Article

Independent features form integrated objects: Using a novel shape-color “conjunction task” to reconstruct memory resolution for multiple object features simultaneously

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Though much progress has been made to understand feature integration, debate remains regarding how objects are represented in mind based on their constituent features. Here, we advance this debate by introducing a novel shape-color “conjunction task” to reconstruct memory resolution for multiple object features simultaneously. In a first experiment, we replicate and extend a classic paradigm originally tested using a change detection task. Replicating previous work, memory resolution for individual features was reduced when the number of objects increased, regardless of the number of to-be-remembered features. Extending previous work, we found that high resolution memory near perfect in resemblance to the target was selectively impacted by the number of to-be-remembered features. Applying a data-driven statistical model of stochastic dependence, we found robust evidence of integration for lower-resolution feature memories, but less evidence for integration of high-resolution feature memories. These results suggest that memory resolution for individual features can be higher than memory resolution for their integration. In a second experiment which manipulated the nature of distracting information, we examined whether object features were directly bound to each other or by virtue of shared spatial location. Feature integration was disrupted by distractors sharing visual features of target objects but not when distractors shared spatial location – suggesting that feature integration can be driven by direct binding between shape and color features in memory. Our results constrain theoretical models of object representation, providing empirical support for hierarchical representations of both integrated and independent features.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... In [25], a memory model is confirmed in which both the features of the object and the object in the form of bound features are presented. Moreover, the features of an object can be bound not only through their common position, but directly with each other. ...
... In [26], a hierarchy of episode representations is considered, in which the levels of objects, events, and narratives are distinguished. Moreover, the representations of both objects and events also have a hierarchical structure: there exists a representation of both the object-whole and its features [25], and events are represented both by their details as well as by coarser global information. Perhaps there are different mechanisms at work to memorize these different levels of hierarchy. ...
Article
Full-text available
When storing multiple objects in visual working memory, observers sometimes misattribute perceived features to incorrect locations or objects. These misattributions are called binding errors (or swaps) and have been previously demonstrated mostly in simple objects whose features are easy to encode independently and arbitrarily chosen, like colors and orientations. Here, we tested whether similar swaps can occur with real-world objects, where the connection between features is meaningful rather than arbitrary. In Experiments 1 and 2, observers were simultaneously shown four items from two object categories. Within a category, the two exemplars could be presented in either the same or different states (e.g., open/closed; full/empty). After a delay, both exemplars from one of the categories were probed, and participants had to recognize which exemplar went with which state. We found good memory for state information and exemplar information on their own, but a significant memory decrement for exemplar-state combinations, suggesting that binding was difficult for observers and swap errors occurred even for meaningful real-world objects. In Experiment 3, we used the same task, but in one-half of the trials, the locations of the exemplars were swapped at test. We found that there are more errors in general when the locations of exemplars were swapped. We concluded that the internal features of real-world objects are not perfectly bound in working memory, and location updates impair object and feature representations. Overall, we provide evidence that even real-world objects are not stored in an entirely unitized format in working memory.
Article
Full-text available
Successful retrieval of an item from visual working memory (VWM) often requires an associated representation of the trial-unique context in which that item was presented. In experiment 1, fMRI of 16 male and female humans replicated a previous dissociation of the effects of manipulating memory load in comparison to the effects of manipulating context binding, by comparing VWM for one oriented line versus for three lines individuated by their location versus for three "heterogeneous" items drawn from different categories (orientation, color, and luminance): delay-period fMRI signal in frontal cortex and intraparietal sulcus (IPS) was sensitive to stimulus homogeneity rather than to memory load per se. Additionally, inspection of behavioral performance revealed a broad range of individual differences in the probability of responses to nontargets (also known as "swap errors"), and a post hoc comparison of high swap-error versus low swap-error groups generated several intriguing results: at recall, high swap-error subjects were seen to represent both the orientation and the location of the probed item less strongly, and with less differentiation from nonprobed items, and delay-period signal in IPS predicted behavioral and neural correlates of context binding at recall. In experiment 2, which was a preregistered replication, the 27 male and female humans were grouped into low and high swap-error groups by median split, and the results were broadly consistent with experiment 1. These results present a neural correlate of swap errors, and suggest that delay-period activity of the IPS may be more important for the operation of context binding than for representation per se of stimulus identity.
Article
Full-text available
Almost all models of visual memory implicitly assume that errors in mnemonic representations are linearly related to distance in stimulus space. Here we show that neither memory nor perception are appropriately scaled in stimulus space; instead, they are based on a transformed similarity representation that is nonlinearly related to stimulus space. This result calls into question a foundational assumption of extant models of visual working memory. Once psychophysical similarity is taken into account, aspects of memory that have been thought to demonstrate a fixed working memory capacity of around three or four items and to require fundamentally different representations—across different stimuli, tasks and types of memory—can be parsimoniously explained with a unitary signal detection framework. These results have substantial implications for the study of visual memory and lead to a substantial reinterpretation of the relationship between perception, working memory and long-term memory.
Article
Full-text available
Rich episodic experiences are represented in a hierarchical manner across a diverse network of brain regions, and as such, the way in which episodes are forgotten is likely to be similarly diverse. Using novel experimental approaches and statistical modelling, recent research has suggested that item-based representations, such as ones related to the colour and shape of an object, fragment over time, whereas higher-order event-based representations may be forgotten in a more ‘holistic’ uniform manner. We propose a framework that reconciles these findings, where complex episodes are represented in a hierarchical manner, from individual items, to small-scale events, to large-scale episodic narratives. Each level in the hierarchy is represented in distinct brain regions, from the perirhinal cortex, to posterior hippocampus, to anterior hippocampus and ventromedial prefrontal cortex. Critically, forgetting may be underpinned by different mechanisms at each level in the hierarchy, leading to different patterns of behaviour.
Article
Full-text available
Forgetting involves the loss of information over time; however, we know little about what form this information loss takes. Do memories become less precise over time, or do they instead become less accessible? Here we assessed memory for word–location associations across four days, testing whether forgetting involves losses in precision versus accessibility and whether such losses are modulated by learning a generalizable pattern. We show that forgetting involves losses in memory accessibility with no changes in memory precision. When participants learned a set of related word–location associations that conformed to a general pattern, we saw a strong trade-off; accessibility was enhanced, whereas precision was reduced. However, this trade-off did not appear to be modulated by time or confer a long-term increase in the total amount of information maintained in memory. Our results place theoretical constraints on how models of forgetting and generalization account for time-dependent memory processes. Protocol registration The stage 1 protocol for this Registered Report was accepted in principle on 4 June 2019. The protocol, as accepted by the journal, can be found at 10.6084/m9.figshare.c.4368464.v1.
