadrants of the monitor. In Experiment 2, search-arrays were composed of four shapes, displayed at the corners of an imaginary square centered on fixation. In pre-cue trials, we observed an N2pc of about equal amplitude and latency for search-arrays composed of two or four shapes. In post-cue trials, we observed N2pc-like activity with search-arrays composed of two shapes, that was however substantially attenuated with search-arrays composed of four shapes. For many aspects, attending to a perceptual object was functionally and neurally analogous to attending to an object held in VSTM, suggesting that spatial selective attention biases search of objects during both ongoing perception and retention.     

Functional connectivity during top-down modulation of visual short-term memory representations

Recent evidence has revealed that short-lived internal representations held in visual short-term memory (VSTM) can be modulated by top-down control via retrospective attention which impacts subsequent behavioral performance. However, the functional inter-regional interactions underlying these top-down modulatory effects are not fully characterized. Here we used event-related functional magnetic imaging to investigate whether the strength of functional connectivity between the frontal cortex and posterior visual areas varies with the efficacy of top-down modulation of memory traces. Top-down modulation was manipulated by the timing of retro-cuing (early or late) in a VSTM task. Univariate analyses revealed that more effective top-down modulation (early cueing vs. late cueing) increased activity in early visual areas. Importantly, coherency analyses revealed that top-down modulation produced stronger functional connectivity between frontal and posterior occipital regions. Also, participants with stronger functional connectivity exhibit better memory performance. These results suggest that augmented functional connectivity between frontal and posterior visual areas strengthens the VSTM representations of importance to behavioral goals.     

Distinct electrophysiological indices of maintenance in auditory and visual short-term memory

We compared the electrophysiological correlates for the maintenance of non-musical tones sequences in auditory short-term memory (ASTM) to those for the short-term maintenance of sequences of coloured disks held in visual short-term memory (VSTM). The visual stimuli yielded a sustained posterior contralateral negativity (SPCN), suggesting that the maintenance of sequences of coloured stimuli engaged structures similar to those involved in the maintenance of simultaneous visual displays. On the other hand, maintenance of acoustic sequences produced a sustained negativity at fronto-central sites. This component is named the Sustained Anterior Negativity (SAN). The amplitude of the SAN increased with increasing load in ASTM and predicted individual differences in the performance. There was no SAN in a control condition with the same auditory stimuli but no memory task, nor one associated with visual memory. These results suggest that the SAN is an index of brain activity related to the maintenance of representations in ASTM that is distinct from the maintenance of representations in VSTM.     

Manipulations of attention dissociate fragile visual short-term memory from visual working memory

People often rely on information that is no longer in view, but maintained in visual short-term memory (VSTM). Traditionally, VSTM is thought to operate on either a short time-scale with high capacity - iconic memory - or a long time scale with small capacity - visual working memory. Recent research suggests that in addition, an intermediate stage of memory in between iconic memory and visual working memory exists. This intermediate stage has a large capacity and a lifetime of several seconds, but is easily overwritten by new stimulation. We therefore termed it fragile VSTM. In previous studies, fragile VSTM has been dissociated from iconic memory by the characteristics of the memory trace. In the present study, we dissociated fragile VSTM from visual working memory by showing a differentiation in their dependency on attention. A decrease in attention during presentation of the stimulus array greatly reduced the capacity of visual working memory, while this had only a small effect on the capacity of fragile VSTM. We conclude that fragile VSTM is a separate memory store from visual working memory. Thus, a tripartite division of VSTM appears to be in place, comprising iconic memory, fragile VSTM and visual working memory.     

Reduced prefrontal oxygenation during object and spatial visual working memory in unpolar and bipolar depression

Altered prefrontal brain activity (e.g. hypofrontality) during cognitive tasks such as working memory is a core neuroimaging marker in unipolar (UNI) and bipolar (BI) depression. The present study investigated for the first time UNI (n = 16) and BI patients (n = 14) in a working memory task including different processes (storage and matching) and components (object and spatial visual) with functional near-infrared spectroscopy (fNIRS) over the prefrontal cortex. In healthy controls (n = 15) comparable to both patient groups, changes of oxygenated and deoxygenated haemoglobin indicated increased ventro-lateral, dorso-lateral prefrontal and superior frontal cortex activity for object and spatial visual working memory storage as compared to the control condition. In contrast, both patient groups showed diminished brain activity in all working memory conditions. Results revealed unspecific deficits that did not allow the differentiation between unipolar and bipolar depression in dependence of working memory processes or components. However, fNIRS can be considered as a valid, easy manageable, low cost and rapid tool for measuring (diminished) prefrontal cortex functions.     

