Perception and memory have distinct spatial tuning properties in human visual cortex

Beyond Routine Maintenance: Current Trends in Working Memory Research

Working memory (WM) is an evolving concept. Our understanding of the neural functions that support WM develops iteratively alongside the approaches used to study it, and both can be profoundly shaped by available tools and prevailing theoretical paradigms. Here, the organizers of the 2024 Working Memory Symposium-inspired by this year's meeting-highlight current trends and looming questions in WM research. This review is organized into sections describing (1) ongoing efforts to characterize WM function across sensory modalities, (2) the growing appreciation that WM representations are malleable to context and future actions, (3) the enduring problem of how multiple WM items and features are structured and integrated, and (4) new insights about whether WM shares function with other cognitive processes that have conventionally been considered distinct. This review aims to chronicle where the field is headed and calls attention to issues that are paramount for future research.

Visual imagery of familiar people and places in category selective cortex

Visual imagery is a dynamic process recruiting a network of brain regions. We used electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) fusion to investigate the dynamics of category selective imagery in medial parietal cortex (MPC), ventral temporal cortex (VTC), and primary visual cortex (V1). Subjects attended separate EEG and fMRI sessions where they created mental images of personally familiar people and place stimuli. The fMRI contrast comparing people and place imagery replicated previous findings of category-selectivity in the medial parietal cortex. In addition, greater activity for places was found in the ventral and lateral place memory areas, the frontal eye fields, the inferior temporal sulcus, and the intraparietal sulcus. In contrast, greater activity for people was found in the fusiform face area and the right posterior superior temporal sulcus. Using multivariate decoding analysis in fMRI, we could decode individual stimuli within the preferred category in VTC. A more complex pattern emerged in MPC, which represented information that was not restricted to the preferred category. We were also able to decode category and individual stimuli in the EEG data. EEG-fMRI fusion indicated similar timings in MPC and VTC activity during imagery. However, in the VTC, fusion was higher in place selective regions during an early time window, and higher in face selective regions in a later time window. In contrast, fusion correlations in V1 occurred later during the imagery period, possibly reflecting the top-down progression of mental imagery from category-selective regions to primary visual cortex.

Advanced concentrative absorption meditation reorganizes functional connectivity gradients of the brain: 7T MRI and phenomenology case study of jhana meditation

There is growing scientific interest in advanced meditation, and particularly the Theravada Buddhist advanced concentrative absorption meditation known as jhana (ACAM-J). ACAM-J includes a series of eight consecutive meditative states, which are radically altered states of consciousness. The neuroscience of ACAM-J, specifically brain reorganization, remains underspecified in part due to the difficulty of finding and studying expert ACAM-J meditators and challenges related to laboratory investigation of ACAM-J. Using a nonlinear dimensionality reduction technique applied to human functional neuroimaging in an intensive case study, we investigated brain reorganization during ACAM-J. We applied linear mixed models and correlations to explore relations among brain reorganization and ACAM-J phenomenology. Results demonstrated that ACAM-J induces disruption of the hierarchical organization of the brain by shifting the gradients toward a more globally integrated rather than segregated state between sensory-related and higher-order cognitive regions. Additionally, ACAM-J induces a separation between sensory-related and attention modulation-related regions, resulting in greater differentiation in functional organization of these regions, consistent with phenomenological reports. This study highlights the need for further research into brain reorganization and health-related implications of both short-term and long-term practice of ACAM-J. Key points/highlights The neuroscience of advanced concentrative absorption meditation (ACAM) has the potential to improve our knowledge of well-being and altered states of consciousness but remains underexplored due to methodological challenges. We investigated functional reorganization of the brain during ACAM-J using gradient analysis and demonstrated that ACAM-J disrupts the hierarchical organization of the brain during meditation. Additionally, we demonstrated that ACAM-J increases differentiation between primary sensory areas and areas related to attention modulation.