Article
Full-text available
Experimental findings show the ubiquitous presence of graded responses and tuning curves in the neocortex, particularly in visual areas [1-15]. Among these, inferotemporal-cortex (IT) neurons respond to complex visual stimuli, but differences in the neurons' responses can be used to distinguish the stimuli eliciting the responses [8, 9, 16-18]. The IT projects directly to the medial temporal lobe (MTL) [19], where neurons respond selectively to different pictures of specific persons and even to their written and spoken names [20-22]. However, it is not clear whether this is done through a graded coding, as in the neocortex, or a truly invariant code, in which the response-eliciting stimuli cannot be distinguished from each other. To address this issue, we recorded single neurons during the repeated presentation of different stimuli (pictures and written and spoken names) corresponding to the same persons. Using statistical tests and a decoding approach, we found that only in a minority of cases can the different pictures of a given person be distinguished from the neurons' responses and that in a larger proportion of cases, the responses to the pictures were different to the ones to the written and spoken names. We argue that MTL neurons tend to lack a representation of sensory features (particularly within a sensory modality), which can be advantageous for the memory function attributed to this area [23-25], and that a full representation of memories is given by a combination of mostly invariant coding in the MTL with a representation of sensory features in the neocortex.
Article
Full-text available
Endel Tulving's proposal that episodic memory is distinct from other memory systems like semantic memory remains an extremely influential idea in cognitive neuroscience research. As originally suggested by Tulving, episodic memory involves three key components that differentiate it from all other memory systems: spatiotemporal binding, mental time travel, and autonoetic consciousness. Here, we focus on the idea of spatiotemporal binding in episodic memory and, in particular, how consideration of the precision of spatiotemporal context helps expand our understanding of episodic memory. Precision also helps shed light on another key issue in cognitive neuroscience, the role of the hippocampus outside of episodic memory in perception, attention, and working memory. By considering precision alongside item-context bindings, we attempt to shed new light on both the nature of how we represent context and what roles the hippocampus plays in episodic memory and beyond.
Article
Full-text available
One view is that conceptual knowledge is organized using the circuitry in the medial temporal lobe (MTL) that supports spatial processing and navigation. In contrast, we find that a domain-general learning algorithm explains key findings in both spatial and conceptual domains. When the clustering model is applied to spatial navigation tasks, so-called place and grid cell-like representations emerge because of the relatively uniform distribution of possible inputs in these tasks. The same mechanism applied to conceptual tasks, where the overall space can be higher-dimensional and sampling sparser, leading to representations more aligned with human conceptual knowledge. Although the types of memory supported by the MTL are superficially dissimilar, the information processing steps appear shared. Our account suggests that the MTL uses a general-purpose algorithm to learn and organize context-relevant information in a useful format, rather than relying on navigation-specific neural circuitry. Spatial maps in the medial temporal lobe (MTL) have been proposed to map abstract conceptual knowledge. Rather than grounding abstract knowledge in a spatial map, the authors propose a general-purpose clustering algorithm that explains how both spatial (including place and grid cells) and higher-dimensional conceptual representations arise during learning.
Article
Full-text available
Subjective similarity holds a prominent place in many psychological theories, influencing diverse cognitive processes ranging from attention and categorization to memory and problem solving. Despite the known importance of subjective similarity, there are few resources available to experimenters interested in manipulating the visual similarity of shape, one common type of subjective similarity. Here, across seven validation iterations, we incrementally developed a stimulus space consisting of 360 shapes using a novel image-processing method in conjunction with collected similarity judgments. The result is the Validated Circular Shape space, the first Validated Circular Shape space comparable to the commonly used "color wheel", whereby angular distance along a 2D circle is a proxy for visual similarity. This extensively validated resource is freely available to experimenters wishing to precisely manipulate the visual similarity of shape. (PsycINFO Database Record (c) 2019 APA, all rights reserved).
Article
Full-text available
In the history of the quest for human-level artificial intelligence, a number of rival paradigms have vied for supremacy. Symbolic artificial intelligence was dominant for much of the 20th century, but currently a connectionist paradigm is in the ascendant, namely machine learning with deep neural networks. However, both paradigms have strengths and weaknesses, and a significant challenge for the field today is to effect a reconciliation. A central tenet of the symbolic paradigm is that intelligence results from the manipulation of abstract compositional representations whose elements stand for objects and relations. If this is correct, then a key objective for deep learning is to develop architectures capable of discovering objects and relations in raw data, and learning how to represent them in ways that are useful for downstream processing. This short review highlights recent progress in this direction.
Article
Full-text available
Traversing sensory environments requires keeping relevant information in mind while simultaneously processing new inputs. Visual information is kept in working memory via feature-selective responses in early visual cortex, but recent work has suggested that new sensory inputs obligatorily wipe out this information. Here we show region-wide multiplexing abilities in classic sensory areas, with population-level response patterns in early visual cortex representing the contents of working memory alongside new sensory inputs. In a second experiment, we show that when people get distracted, this leads to both disruptions of mnemonic information in early visual cortex and decrements in behavioral recall. Representations in the intraparietal sulcus reflect actively remembered information encoded in a transformed format, but not task-irrelevant sensory inputs. Together, these results suggest that early visual areas play a key role in supporting high-resolution working memory representations that can serve as a template for comparison with incoming sensory information.
Article
Full-text available
Thanks to patients Phineas Gage and Henry Molaison, we have long known that behavioral control depends on the frontal lobes, whereas declarative memory depends on the medial temporal lobes (MTL). For decades, cognitive functions-behavioral control, declarative memory-have served as labels for characterizing the division of labor in cortex. This approach has made enormous contributions to understanding how the brain enables the mind, providing a systems-level explanation of brain function that constrains lower-level investigations of neural mechanism. Today, the approach has evolved such that functional labels are often applied to brain networks rather than focal brain regions. Furthermore, the labels have diversified to include both broadly-defined cognitive functions (declarative memory, visual perception) and more circumscribed mental processes (recollection, familiarity, priming). We ask whether a process-a high-level mental phenomenon corresponding to an introspectively-identifiable cognitive event-is the most productive label for dissecting memory. For example, recollection conflates a neurocomputational operation (pattern completion-based retrieval) with a class of representational content (associative, high-dimensional memories). Because a full theory of memory must identify operations and representations separately, and specify how they interact, we argue that processes like recollection constitute inadequate labels for characterizing neural mechanisms. Instead, we advocate considering the component operations and representations of processes like recollection in isolation. For the organization of memory, the evidence suggests that pattern completion is recapitulated widely across the ventral visual stream and MTL, but the division of labor between sites within this pathway can be explained by representational content.