Verbal, visual, and spatio-sequential short-term memory: assessment of the storage capacities of children and teenagers with Down's syndrome

Background   It is recognized that individuals with Down's syndrome have a specific deficit in verbal short-term memory. On the other hand, non-verbal short-term memory seems to be preserved or even be a strong point for these persons. Nevertheless, the extent and specificity of the deficit must be determined. To do so, we carried out a research programme that allowed us to simultaneously assess various short-term memory systems in a developmental perspective, and to compare our participants' performance to that obtained by typically developing individuals of the same mental age.
Method   Three span tasks are used (auditory word span/visual patterns test/Corsi blocks task) with 54 children and teenagers with Down's syndrome and 54 typically developing children as control group. Participants were matched according to their cognitive level.
Results   For the auditory word span task, participants with Down's syndrome obtained performances significantly lower than those of the typically developing participants. On the other hand, compared with typically developing children, children and teenagers with Down's syndrome have a spatio-sequential span significantly higher for the lowest developmental ages. No significant differences were found for visual span.
Conclusions   Individuals with Down's syndrome exhibited a distinctive pattern of memory performance, in addition to their developmental specificities.     

Deficits in visual short-term memory binding in children at risk of non-verbal learning disabilities

It has been hypothesized that learning disabled children meet short-term memory (STM) problems especially when they must bind different types of information, however the hypothesis has not been systematically tested. This study assessed visual STM for shapes and colors and the binding of shapes and colors, comparing a group of children (aged between 8 and 10 years) at risk of non-verbal learning disabilities (NLD) with a control group of children matched for general verbal abilities, age, gender, and socioeconomic level. Results revealed that groups did not differ in retention of either shapes or colors, but children at risk of NLD were poorer than controls in memory for shape-color bindings.     

Contrasting early visual cortical activation states causally involved in visual imagery and short-term memory

Whether visual imagery and visual short-term memory (STM) share the same neural resources, and the extent to which the early visual cortex (V1/V2) is involved in these processes, has been the subject of much debate. Here, we used transcranial magnetic stimulation (TMS) in two separate experiments to contrast the neural states associated with visual imagery and visual STM in the early visual cortex. In Experiment 1, we investigated V1/V2 activation states at the end of the retention phase in a visual imagery and a visual STM task. V1/V2 TMS facilitated performance in both tasks; the finding that imagery and STM interacted with TMS in the same way suggests that the two processes have similar effects on early visual cortical excitability. In Experiment 2, we investigated V1/V2 activation states at the beginning of the retention phase. V1/V2 TMS impaired performance in the visual STM task, whereas it had no effect on the imagery task. Taken together, our findings show that the late phases of the early visual cortical activation state associated with visual imagery and visual STM are similar; differences between the two processes are apparent in the early phases of the tasks. Our results also suggest that the causal role of the early visual cortex in visual STM includes both the initial translation of the visual input into working memory and the subsequent maintenance of the mental representation. Finally, our findings indicate that visual STM sensory recruitment in working memory might act via excitability modulation of V1/V2 neurons.     