Feedback scales the spatial tuning of cortical responses during both visual working memory and long-term memory

Perception, working memory, and long-term memory each evoke neural responses in visual cortex. While previous neuroimaging research on the role of visual cortex in memory has largely emphasized similarities between perception and memory, we hypothesized that responses in visual cortex would differ depending on the origins of the inputs. Using fMRI, we quantified spatial tuning in visual cortex while participants (both sexes) viewed, maintained in working memory, or retrieved from long-term memory a peripheral target. In each condition, BOLD responses were spatially tuned and aligned with the target's polar angle in all measured visual field maps including V1. As expected given the increasing sizes of receptive fields, polar angle tuning during perception increased in width up the visual hierarchy from V1 to V2, V3, hV4, and beyond. In stark contrast, the tuned responses were broad across the visual hierarchy during long-term memory (replicating a prior result) and during working memory. This pattern is consistent with the idea that mnemonic responses in V1 stem from top-down sources, even when the stimulus was recently viewed and is held in working memory. Moreover, in long-term memory, trial-to-trial biases in these tuned responses (clockwise or counterclockwise of target), predicted matched biases in memory, suggesting that the reinstated cortical responses influence memory guided behavior. We conclude that feedback widens spatial tuning in visual cortex during memory, where earlier visual maps inherit broader tuning from later maps thereby impacting the precision of memory.Significance Statement We demonstrate that remembering a visual stimulus evokes responses in visual cortex that differ in spatial extent compared to seeing the same stimulus. Perception evokes tuned responses in early visual areas that increase in size up the visual hierarchy. Prior work showed that feedback inputs associated with long-term memory originate from later visual areas with larger receptive fields resulting in uniformly wide spatial tuning even in primary visual cortex. We replicate these results and show that the same pattern holds when maintaining in working memory a recently viewed stimulus. That trial-to-trial difficulty is reflected in the accuracy and precision of these representations suggests that visual cortex is flexibly used for processing visuospatial information, regardless of where that information originates.

Feedback scales the spatial tuning of cortical responses during both visual working memory and long-term memory

Perception, working memory, and long-term memory each evoke neural responses in visual cortex. While previous neuroimaging research on the role of visual cortex in memory has largely emphasized similarities between perception and memory, we hypothesized that responses in visual cortex would differ depending on the origins of the inputs. Using fMRI, we quantified spatial tuning in visual cortex while participants (both sexes) viewed, maintained in working memory, or retrieved from long-term memory a peripheral target. In each condition, BOLD responses were spatially tuned and aligned with the target's polar angle in all measured visual field maps including V1. As expected given the increasing sizes of receptive fields, polar angle tuning during perception increased in width up the visual hierarchy from V1 to V2, V3, hV4, and beyond. In stark contrast, the tuned responses were broad across the visual hierarchy during long-term memory (replicating a prior result) and during working memory. This pattern is consistent with the idea that mnemonic responses in V1 stem from top-down sources, even when the stimulus was recently viewed and is held in working memory. Moreover, in long-term memory, trial-to-trial biases in these tuned responses (clockwise or counterclockwise of target), predicted matched biases in memory, suggesting that the reinstated cortical responses influence memory guided behavior. We conclude that feedback widens spatial tuning in visual cortex during memory, where earlier visual maps inherit broader tuning from later maps thereby impacting the precision of memory.

Dynamic estimation of the attentional field from visual cortical activity

Navigating around the world, we must adaptively allocate attention to our surroundings based on anticipated future stimuli and events. This allocation of spatial attention boosts visuocortical representations at attended locations and locally enhances perception. Indeed, spatial attention has often been analogized to a "spotlight" shining on the item of relevance. Although the neural underpinnings of the locus of this attentional spotlight have been relatively well studied, less is known about the size of the spotlight: to what extent can the attentional field be broadened and narrowed in accordance with behavioral demands? In this study, we developed a paradigm for dynamically estimating the locus and spread of covert spatial attention, inferred from visuocortical activity using fMRI in humans. We measured BOLD activity in response to an annulus while participants (4 female, 4 male) used covert visual attention to determine whether more numbers or letters were present in a cued region of the annulus. Importantly, the width of the cued area was systematically varied, calling for different sizes of the attentional spotlight. The deployment of attention was associated with an increase in BOLD activity in corresponding retinotopic regions of visual areas V1-V3. By modeling the visuocortical attentional modulation, we could reliably recover the cued location, as well as a broadening of the attentional enhancement with wider attentional cues. This modeling approach offers a useful window into the dynamics of attention and spatial uncertainty.