Article
Full-text available
The question of whether visual working memory (VWM) stores individual features or bound objects as basic units is actively debated. Evidence exists for both feature-based and object-based storages, as well as hierarchically organized representations maintaining both types of information at different levels. One argument for feature-based storage is that features belonging to different dimensions (e.g., color and orientations) can be stored without interference suggesting independent capacities for every dimension. Here, we studied whether the lack of cross-dimensional interference reflects genuinely independent feature storages or mediated by common objects. In three experiments, participants remembered and recalled the colors and orientations of sets of objects. We independently manipulated set sizes within each feature dimension (making colors and orientations either identical or differing across objects). Critically, we assigned to-be-remembered colors and orientations either to same spatially integrated or to different spatially separated objects. We found that the precision and recall probability within each dimension was not affected by set size manipulations in a different dimension when the features belonged to integrated objects. However, manipulations with color set sizes did affect orientation memory when the features were separated. We conclude therefore that different feature dimensions can be encoded and stored independently but the advantage of the independent storages are mediated at the object-based level. This conclusion is consistent with the idea of hierarchically organized VWM.
Article
Full-text available
Episodic memories reflect a bound representation of multimodal features that can be reinstated with varying precision. Yet little is known about how brain networks involved in memory, including the hippocampus and posterior-medial (PM) and anterior-temporal (AT) systems, interact to support the quality and content of recollection. Participants learned color, spatial, and emotion associations of objects, later reconstructing the visual features using a continuous color spectrum and 360-degree panorama scenes. Behaviorally, dependencies in memory were observed for the gist but not precision of event associations. Supporting this integration, hippocampus, AT, and PM regions showed increased connectivity and reduced modularity during retrieval compared to encoding. These inter-network connections tracked a multidimensional, objective measure of memory quality. Moreover, distinct patterns of connectivity tracked item color and spatial memory precision. These findings demonstrate how hippocampal-cortical connections reconfigure during episodic retrieval, and how such dynamic interactions might flexibly support the multidimensional quality of remembered events.
Article
Full-text available
Episodic memory reflects the ability to recollect the temporal and spatial context of past experiences. Episodic memories depend on the hippocampus but have been proposed to undergo rapid forgetting unless consolidated during offline periods such as sleep to neocortical areas for long-term storage. Here, we propose an alternative to this standard systems consolidation theory (SSCT) — a contextual binding account — in which the hippocampus binds item-related and context-related information. We compare these accounts in light of behavioural, lesion, neuroimaging and sleep studies of episodic memory and contend that forgetting is largely due to contextual interference, episodic memory remains dependent on the hippocampus across time, contextual drift produces post-encoding activity and sleep benefits memory by reducing contextual interference.
Article
Full-text available
In the somatosensory nerves, the tactile perception of texture is driven by spatial and temporal patterns of activation distributed across three populations of afferents. These disparate streams of information must then be integrated centrally to achieve a unified percept of texture. To investigate the representation of texture in somatosensory cortex, we scanned a wide range of natural textures across the fingertips of rhesus macaques and recorded the responses evoked in Brodmann’s areas 3b, 1, and 2. We found that texture identity is reliably encoded in the idiosyncratic responses of populations of cortical neurons, giving rise to a high-dimensional representation of texture. Cortical neurons fall along a continuum in their sensitivity to fine vs. coarse texture, and neurons at the extrema of this continuum seem to receive their major input from different afferent populations. Finally, we show that cortical responses can account for several aspects of texture perception in humans.
Article
Full-text available
Face processing in autism spectrum disorder (ASD) is thought to be atypical, but it is unclear whether differences in visual conjunctive processing are specific to faces. To address this, we adapted a previously established eye-tracking paradigm which modulates the need for conjunctive processing by varying the degree of feature ambiguity in faces and objects. Typically-developed (TD) participants showed a canonical pattern of conjunctive processing: High-ambiguity objects were processed more conjunctively than low-ambiguity objects, and faces were processed in an equally conjunctive manner regardless of ambiguity level. In contrast, autistic individuals did not show differences in conjunctive processing based on stimulus category, providing evidence that atypical visual conjunctive processing in ASD is the result of a domain general lack of perceptual specialization. © 2019 Stevenson, Philipp-Muller, Hazlett, Wang, Luk, Lee, Black, Yeung, Shafai, Segers, Feber and Barense.
Article
Full-text available
How the neocortex works is a mystery. In this paper we propose a novel framework for understanding its function. Grid cells are neurons in the entorhinal cortex that represent the location of an animal in its environment. Recent evidence suggests that grid cell-like neurons may also be present in the neocortex. We propose that grid cells exist throughout the neocortex, in every region and in every cortical column. They define a location-based framework for how the neocortex functions. Whereas grid cells in the entorhinal cortex represent the location of one thing, the body relative to its environment, we propose that cortical grid cells simultaneously represent the location of many things. Cortical columns in somatosensory cortex track the location of tactile features relative to the object being touched and cortical columns in visual cortex track the location of visual features relative to the object being viewed. We propose that mechanisms in the entorhinal cortex and hippocampus that evolved for learning the structure of environments are now used by the neocortex to learn the structure of objects. Having a representation of location in each cortical column suggests mechanisms for how the neocortex represents object compositionality and object behaviors. It leads to the hypothesis that every part of the neocortex learns complete models of objects and that there are many models of each object distributed throughout the neocortex. The similarity of circuitry observed in all cortical regions is strong evidence that even high-level cognitive tasks are learned and represented in a location-based framework.