Markers of preparatory attention predict visual short-term memory performance

Visual short-term memory (VSTM) is limited in capacity. Therefore, it is important to encode only visual information that is most likely to be relevant to behaviour. Here we asked which aspects of selective biasing of VSTM encoding predict subsequent memory-based performance. We measured EEG during a selective VSTM encoding task, in which we varied parametrically the memory load and the precision of recall required to compare a remembered item to a subsequent probe item. On half the trials, a spatial cue indicated that participants only needed to encode items from one hemifield. We observed a typical sequence of markers of anticipatory spatial attention: early attention directing negativity (EDAN), anterior attention directing negativity (ADAN), late directing attention positivity (LDAP); as well as of VSTM maintenance: contralateral delay activity (CDA). We found that individual differences in preparatory brain activity (EDAN/ADAN) predicted cue-related changes in recall accuracy, indexed by memory-probe discrimination sensitivity (d′). Importantly, our parametric manipulation of memory-probe similarity also allowed us to model the behavioural data for each participant, providing estimates for the quality of the memory representation and the probability that an item could be retrieved. We found that selective encoding primarily increased the probability of accurate memory recall; that ERP markers of preparatory attention predicted the cue-related changes in recall probability.     

Patient assessment based on a theory of visual attention (TVA): subtle deficits after a right frontal-subcortical lesion

We report on a patient who complained of reduced awareness in the left visual field, but showed no visual neglect or extinction in clinical testing. By MR scanning, the brain damage was localized to the right basal ganglia, also involving structures in right frontal cortex. Using psychophysical testing and mathematical modeling based on Bundesen's theory of visual attention [TVA; Psychol. Rev. 97 (1990) 523], the patient's subjective experience of attentional disturbance was confirmed, and the deficit was specified into several components. At very short exposure durations, two effects were shown. The detection threshold was elevated, particularly in the left visual field, and stimuli in this side were given less attentional weight. In addition, the capacity of visual short-term memory (VSTM) was markedly reduced in both visual fields. The robustness of the test results was evaluated by bootstrap analysis. The study demonstrates the sensitivity and specificity gained by combining psychophysical testing with TVA modeling in the analysis of visual attention disorders. Extending the results of a pioneer study of parietal neglect patients by Duncan et al. [J. Exp. Psychol.: Gen. 128 (1999) 450], this study demonstrates the strength of the method in a single case, with a lesion outside parietal cortex, and only minor clinical symptoms.     

Task-irrelevant memory load induces inattentional blindness without temporo-parietal suppression

We often fail to consciously detect an unexpected object when we are engaged in an attention-demanding task (inattentional blindness). The inattentional blindness which is induced by visual short-term memory (VSTM) load has been proposed to result from a suppression of temporo-parietal junction (TPJ) activity that involves stimulus-driven attention. However, the fact that, inversely proportional to TPJ activity, intraparietal sulcus (IPS) activity correlates with VSTM load renders questionable the account of inattentional blindness based only on TPJ activity. Here, we investigated whether the TPJ is solely responsible for inattentional blindness by decoupling IPS and TPJ responses to VSTM load and then using the same manipulation to test the behavioral inattentional blindness performance. Experiment 1 showed that TPJ activity was not suppressed by task-irrelevant load while the IPS responded to both task-relevant and task-irrelevant load. Although the TPJ account of inattentional blindness predicts that the degree of inattentional blindness should track TPJ activity, we found in Experiment 2 that inattentional blindness was induced not only by task-relevant load but also by task-irrelevant load, showing inconsistency between the extent of inattentional blindness and TPJ response. These findings suggest that inattentional blindness can be induced without suppression of TPJ activity and seem to offer the possibility that the IPS contributes to conscious perception.     

Phonological recoding, visual short-term store and the effect of unattended speech: evidence from a case of slowly progressive anarthria

Phonological processing and short-term memory were investigated in a patient with slowly progressive anarthria. The patient, who had an auditory-verbal span in the lower unimpaired range, showed preserved phonological similarity and word length effects with auditory presentation. These phonological effects of immediate retention were absent with visual input. The patient was also unable to perform phonological judgments on written verbal material. Under unattended speech interference her visual span improved, with an increase of visual confusion errors. In the light of a model including a main auditory-verbal storage component (the phonological short-term input store, STS), and a rehearsal process, that, after phonological recoding, conveys visually presented verbal material to the phonological STS, the patient's pattern of impairment is interpreted as a selective deficit in the process of phonological recoding. This prevents visual-verbal material from accessing the phonological STS, that is, therefore, insulated from visual inputs.     