A gradual transition toward categorical representations along the visual hierarchy during working memory, but not perception

The ability to stably maintain visual information over brief delays is central to healthy cognitive functioning, as is the ability to differentiate such internal representations from external inputs. One possible way to achieve both is via multiple concurrent mnemonic representations along the visual hierarchy that differ systematically from the representations of perceptual inputs. To test this possibility, we examine orientation representations along the visual hierarchy during perception and working memory. Human participants directly viewed, or held in mind, oriented grating patterns, and the similarity between fMRI activation patterns for different orientations was calculated throughout retinotopic cortex. During direct viewing of grating stimuli, similarity was relatively evenly distributed amongst all orientations, while during working memory the similarity was higher around oblique orientations. We modeled these differences in representational geometry based on the known distribution of orientation information in the natural world: The "veridical" model uses an efficient coding framework to capture hypothesized representations during visual perception. The "categorical" model assumes that different "psychological distances" between orientations result in orientation categorization relative to cardinal axes. During direct perception, the veridical model explained the data well. During working memory, the categorical model gradually gained explanatory power over the veridical model for increasingly anterior retinotopic regions. Thus, directly viewed images are represented veridically, but once visual information is no longer tethered to the sensory world there is a gradual progression to more categorical mnemonic formats along the visual hierarchy.

The dynamic reconfiguration of the functional network during episodic memory task predicts the memory performance

Episodic memory is essential for forming and retaining personal experiences, representing a fundamental aspect of human cognition. Traditional studies of episodic memory have typically used static analysis methods, viewing the brain as an unchanging entity and overlooking its dynamic properties over time. In this study, we utilized dynamic functional connectivity analysis on fMRI data from healthy adults performing an episodic memory task. We quantified integration and recruitment metrics and examined their correlation with memory performance using Pearson correlation. During encoding, integration across the entire brain, especially within the frontoparietal subnetwork, was significantly correlated with memory performance. During retrieval, recruitment becomes significantly associated with memory performance in visual subnetwork, somatomotor subnetwork, and ventral attention subnetwork. At the nodal level, a significant negative correlation was observed between memory scores and integration of the anterior cingulate gyrus, precentral gyrus, and inferior frontal gyrus within the frontoparietal network during encoding task. During retrieval task, a significant negative correlation was found between memory scores and recruitment in the left progranular cortex and right transverse gyral ventral, whereas positive correlations were seen in the right posterior inferior temporal, left middle temporal, right frontal operculum, and left operculum nodes. Moreover, the dynamic reconfiguration of the functional network was predictive of predict memory performance, as demonstrated by a significant correlation between actual and predicted memory scores. These findings advance our understanding network mechanisms underlying memory processes and developing intervention approaches for memory-related disorders as they shed light on critical factors involved in cognitive processes and provide a deeper understanding of the underlying mechanisms driving cognitive function.

Centering cognitive neuroscience on task demands and generalization

Cognitive neuroscience seeks generalizable theories explaining the relationship between behavioral, physiological and mental states. In pursuit of such theories, we propose a theoretical and empirical framework that centers on understanding task demands and the mutual constraints they impose on behavior and neural activity. Task demands emerge from the interaction between an agent's sensory impressions, goals and behavior, which jointly shape the activity and structure of the nervous system on multiple spatiotemporal scales. Understanding this interaction requires multitask studies that vary more than one experimental component (for example, stimuli and instructions) combined with dense behavioral and neural sampling and explicit testing for generalization across tasks and data modalities. By centering task demands rather than mental processes that tasks are assumed to engage, this framework paves the way for the discovery of new generalizable concepts unconstrained by existing taxonomies, and moves cognitive neuroscience toward an action-oriented, dynamic and integrated view of the brain.