Article
Full-text available
Neural oscillations have been recorded and implicated in many different basic brain and cognitive processes. For example, oscillatory neural activity has been suggested to play a role in binding and in the maintenance of information in working memory. With respect to the latter, the majority of work has focused primarily on oscillations in terms of providing a “code” in working memory. However, oscillations may additionally play a fundamental role by enabling or facilitating essential properties and behaviors that neuronal networks must exhibit in order to produce functional working memory and the processes it supports, such as combining items in memory into bound objects or separating bound objects into distinct items. In the present work, we present a biologically plausible working memory model and demonstrate that specific types of stable oscillatory dynamics that arise may play critical roles in providing mechanisms for working memory and the cognitive functions that it supports. Specifically, these roles include (1) enabling a range of different types of binding, (2) both enabling and limiting capacities of bound and distinct items held active in working memory, and (3) facilitating transitions between active working memory states as required in cognitive function. Several key results arise within the examinations, such as the occurrence of different network capacities for working memory and binding, differences in processing times for transitions in working memory states, and the emergence of a combinatorially rich and complex range of oscillatory states that are sufficient to map onto a wide range of cognitive operations supported by working memory, such as variable binding, reasoning, and language. In particular, we show that these oscillatory states and their transitions can provide a specific instantiation of current established connectionist models in representing these functions. Finally, we further characterize the dependence of the relevant oscillatory solutions on certain critical parameters, including mutual inhibition and synaptic timescales.
Article
Full-text available
Visual short-term memory (VSTM) ability falls throughout the life span in healthy adults. Using a continuous report task, in a large, population-based sample, we first confirmed that this decline affects the quality and quantity of reported memories as well as knowledge of which item went where. Visual and sensorimotor precision also worsened with advancing age, but this did not account for the reduced memory performance. We then considered two strategies that older individuals might be able to adopt, to offset these memory declines: the use of contextual encoding, and metacognitive monitoring of performance. Context and metacognitive awareness were both associated with significantly better performance, however these effects did not interact with age in our sample. This suggests that older adults retain their capacity to boost memory performance through attention to external context and monitoring of their performance. Strategies that focus on taking advantage of these preserved abilities may therefore help to maintain VSTM performance with advancing age. The article reports on analysis of the Cambridge Centre for Ageing and Neuroscience (Cam-CAN) data.
Article
Full-text available
Natural environments convey information through multiple sensory modalities, all of which contribute to people’s percepts. Although it has been shown that visual or auditory content of scene categories can be decoded from brain activity, it remains unclear how humans represent scene information beyond a specific sensory modality domain. To address this question, we investigated how categories of scene images and sounds are represented in several brain regions. A group of healthy human subjects (both sexes) participated in the present study, where their brain activity was measured with fMRI while viewing images or listening to sounds of different real-world environments. We found that both visual and auditory scene categories can be decoded not only from modality-specific areas, but also from several brain regions in the temporal, parietal, and prefrontal cortex (PFC). Intriguingly, only in the PFC, but not in any other regions, categories of scene images and sounds appear to be represented in similar activation patterns, suggesting that scene representations in PFC are modality-independent. Furthermore, the error patterns of neural decoders indicate that category-specific neural activity patterns in the middle and superior frontal gyri are tightly linked to categorization behavior. Our findings demonstrate that complex scene information is represented at an abstract level in the PFC, regardless of the sensory modality of the stimulus.
Article
Full-text available
A significant body of research in cognitive neuroscience is aimed at understanding how object concepts are represented in the human brain. However, it remains unknown whether and where the visual and abstract conceptual features that define an object concept are integrated. We addressed this issue by comparing the neural pattern similarities among object-evoked fMRI responses with behavior-based models that independently captured the visual and conceptual similarities among these stimuli. Our results revealed evidence for distinctive coding of visual features in lateral occipital cortex, and conceptual features in the temporal pole and parahippocampal cortex. By contrast, we found evidence for integrative coding of visual and conceptual object features in perirhinal cortex. The neuroanatomical specificity of this effect was highlighted by results from a searchlight analysis. Taken together, our findings suggest that perirhinal cortex uniquely supports the representation of fully specified object concepts through the integration of their visual and conceptual features.
Article
Full-text available
It has been challenging to identify clinical cognitive markers that can differentiate patients with Alzheimer’s disease (AD) from those with behavioral variant frontotemporal dementia (bvFTD). The short-term memory binding (STMB) test assesses the ability to integrate colors and shapes into unified representations and to hold them temporarily during online performance. The objective of this study is to investigate whether free recall deficits during short-term memory binding (STMB) test can differentiate patients with AD from those with bvFTD and controls. Participants were 32 cognitively intact adults, 35 individuals with AD and 18 with bvFTD. All patients were in the mild dementia stage. Receiver-operating characteristic (ROC) analyses were used to examine the diagnostic accuracy of the STMB. The results showed that AD patients performed significantly worse than controls and bvFTD patients in the STMB test, while the latter groups showed equivalent performance. The bound condition of the STMB test showed an AUC of 0.853, with 84.4% of sensitivity and 80% of specificity to discriminate AD from controls and an AUC of 0.794, with 72.2% of sensitivity and 80% of specificity to differentiate AD from bvFTD. Binding deficits seem specific to AD. The free recall version of the STMB test can be used for clinical purposes and may aid in the differential diagnosis of AD. Findings support the view that the STMB may be a suitable cognitive marker for AD.
Article
Full-text available
Binding refers to the operation that groups different features together into objects.Wepropose a neural architecture for feature binding in visual working memory that employs populations of neurons with conjunction responses. We tested this model using cued recall tasks, in which subjects had to memorize object arrays composed of simple visual features (color, orientation, and location). After a brief delay, one feature of one item was given as a cue, and the observer had to report, on a continuous scale, one or two other features of the cued item. Binding failure in this task is associated with swap errors, in which observers report an item other than the one indicated by the cue. We observed that the probability of swapping two items strongly correlated with the items’ similarity in the cue feature dimension, and found a strong correlation between swap errors occurring in spatial and nonspatial report. The neural model explains both swap errors and response variability as results ofdecodingnoisy neural activity, and can account for the behavioral results in quantitative detail. We then used the model to compare alternative mechanisms for binding nonspatial features. We found the behavioral results fully consistent with a model in which nonspatial features are boundexclusively via their shared location,with no indication of direct binding between color and orientation. Theseresultsprovideevidencefor a special role of location in feature binding, and the model explains how this special role could be realized in the neural system.