Understanding the allocation of attention when faced with varying perceptual load in partial report: a computational approach

The allocation of visual processing capacity is a key topic in studies and theories of visual attention. The load theory of Lavie (1995) proposes that allocation happens in two steps where processing resources are first allocated to task-relevant stimuli and secondly remaining capacity ‘spills over’ to task-irrelevant distractors. In contrast, the Theory of Visual Attention (TVA) proposed by Bundesen (1990) assumes that allocation happens in a single step where processing capacity is allocated to all stimuli, both task-relevant and task-irrelevant, in proportion to their relative attentional weight. Here we present data from two partial report experiments where we varied the number and discriminability of the task-irrelevant stimuli (Experiment 1) and perceptual load (Experiment 2). The TVA fitted the data of the two experiments well thus favoring the simple explanation with a single step of capacity allocation. We also show that the effects of varying perceptual load can only be explained by a combined effect of allocation of processing capacity as well as limits in visual working memory. Finally, we link the results to processing capacity understood at the neural level based on the neural theory of visual attention by Bundesen et al. (2005).     

Maternal speech shapes the cerebral frontotemporal network in neonates: A hemodynamic functional connectivity study

Language development and the capacity for communication in infants are predominantly supported by their mothers, beginning when infants are still in utero. Although a mother’s speech should thus have a significant impact on her neonate’s brain, neurocognitive evidence for this hypothesis remains elusive. The present study examined 37 neonates using near-infrared spectroscopy and observed the interactions between multiple cortical regions while neonates heard speech spoken by their mothers or by strangers. We analyzed the functional connectivity between regions whose response-activation patterns differed between the two types of speakers. We found that when hearing their mothers’ speech, functional connectivity was enhanced in both the neonatal left and right frontotemporal networks. On the left it was enhanced between the inferior/middle frontal gyrus and the temporal cortex, while on the right it was enhanced between the frontal pole and temporal cortex. In particular, the frontal pole was more strongly connected to the left supramarginal area when hearing speech from mothers. These enhanced frontotemporal networks connect areas that are associated with language (left) and voice processing (right) at later stages of development. We suggest that these roles are initially fostered by maternal speech.     

Losing memories overnight: A unique form of human amnesia

Since an automobile accident in 2005, patient FL has reported difficulty retaining information from one day to the next. During the course of any given day, she describes her memory as normal. However, memory for each day disappears during a night of sleep. She reports good memory for events that occurred before the accident. Although this pattern of memory impairment is, to our knowledge, unique to the medical literature, it was depicted in the fictional film “50 First Dates”. On formal testing, FL performed moderately well when trying to remember material that she had learned during the same day, but she exhibited no memory at all for material that she knew had been presented on a previous day. For some tests, unbeknownst to FL, material learned on the previous day was intermixed with material learned on the same day as the test. On these occasions, FL's memory was good. Thus, she was able to remember events from earlier days when memory was tested covertly. FL performed differently in a number of ways from individuals who were instructed to consciously feign her pattern of memory impairment. It was also the impression of those who worked with FL that she believed she had the memory impairment that she described and that she was not intentionally feigning amnesia. On the basis of her neuropsychological findings, together with a normal neurological exam, normal MRI findings, and psychiatric evaluation, we suggest that FL exhibits a unique form of functional amnesia and that its characterization may have been influenced by knowledge of how amnesia was depicted in a popular film. She subsequently improved (and began retaining day-to-day memory) at Johns Hopkins University where she was in a supportive in-patient environment and was shown how to take control of her condition by interrupting her sleep at 4-h intervals.     