Neuronal representation of visual working memory content in the primate primary visual cortex

The human ability to perceive vivid memories as if they "float" before our eyes, even in the absence of actual visual stimuli, captivates the imagination. To determine the neural substrates underlying visual memories, we investigated the neuronal representation of working memory content in the primary visual cortex of monkeys. Our study revealed that neurons exhibit unique responses to different memory contents, using firing patterns distinct from those observed during the perception of external visual stimuli. Moreover, this neuronal representation evolves with alterations in the recalled content and extends beyond the retinotopic areas typically reserved for processing external visual input. These discoveries shed light on the visual encoding of memories and indicate avenues for understanding the remarkable power of the mind's eye.

Visual working memories are abstractions of percepts

During perception, decoding the orientation of gratings depends on complex interactions between the orientation of the grating, aperture edges, and topographic structure of the visual map. Here, we aimed to test how aperture biases described during perception affect working memory (WM) decoding. For memoranda, we used gratings multiplied by radial and angular modulators to generate orthogonal aperture biases for identical orientations. Therefore, if WM representations are simply maintained sensory representations, they would have similar aperture biases. If they are abstractions of sensory features, they would be unbiased and the modulator would have no effect on orientation decoding. Neural patterns of delay period activity while maintaining the orientation of gratings with one modulator (e.g. radial) were interchangeable with patterns while maintaining gratings with the other modulator (e.g. angular) in visual and parietal cortex, suggesting that WM representations are insensitive to aperture biases during perception. Then, we visualized memory abstractions of stimuli using models of visual field map properties. Regardless of aperture biases, WM representations of both modulated gratings were recoded into a single oriented line. These results provide strong evidence that visual WM representations are abstractions of percepts, immune to perceptual aperture biases, and compel revisions of WM theory.

Vision matters for shape representation: Evidence from sculpturing and drawing in the blind

Shape is a property that could be perceived by vision and touch, and is classically considered to be supramodal. While there is mounting evidence for the shared cognitive and neural representation space between visual and tactile shape, previous research tended to rely on dissimilarity structures between objects and had not examined the detailed properties of shape representation in the absence of vision. To address this gap, we conducted three explicit object shape knowledge production experiments with congenitally blind and sighted participants, who were asked to produce verbal features, 3D clay models, and 2D drawings of familiar objects with varying levels of tactile exposure, including tools, large nonmanipulable objects, and animals. We found that the absence of visual experience (i.e., in the blind group) led to stronger differences in animals than in tools and large objects, suggesting that direct tactile experience of objects is essential for shape representation when vision is unavailable. For tools with rich tactile/manipulation experiences, the blind produced overall good shapes comparable to the sighted, yet also showed intriguing differences. The blind group had more variations and a systematic bias in the geometric property of tools (making them stubbier than the sighted), indicating that visual experience contributes to aligning internal representations and calibrating overall object configurations, at least for tools. Taken together, the object shape representation reflects the intricate orchestration of vision, touch and language.

Inner speech and the body error theory

Inner speech is commonly understood as the conscious experience of a voice within the mind. One recurrent theme in the scientific literature is that the phenomenon involves a representation of overt speech, for example, a representation of phonetic properties that result from a copy of speech instructions that were ultimately suppressed. I propose a larger picture that involves some embodied objects and their misperception. I call it "the Body Error Theory," or BET for short. BET is a form of illusionism, but the particular version I favor is a cross-modal illusion. Newly described here, my hypothesis is that the experience of inner speech arises from a mix of interoception and audition. Specifically, there is the detection of slight but well-confirmed activities in the speech musculature that occur during inner speech, which helps to transform representations of normal but quiet nonverbal sounds that inevitably occur during inner speech, from breathing to background noise, into a mistaken perception of inner speech. Simply put, activities in the speech musculature mix with sounds to create the appearance of speech sounds, which thus explains the "voice within the mind." I also show how BET's cross-modal system fits with standard information processing accounts for speech monitoring and how it accommodates the central insights of leading theories of inner speech. In addition, I show how BET is supported by data from experience-sampling surveys and how it can be empirically tested against its rivals.