Article
Full-text available
A growing literature suggests that the hippocampus is critical for the rapid extraction of regularities from the environment. Although this fits with the known role of the hippocampus in rapid learning, it seems at odds with the idea that the hippocampus specializes in memorizing individual episodes. In particular, the Complementary Learning Systems theory argues that there is a computational trade-off between learning the specifics of individual experiences and regularities that hold across those experiences. We asked whether it is possible for the hippocampus to handle both statistical learning and memorization of individual episodes. We exposed a neural network model that instantiates known properties of hippocampal projections and subfields to sequences of items with temporal regularities. We found that the monosynaptic pathway—the pathway connecting entorhinal cortex directly to region CA1—was able to support statistical learning, while the trisynaptic pathway—connecting entorhinal cortex to CA1 through dentate gyrus and CA3—learned individual episodes, with apparent representations of regularities resulting from associative reactivation through recurrence. Thus, in paradigms involving rapid learning, the computational trade-off between learning episodes and regularities may be handled by separate anatomical pathways within the hippocampus itself. This article is part of the themed issue ‘New frontiers for statistical learning in the cognitive sciences’.
Article
Full-text available
If we view a visual scene that contains many objects, then momentarily close our eyes, some details persist while others seem to fade. Discrete models of visual working memory (VWM) assume that only a few items can be actively maintained in memory, beyond which pure guessing will emerge. Alternatively, continuous resource models assume that all items in a visual scene can be stored with some precision. Distinguishing between these competing models is challenging, however, as resource models that allow for stochastically variable precision (across items and trials) can produce error distributions that resemble random guessing behavior. Here, we evaluated the hypothesis that a major source of variability in VWM performance arises from systematic variation in precision across the stimuli themselves; such stimulus-specific variability can be incorporated into both discrete-capacity and variable-precision resource models. Participants viewed multiple oriented gratings, and then reported the orientation of a cued grating from memory. When modeling the overall distribution of VWM errors, we found that the variable-precision resource model outperformed the discrete model. However, VWM errors revealed a pronounced "oblique effect," with larger errors for oblique than cardinal orientations. After this source of variability was incorporated into both models, we found that the discrete model provided a better account of VWM errors. Our results demonstrate that variable precision across the stimulus space can lead to an unwarranted advantage for resource models that assume stochastically variable precision. When these deterministic sources are adequately modeled, human working memory performance reveals evidence of a discrete capacity limit.
Article
Several measurement models have been proposed for data from the continuous-reproduction paradigm for studying visual working memory (WM): The original mixture model (Zhang & Luck, 2008) and its extension (Bays et al., 2009); the interference measurement model (IMM; Oberauer et al., 2017), and the target confusability competition (TCC) model (Schurgin et al., 2020). This article describes a space of possible measurement models in which all these models can be placed. The space is defined by three dimensions: (a) The choice of an activation function (von-Mises or Laplace), (b) the choice of a response-selection function (variants of Luce's choice rule or of signal-detection theory), (c) and whether or not memory precision is assumed to be a constant over manipulations affecting memory. A factorial combination of these three variables generates all possible models in the model space. Fitting all models to eight data sets revealed a new model as empirically most adequate, which combines a von-Mises activation function with a signal-detection response-selection rule. The precision parameter can be treated as a constant across many experimental manipulations, though it probably varies between individuals. All modeling code and the raw data modeled are available on the OSF: https://osf.io/zwprv/ (PsycInfo Database Record (c) 2021 APA, all rights reserved).
Article
Describing our visual experiences improves their retention in visual working memory, yielding a labeling benefit. Labels vary, however, in categorical distinctiveness: they can be applied broadly or narrowly to categorize stimuli. Does categorical distinctiveness constrain the labeling benefit? Here, we varied the number of terms used to label continuously varying colors (Experiment 1) and shapes (Experiment 2). Participants memorized four items, and later recalled them using a continuous color or shape wheel. During study, participants articulated “bababa” or labeled the items with two, four, or their preferred term. Recall error decreased with increases in the number of labels. Mixture modeling showed that labeling increased the probability of recall. Memory precision, however, varied with categorical distinctiveness: broad labels reduced precision, whereas categorically distinct labels increased precision compared to no-labels. In sum, in-the-moment labeling activates categorical knowledge that facilitates the storage of visual details. Data and analysis scripts are available at: https://osf.io/mqg4k/
Article
Visual working memory (VWM) allows us to actively represent a limited amount of visual information in mind. Although its severe capacity limit is widely accepted, researchers disagree on the nature of its representational unit. Object-based theories argue that VWM organizes feature representations into integrated representations, whereas feature-based theories argue that VWM represents visual features independently. Supporting a feature-based account of VWM, recent studies have demonstrated that features comprising an object can be forgotten independently. Although evidence of feature-based forgetting invalidates a pure object-based account of VWM that assumes perfect integration of feature representations, it is possible that feature representations may be organized in a dependent manner on the basis of objects when they exist in memory. Furthermore, many previous studies prompted participants to recall object features independently by sequentially displaying a response probe for each feature (i.e., sequential estimation procedure), and this task demand might have promoted the independence of feature representations in VWM. To test these possibilities, we created a novel task to simultaneously capture the representational quality of two features of the same object (i.e., simultaneous estimation procedure) and tested their dependence across the entire spectrum of representational quality. Here, we found that the quality of feature representations within the same object covaried reliably in both sequential and simultaneous estimation procedures, but this representational dependence was statistically stronger in the simultaneous estimation procedure than in the sequential estimation procedure. Furthermore, we confirmed that neither the shared spatial location nor simultaneous estimation of two features was sufficient to induce representational dependence in VWM. Thus, our results demonstrate that feature representations in VWM are organized in a dependent manner on the basis of objects, but the degree of dependence can vary based on the current task demand.
Article
Certain transformations must occur within the brain to allow rapid processing of familiar experiences. Complex objects are thought to become unitized, whereby multifeature conjunctions are retrieved as rapidly as a single feature. Behavioral studies strongly support unitization theory, but a compelling neural mechanism is lacking. Here, we examined how unitization transforms conjunctive representations to become more "feature-like" by recruiting posterior regions of the ventral visual stream (VVS) whose architecture is specialized for processing single features. We used functional magnetic resonance imaging to scan humans before and after visual training with novel objects. We implemented a novel multivoxel pattern analysis to measure a conjunctive code, which represented a conjunction of object features above and beyond the sum of the parts. Importantly, a multivoxel searchlight showed that the strength of conjunctive coding in posterior VVS increased posttraining. Furthermore, multidimensional scaling revealed representational separation at the level of individual features in parallel to the changes at the level of feature conjunctions. Finally, functional connectivity between anterior and posterior VVS was higher for novel objects than for trained objects, consistent with early involvement of anterior VVS in unitizing feature conjunctions in response to novelty. These data demonstrate that the brain implements unitization as a mechanism to refine complex object representations over the course of multiple learning experiences.
