One of the major lessons memory research has been that human is fallible, imprecise, and subject to interference. Thus, although observers can remember thousands images, it widely assumed these memories lack detail. Contrary this assumption, here we show long-term capable storing a massive number objects with details from image. Participants viewed pictures 2,500 over course 5.5 h. Afterward, they were shown pairs images indicated which two had seen. The previously item could be paired...

Humans have a massive capacity to store detailed information in visual long-term memory. The present studies explored the fidelity of these memory representations and examined how conceptual perceptual features object categories support this capacity. Observers viewed 2,800 images with different number exemplars presented from each category. At test, observers indicated which 2 they had previously studied. Memory performance was high remained quite (82% accuracy) 16 category memory,...

Occipito-temporal cortex is known to house visual object representations, but the organization of neural activation patterns along this still being discovered. Here we found a systematic, large-scale structure in responses related interaction between two major cognitive dimensions representation: animacy and real-world size. Neural were measured with functional magnetic resonance imaging while human observers viewed images big small animals objects. We that size drives differential only...

Observers can store thousands of object images in visual long-term memory with high fidelity, but the fidelity scene representations is not known. Here, we probed scene-representation by varying number studied exemplars different categories and testing using exemplar-level foils. viewed scenes over 5.5 hr then completed a series forced-choice tests. Memory performance was high, even up to 64 from same category memory. Moreover, there only 2% decrease accuracy for each doubling exemplars....

The information that individuals can hold in working memory is quite limited, but researchers have typically studied this capacity using simple objects or letter strings with no associations between them. However, the real world there are strong and regularities input. In an theoretic sense, introduce redundancies make input more compressible. current study shows observers take advantage of these redundancies, enabling them to remember items memory. 2 experiments, covariance was introduced...

Visual long-term memory can store thousands of objects with surprising visual detail, but just how detailed are these representations, and one quantify this fidelity? Using the property color as a case study, we estimated precision information in memory, compared same working memory. Observers were shown real-world random colors asked to recall after delay. We quantified two parameters performance: variability internal representations (fidelity) probability forgetting an object’s altogether....

Human object-selective cortex shows a large-scale organization characterized by the high-level properties of both animacy and object size. To what extent are these neural responses explained primitive perceptual features that distinguish animals from objects big small objects? address this question, we used texture synthesis algorithm to create class stimuli-texforms-which preserve some mid-level form information while rendering them unrecognizable. We found unrecognizable texforms were...

Abstract Anterior regions of the ventral visual stream encode substantial information about object categories. Are top-down category-level forces critical for arriving at this representation, or can representation be formed purely through domain-general learning natural image structure? Here we present a fully self-supervised model which learns to represent individual images, rather than categories, such that views same are embedded nearby in low-dimensional feature space, distinctly from...

A large body of literature has shown that observers often fail to notice significant changes in visual scenes, even when these happen right front their eyes. For instance, people if conversation partner is switched another person, or background objects suddenly disappear.1,2 These 'change blindness' studies have led the inference amount information we remember about each item a scene may be quite low.1 However, recent work demonstrated long-term memory capable storing massive number with...

Real-world objects can be viewed at a range of distances and thus experienced visual angles within the field. Given large amount size variation possible when observing objects, we examined how internal object representations represent information. In series experiments which required observers to access existing knowledge, observed that real-world have consistent they are drawn, imagined, preferentially viewed. Importantly, this is proportional logarithm assumed in world, best characterized...

Are real-world objects represented as bound units?While a great deal of research has examined binding between the feature dimensions simple shapes, little work whether featural properties are stored in single unitary object representation.In first experiment, we find that information about an object's color is forgotten more rapidly than state (e.g.open, closed), suggesting observers do not forget entirely units.In second and third examine exemplar separately or together.If these separately,...

Significance Human cognition is inherently limited: only a finite amount of visual information can be processed at given instant. What determines those limits? Here, we show that more objects when they are from different stimulus categories than the same category. This behavioral benefit maps directly onto functional organization ventral pathway. These results suggest our ability to process multiple items once limited by extent which compete with one another for neural representation....

Understanding how perceptual and conceptual representations are connected is a fundamental goal of cognitive science. Here, we focus on broad distinction that constrains interact with objects--real-world size. Although there appear to be clear correlates for basic-level categories (apples look like other apples, oranges oranges), the broader categorical distinctions largely unexplored, i.e., do small objects objects? Because many kinds (e.g., cups, keys), may no reliable features distinguish...

Estimating the size of a space and its degree clutter are effortless ubiquitous tasks moving agents in natural environment. Here, we examine how regions along occipital–temporal lobe respond to pictures indoor real-world scenes that parametrically vary their physical "size" (the spatial extent bounded by walls) functional "clutter" organization quantity objects fill up space). Using linear regression model on multivoxel pattern activity across interest, find evidence both properties...

Visual search is a ubiquitous visual behavior, and efficient essential for survival. Different cognitive models have explained the speed accuracy of based either on dynamics attention or similarity item representations. Here, we examined extent to which performance task can be predicted from stable representational architecture system, independent attentional dynamics. Participants performed with 28 conditions reflecting different pairs categories (e.g., searching face among cars, body...

Abstract The rapid development and open-source release of highly performant computer vision models offers new potential for examining how different inductive biases impact representation learning emergent alignment with the high-level human ventral visual system. Here, we assess a diverse set 224 models, curated to enable controlled comparison model properties, testing their brain predictivity using large-scale functional magnetic resonance imaging data. We find that qualitatively...

Modular and distributed coding theories of category selectivity along the human ventral visual stream have long existed in tension. Here, we present a reconciling framework—contrastive coding—based on series analyses relating within biological artificial neural networks. We discover that, models trained with contrastive self-supervised objectives over rich natural image diet, category-selective tuning naturally emerges for faces, bodies, scenes, words. Further, lesions these model units lead...

When we recognize an object, do automatically know how big it is in the world? We employed a Stroop-like paradigm, which two familiar objects were presented at different visual sizes on screen. Observers faster to indicate was bigger or smaller screen when real-world size of congruent with than incongruent--demonstrating familiar-size Stroop effect. Critically, irrelevant for task. This effect also present only one item incongruent display. In contrast, no observed participants who simply...

Abstract Humans observe a wide range of actions in their surroundings. How is the visual cortex organized to process this diverse input? Using functional neuroimaging, we measured brain responses while participants viewed short videos everyday actions, then probed structure these using voxel-wise encoding modeling. Responses are well fit by feature spaces that capture body parts involved an action and action’s targets (i.e. whether was directed at object, another person, actor, space)....

The human ventral visual stream has a highly systematic organization of object information, but the causal pressures driving these topographic motifs are debated. Here, we use self-organizing principles to learn representation data manifold deep neural network representational space. We find that smooth mapping this space showed many brain-like motifs, with large-scale by animacy and real-world size, supported mid-level feature tuning, naturally emerging face- scene-selective regions. While...