A multiple case study of verbal short-term memory in velo-cardio-facial syndrome

Background Velo‐cardio‐facial syndrome (VCFS, 22q11.2 deletion) is characterized by severely delayed language development. The current study explored the integrity of verbal short‐term memory (STM), a cognitive function critically involved in language development, in eight children with VCFS. Methods Using a multiple case study design, we presented a series of STM tasks exploring immediate serial recall for word and non‐word lists to eight children with VCFS (aged 8–12 years) and to chronological‐age‐matched control groups. A first task assessed the integrity of phonological coding in verbal STM by comparing recall for phonologically similar and dissimilar words. Subsequently, the interaction between verbal knowledge and STM capacity was investigated by comparing recall for high‐ and low‐imageability words, for high‐ and low‐frequency words, and for words and non‐words. A final task assessed short‐term serial order recognition for digit sequences. Results When computing the number of items recalled in the word recall tasks, independently of their serial position, only one child presented consistent difficulties. Short‐term recall of non‐words was normal in each child. Phonological similarity and verbal knowledge influenced STM performance to a similar extent in children with VCFS and controls. On the other hand, when applying a strict serial recall criterion, difficulties with the word and non‐word recall tasks were observed in most children. Half of the patients were also impaired in the serial order recognition task. Conclusions Despite mild intellectual disability, it is possible for short‐term retention capacities for verbal item information to be at an age‐appropriate level in VCFS. However, STM for serial order information could be impaired more specifically.     

An event-related fMRI study of the neural networks underlying repetition suppression and reaction time priming in implicit visual memory

Unfamiliar line drawings were presented to subjects three times during BOLD fMRI scanning. A set of brain areas was detected in which the effect of stimulus repetition on the evoked fMRI response depended on whether or not the drawing could be conceived as a coherent three-dimensional structure. Differential repetition effects were found in the neural response to drawings of both structurally possible and impossible objects. This differential effect of repetition was related to the amount of reaction time priming on the concurrent task involving decisions about three-dimensional structure in the possible but not in the impossible objects. These results point to different neurophysiological processing mechanisms for structurally possible and impossible images and demonstrate neural plasticity that predicts behavioral priming for structurally possible images.     

A combined neuropsychological and neuroimaging study of topographical and non-verbal memory in semantic dementia

A combined neuropsychological and neuroimaging investigation was carried out on a patient (O.I.) with semantic dementia who had asymmetrical temporal lobe atrophy, greater on the left. His performance on tests of verbal memory was gravely impaired. Similarly, his visual memory as indexed by recognition of unfamiliar faces was impaired. By contrast, his recognition memory for topographical memoranda (e.g. buildings, landscapes) and ability to find his way around was preserved. In order to identify the neural substrates supporting the preserved recognition of static topographical memoranda, O.I. was scanned using positron emission tomography (PET) during the encoding and recognition of building and landscape stimuli. In common with control subjects, during encoding O.I. activated parahippocampal cortex bilaterally, along with bilateral temporo-parietal, retrosplenial and left frontal cortices. During recognition, both patient and controls activated right parahippocampal, right superior parietal and right frontal cortices. Notably, control subjects, but not O.I., also activated at encoding the precuneus and at recognition the retrosplenial cortex. This allows the conclusion that these two areas while involved may not be necessary for topographical memory. Interestingly, the patient also activated regions that were not evident in control subjects both during encoding and recognition. These additional areas of activation may be necessary in a compensatory role. Overall, these data represent the first reported assessment of the functional integrity of degenerating brain tissue and its contribution to preserved topographical memory. The combination of the neuropsychological and neuroimaging approaches may provide insights into the functional-anatomy of memory while having clinical utility for the assessment of residual brain tissue.     

Functional neuroimaging of emotionally intense autobiographical memories in post-traumatic stress disorder

Post-traumatic stress disorder (PTSD) affects regions that support autobiographical memory (AM) retrieval, such as the hippocampus, amygdala and ventral medial prefrontal cortex (PFC). However, it is not well understood how PTSD may impact the neural mechanisms of memory retrieval for the personal past. We used a generic cue method combined with parametric modulation analysis and functional MRI (fMRI) to investigate the neural mechanisms affected by PTSD symptoms during the retrieval of a large sample of emotionally intense AMs. There were three main results. First, the PTSD group showed greater recruitment of the amygdala/hippocampus during the construction of negative versus positive emotionally intense AMs, when compared to controls. Second, across both the construction and elaboration phases of retrieval the PTSD group showed greater recruitment of the ventral medial PFC for negatively intense memories, but less recruitment for positively intense memories. Third, the PTSD group showed greater functional coupling between the ventral medial PFC and the amygdala for negatively intense memories, but less coupling for positively intense memories. In sum, the fMRI data suggest that there was greater recruitment and coupling of emotional brain regions during the retrieval of negatively intense AMs in the PTSD group when compared to controls.