Article
The ability to remember associations among components of an event, which is central to episodic memory, declines with normal aging. In accord with the specificity principle of memory, these declines may occur because associative memory requires retrieval of specific information. Guided by this principle, we endeavored to determine whether ubiquitous age-related deficits in associative memory are restricted to specific representations or extend to the gist of associations. Young and older adults (30 each in Experiment 1, 40 each in Experiment 2) studied face–scene pairs and then performed associative-recognition tests following variable delays. Whereas both young and older adults could retrieve the gist of associations, older adults were impaired in their ability to retrieve more specific representations. Our results also show that associations can be retrieved from multiple levels of specificity, suggesting that episodic memory might be accessed on a continuum of specificity.
Article
Episodic memory declines with older age, but it is unresolved whether this decline reflects reduced probability of successfully retrieving information from memory, or decreased precision of the retrieved information. Here, we used continuous measures of episodic memory retrieval in combination with computational mixture modeling of participants' retrieval errors to distinguish between these two potential accounts of age-related memory deficits. In three experiments, young and older participants encoded stimulus displays consisting of everyday objects varying along different perceptual features (e.g., location, color and orientation) in a circular space. At test, participants recreated the features of studied objects using a continuous response dial. Across all 3 experiments, we observed significant age-related declines in the precision of episodic memory retrieval, whereas significant age differences in retrieval success were limited to the most challenging task condition. Reductions in mnemonic precision were evident across different object features retained in long-term memory and persisted after controlling for age-related decreases in the fidelity of perception and working memory. The findings highlight impoverished precision of memory representations as one factor contributing to age-related episodic memory loss and suggest that the cognitive and neural changes associated with older age may differentially affect distinct aspects of episodic retrieval. (PsycINFO Database Record (c) 2020 APA, all rights reserved).
Article
The hallmark of episodic memory is recollecting multiple perceptual details tied to a specific spatial-temporal context. To remember an event, it is therefore necessary to integrate such details into a coherent representation during initial encoding. Here we tested how the brain encodes and binds multiple, distinct kinds of features in parallel, and how this process evolves over time during the event itself. We analyzed data from 27 human subjects (16 females, 11 males) who learned a series of objects uniquely associated with a color, a panoramic scene location, and an emotional sound while fMRI data were collected. By modeling how brain activity relates to memory for upcoming or just-viewed information, we were able to test how the neural signatures of individual features as well as the integrated event changed over the course of encoding. We observed a striking dissociation between early and late encoding processes: left inferior frontal and visuo-perceptual signals at the onset of an event tracked the amount of detail subsequently recalled and were dissociable based on distinct remembered features. In contrast, memory-related brain activity shifted to the left hippocampus toward the end of an event, which was particularly sensitive to binding item color and sound associations with spatial information. These results provide evidence of early, simultaneous feature-specific neural responses during episodic encoding that predict later remembering and suggest that the hippocampus integrates these features into a coherent experience at an event transition.
Article
In the present study, we examined how real-world objects are represented in long-term memory. Two contrasting views exist with regard to this question: one argues that real-world objects are represented as a set of independent features, and the other argues that they form bound integrate representations. In 5 experiments, we tested the different predictions of each view, namely whether the different features of real-world items are remembered and forgotten independently from each other, in a feature-based manner, or conversely are stored and lost in an object-based manner, with all features depending upon each other. Across various stimuli, learning tasks (incidental or explicit), experimental setups (within- or between-subjects design), feature-dimensions, and encoding times, we consistently found that information is forgotten in an object-based manner. When an object ceases to be fully remembered, all of its features are lost, instead of only some of the object's features being lost whereas other features are still remembered. Furthermore, we found support for a strong form of dependency among the different features, namely a hierarchical structure. We conclude that visual long-term memory is object-based, challenging previous findings. (PsycINFO Database Record (c) 2019 APA, all rights reserved).
Article
People can store thousands of real-world objects in visual long-term memory with high precision. But are these objects stored as unitary, bound entities, as often assumed, or as bundles of separable features? We tested this in several experiments. In the first series of studies, participants were instructed to remember specific exemplars of real-world objects presented in a particular state (e.g., open/closed, full/empty, etc.), and then were asked to recognize either which exemplars they had seen (e.g., I saw this coffee mug), or which exemplar-state conjunctions they had seen (e.g., I saw this coffee mug and it was full). Participants had a large number of within-category confusions, for example misremembering which states went with which exemplars, while simultaneously showing strong memory for the features themselves (e.g., which states they had seen, which exemplars they had seen). In a second series of studies, we found further evidence of independence: participants were very good at remembering which exemplars they had seen independently of whether these items were presented in a new or old state, but the same did not occur for features known to be truly holistically represented. Thus, we find through 2 lines of evidence that the features of real-world objects that support exemplar discrimination and state discrimination are not bound, suggesting visual objects are not inherently unitary entities in memory. (PsycINFO Database Record (c) 2019 APA, all rights reserved).
Article
Do complex event representations fragment over time, or are they instead forgotten in an all-or-none manner? For example, if we met a friend in a café and they gave us a present, do we forget the constituent elements of this event (location, person, and object) independently, or would the whole event be forgotten? Research suggests that item-based memories are forgotten in a fragmented manner. However, we do not know how more complex episodic, event-based memories are forgotten. We assessed both retrieval accuracy and dependency-the statistical association between the retrieval successes of different elements from the same event-for complex events. Across 4 experiments, we show that retrieval dependency is found both immediately after learning and following a 12-hr and 1-week delay. Further, the amount of retrieval dependency after a delay is greater than that predicted by a model of independent forgetting. This dependency was only seen for coherent "closed-loops," where all pairwise associations between locations, people, and objects were encoded. When "open-loops" were learned, where only 2 out of the 3 possible associations were encoded, no dependency was seen immediately after learning or after a delay. Finally, we also provide evidence for higher retention rates for closed-loops than for open-loops. Therefore, closed-loops do not fragment as a function of forgetting and are retained for longer than are open-loops. Our findings suggest that coherent episodic events are not only retrieved, but also forgotten, in an all-or-none manner. (PsycINFO Database Record (c) 2019 APA, all rights reserved).
Article
Anne Treisman’s Feature Integration Theory (FIT) is a landmark in cognitive psychology and vision research. While many have discussed how Treisman’s theory has fared since it was first proposed, it is less common to approach FIT from the other side in time: to examine what experimental findings, theoretical concepts, and ideas inspired it. The theory did not enter into a theoretical vacuum. Treisman’s ideas were inspired by a large literature on a number of topics within visual psychophysics, cognitive psychology, and visual neurophysiology. Several key ideas developed contemporaneously within these fields that inspired FIT, and the theory involved an attempt at integrating them. Our aim here was to highlight the conceptual problems, experimental findings, and theoretical positions that Treisman was responding to with her theory and that the theory was intended to explain. We review a large number of findings from the decades preceding the proposal of feature integration theory showing how the theory integrated many ideas that developed in parallel within neurophysiology, visual psychophysics, and cognitive psychology. Our conclusion is that FIT made sense of many preceding findings, integrating them in an elegant way within a single theoretical account.
Article
To accomplish even rudimentary tasks, our cognitive system must update its representation of the changing environment. This process relies on visual working memory (VWM), which can actively modify its representations. We argue that this ability depends on a pointer system, such that each representation is stably and uniquely mapped to a specific stimulus. Without these pointers, VWM representations are inaccessible and therefore unusable. In three Electroencephalogram (EEG) experiments, we examined whether the pointers are allocated in an object-based, featural, or spatial manner: three factors that were confounded in previous studies. We used a feature change-detection task, in which objects moved and could separate into independently-moving parts. Despite the movement and separation being completely task-irrelevant, we found that the separation invalidated the pointers. This happened in a shape task, where the separation changed both the objects and the task-relevant features, but importantly, also in a color task, where the separation destroyed the objects while leaving the task-relevant features intact. Furthermore, even in a color task where all items had identical shapes, object-separation invalidated the pointers. This suggests that objects and not task-relevant features underlie the pointer system. Finally, when each object-part could be individuated already before the separation, the pointers were maintained, suggesting that the pointers are specifically tied to objects rather than locations. These results shed new light on the pointers which underlie VWM performance, demonstrating that the pointer system is object-based regardless of the task requirements.
Article
The lateral portion of the entorhinal cortex is one of the first brain regions affected by tau pathology, an important biomarker for Alzheimer's disease. Improving our understanding of this region's cognitive role may help identify better cognitive tests for early detection of Alzheimer's disease. Based on its functional connections, we tested the idea that the human anterolateral entorhinal cortex (alERC) may play a role in integrating spatial information into object representations. We recently demonstrated that the volume of the alERC was related to processing the spatial relationships of the features within an object [Yeung, L. K., Olsen, R. K., Bild-Enkin, H. E. P., D'Angelo, M. C., Kacollja, A., McQuiggan, D. A., et al. Anterolateral entorhinal cortex volume predicted by altered intra-item configural processing. Journal of Neuroscience, 37, 5527-5538, 2017]. In this study, we investigated whether the human alERC might also play a role in processing the spatial relationships between an object and its environment using an eye-tracking task that assessed visual fixations to a critical object within a scene. Guided by rodent work, we measured both object-in-place memory, the association of an object with a given context [Wilson, D. I., Langston, R. F., Schlesiger, M. I., Wagner, M., Watanabe, S., & Ainge, J. A. Lateral entorhinal cortex is critical for novel object-context recognition. Hippocampus, 23, 352-366, 2013], and object-trace memory, the memory for the former location of objects [Tsao, A., Moser, M. B., & Moser, E. I. Traces of experience in the lateral entorhinal cortex. Current Biology, 23, 399-405, 2013]. In a group of older adults with varying stages of brain atrophy and cognitive decline, we found that the volume of the alERC and the volume of the parahippocampal cortex selectively predicted object-in-place memory, but not object-trace memory. These results provide support for the notion that the alERC may integrate spatial information into object representations.
Article
Location appears to play a vital role in binding discretely processed visual features into coherent objects. Consequently, it has been proposed that objects are represented for cognition by their spatiotemporal location, with other visual features attached to this location index. On this theory, the visual features of an object are only connected via mutual location; direct binding cannot occur. Despite supporting evidence, some argue that direct binding does take over according to task demands and when representing familiar objects. The current study was developed to evaluate these claims, using a brief memory task to test for contingencies between features under different circumstances. Participants were shown a sequence of three items in different colours and locations, and then asked for the colour and/or location of one of them. The stimuli could either be abstract shapes, or familiar objects. Results indicated that location is necessary for binding regardless of the type of stimulus and task demands, supporting the proposed structure. A follow-up experiment assessed an alternate explanation for the apparent importance of location in binding; eye movements may automatically capture location information, making it impossible to ignore and suggesting a contingency that is not representative of cognitive processes. Participants were required to maintain fixation on half of the trials, with an eye tracker for confirmation. Results indicated that the importance of location in binding cannot be attributed to eye movements. Overall, the findings of this study support the claim that location is essential for visual feature binding, due to the structure of object representations.
Article
It is proposed that a cognitive map encoding the relationships between entities in the world supports flexible behavior, but the majority of the neural evidence for such a system comes from studies of spatial navigation. Recent work describing neuronal parallels between spatial and non-spatial behaviors has rekindled the notion of a systematic organization of knowledge across multiple domains. We review experimental evidence and theoretical frameworks that point to principles unifying these apparently disparate functions. These principles describe how to learn and use abstract, generalizable knowledge and suggest that map-like representations observed in a spatial context may be an instance of general coding mechanisms capable of organizing knowledge of all kinds. We highlight how artificial agents endowed with such principles exhibit flexible behavior and learn map-like representations observed in the brain. Finally, we speculate on how these principles may offer insight into the extreme generalizations, abstractions, and inferences that characterize human cognition. Behrens et al. review an emerging field building formalisms for understanding the neural basis of flexible behavior. The authors extend these ideas toward representations useful for generalization and structural abstraction, allowing rapid inferences and flexible behavior with little direct experience.
Article
Efficient navigation from one place to another is facilitated by the ability to use spatial boundaries to segment routes into their component parts. Similarly, memory for individual episodes relies on the ability to use shifts in spatiotemporal contexts to segment the ongoing stream of experience. The segmentation of experiences in spatial and episodic domains may therefore share neural underpinnings, manifesting in similar behavioral phenomena and cognitive biases. Here, we review evidence for such shared mechanisms, focusing on the key role of boundaries in spatial and episodic memory. We propose that a fundamental event boundary detection mechanism enables navigation in both the spatial and episodic domains, and serves to form cohesive representations that can be used to predict and guide future behavior.
Article
Refreshing and elaboration are cognitive processes assumed to underlie verbal working-memory maintenance and assumed to support long-term memory formation. Whereas refreshing refers to the attentional focussing on representations, elaboration refers to linking representations in working memory into existing semantic networks. We measured the impact of instructed refreshing and elaboration on working and long-term memory separately, and investigated to what extent both processes are distinct in their contributions to working as well as long-term memory. Compared with a no-processing baseline, immediate memory was improved by repeating the items, but not by refreshing them. There was no credible effect of elaboration on working memory, except when items were repeated at the same time. Long-term memory benefited from elaboration, but not from refreshing the words. The results replicate the long-term memory benefit for elaboration, but do not support its beneficial role for working memory. Further, refreshing preserves immediate memory, but does not improve it beyond the level achieved without any processing.
Article
Some theories of episodic memory hypothesize that spatial context plays a fundamental role in episodic memory, acting as a scaffold on which episodes are constructed. A prediction based on this hypothesis is that spatial context should play a primary role in the neural representation of an event. To test this hypothesis in humans, male and female participants imagined events, composed of familiar locations, people, and objects, during an fMRI scan. We used multivoxel pattern analysis to determine the neural areas in which events could be discriminated based on each feature. We found that events could be discriminated according to their location in areas throughout the autobiographical memory network, including the parahippocampal cortex and posterior hippocampus, retrosplenial cortex, posterior cingulate cortex, precuneus, and medial prefrontal cortex. Events were also discriminable based on person and object features, but in fewer regions. Comparing classifier performance in regions involved in memory for scenes and events demonstrated that the location of an event was more accurately classified than the person or object involved. These results support theories that suggest that spatial context is a prominent defining feature of episodic memory.
Article
A network of brain regions have been linked with episodic memory retrieval, but limited progress has been made in identifying the contributions of distinct parts of the network. Here, we utilized continuous measures of retrieval to dissociate three components of episodic memory: retrieval success, precision, and vividness. In the fMRI scanner, participants encoded objects that varied continuously on three features: color, orientation, and location. Participants' memory was tested by having them recreate the appearance of the object features using a continuous dial, and continuous vividness judgments were recorded. Retrieval success, precision, and vividness were dissociable both behaviorally and neurally: successful versus unsuccessful retrieval was associated with hippocampal activity, retrieval precision scaled with activity in the angular gyrus, and vividness judgments tracked activity in the precuneus. The ability to dissociate these components of episodic memory reveals the benefit afforded by measuring memory on a continuous scale, allowing functional parcellation of the retrieval network.
Article
How do we find what we are looking for? Even when the desired target is in the current field of view, we need to search because fundamental limits on visual processing make it impossible to recognize everything at once. Searching involves directing attention to objects that might be the target. This deployment of attention is not random. It is guided to the most promising items and locations by five factors discussed here: bottom-up salience, top-down feature guidance, scene structure and meaning, the previous history of search over timescales ranging from milliseconds to years, and the relative value of the targets and distractors. Modern theories of visual search need to incorporate all five factors and specify how these factors combine to shape search behaviour. An understanding of the rules of guidance can be used to improve the accuracy and efficiency of socially important search tasks, from security screening to medical image perception.
Article
A key feature of human thought and language is compositionality, the ability to bind pre-existing concepts or word meanings together in order to express new ideas. Here we ask how newly composed complex concepts are mentally represented and matched to the outside world, by testing whether it is harder to verify if a picture matches the meaning of a phrase, like big pink tree, than the meaning of a single word, like tree. Five sentence-picture verification experiments provide evidence that, in fact, the meaning of a phrase can often be checked just as fast as the meaning of one single word (and sometimes faster), indicating that the phrase’s constituent concepts can be represented and checked in parallel. However, verification times were increased when matched phrases had more complex modification structures, indicating that it is costly to represent structural relations between constituent concepts. This pattern of data can be well-explained if concepts are composed together using two different mechanisms, binding by synchrony and binding by asynchrony, which have been suggested as solutions to the “binding problem” faced in both vision science and higher-level cognition. Our results suggest that they can also explain aspects of compositional language processing.
Article
Harmonicity is a fundamental element of music, speech, and animal vocalizations. How the auditory system extracts harmonic structures embedded in complex sounds and uses them to form a coherent unitary entity is not fully understood. Despite the prevalence of sounds rich in harmonic structures in our everyday hearing environment, it has remained largely unknown what neural mechanisms are used by the primate auditory cortex to extract these biologically important acoustic structures. In this study, we discovered a unique class of harmonic template neurons in the core region of auditory cortex of a highly vocal New World primate, the common marmoset (Callithrix jacchus), across the entire hearing frequency range. Marmosets have a rich vocal repertoire and a similar hearing range to that of humans. Responses of these neurons show nonlinear facilitation to harmonic complex sounds over inharmonic sounds, selectivity for particular harmonic structures beyond two-tone combinations, and sensitivity to harmonic number and spectral regularity. Our findings suggest that the harmonic template neurons in auditory cortex may play an important role in processing sounds with harmonic structures, such as animal vocalizations, human speech, and music.
Article
Over the past decade, it has been debated whether retaining bindings in working memory (WM) requires more attention than retaining constituent features, focusing on domain-general attention and space-based attention. Recently, we proposed that retaining bindings in WM needs more object-based attention than retaining constituent features (Shen, Huang, & Gao, 2015, Journal of Experimental Psychology: Human Perception and Performance, doi:10.1037/xhp0000018). However, only unitized visual bindings were examined; to establish the role of object-based attention in retaining bindings in WM, more emperical evidence is required. We tested 4 new bindings that had been suggested requiring no more attention than the constituent features in the WM maintenance phase: The two constituent features of binding were stored in different WM modules (cross-module binding, Experiment 1), from auditory and visual modalities (cross-modal binding, Experiment 2), or temporally (cross-time binding, Experiments 3) or spatially (cross-space binding, Experiments 4–6) separated. In the critical condition, we added a secondary object feature-report task during the delay interval of the change-detection task, such that the secondary task competed for object-based attention with the to-be-memorized stimuli. If more object-based attention is required for retaining bindings than for retaining constituent features, the secondary task should impair the binding performance to a larger degree relative to the performance of constituent features. Indeed, Experiments 1–6 consistently revealed a significantly larger impairment for bindings than for the constituent features, suggesting that object-based attention plays a pivotal role in retaining bindings in WM.