Top-down generation of low-resolution representations improves visual perception and imagination

Perception or imagination requires top-down signals from high-level cortex to primary visual cortex (V1) to reconstruct or simulate the representations bottom-up stimulated by the seen images. Interestingly, top-down signals in V1 have lower spatial resolution than bottom-up representations. It is unclear why the brain uses low-resolution signals to reconstruct or simulate high-resolution representations. By modeling the top-down pathway of the visual system using the decoder of a variational auto-encoder (VAE), we reveal that low-resolution top-down signals can better reconstruct or simulate the information contained in the sparse activities of V1 simple cells, which facilitates perception and imagination. This advantage of low-resolution generation is related to facilitating high-level cortex to form geometry-respecting representations observed in experiments. Furthermore, we present two findings regarding this phenomenon in the context of AI-generated sketches, a style of drawings made of lines. First, we found that the quality of the generated sketches critically depends on the thickness of the lines in the sketches: thin-line sketches are harder to generate than thick-line sketches. Second, we propose a technique to generate high-quality thin-line sketches: instead of directly using original thin-line sketches, we use blurred sketches to train VAE or GAN (generative adversarial network), and then infer the thin-line sketches from the VAE- or GAN-generated blurred sketches. Collectively, our work suggests that low-resolution top-down generation is a strategy the brain uses to improve visual perception and imagination, which inspires new sketch-generation AI techniques.

Visual working memories are abstractions of percepts

Pioneering studies demonstrating that the contents of visual working memory (WM) can be decoded from the patterns of multivoxel activity in early visual cortex transformed not only how we study WM, but theories of how memories are stored. For instance, the ability to decode the orientation of memorized gratings is hypothesized to depend on the recruitment of the same neural encoding machinery used for perceiving orientations. However, decoding evidence cannot be used to test the so-called sensory recruitment hypothesis without understanding the underlying nature of what is being decoded. Although unknown during WM, during perception decoding the orientation of gratings does not simply depend on activities of orientation tuned neurons. Rather, it depends on complex interactions between the orientation of the grating, the aperture edges, and the topographic structure of the visual map. Here, our goals are to 1) test how these aperture biases described during perception may affect WM decoding, and 2) leverage carefully manipulated visual stimulus properties of gratings to test how sensory-like are WM codes. For memoranda, we used gratings multiplied by radial and angular modulators to generate orthogonal aperture biases despite having identical orientations. Therefore, if WM representations are simply maintained sensory representations, they would have similar aperture biases. If they are abstractions of sensory features, they would be unbiased and the modulator would have no effect on orientation decoding. Results indicated that fMRI patterns of delay period activity while maintaining the orientation of a grating with one modulator (eg, radial) were interchangeable with patterns while maintaining a grating with the other modulator (eg, angular). We found significant cross-classification in visual and parietal cortex, suggesting that WM representations are insensitive to aperture biases during perception. Then, we visualized memory abstractions of stimuli using a population receptive field model of the visual field maps. Regardless of aperture biases, WM representations of both modulated gratings were recoded into a single oriented line. These results provide strong evidence that visual WM representations are abstractions of percepts, immune to perceptual aperture biases, and compel revisions of WM theory.

A retinotopic code structures the interaction between perception and memory systems

Conventional views of brain organization suggest that regions at the top of the cortical hierarchy processes internally oriented information using an abstract amodal neural code. Despite this, recent reports have described the presence of retinotopic coding at the cortical apex, including the default mode network. What is the functional role of retinotopic coding atop the cortical hierarchy? Here we report that retinotopic coding structures interactions between internally oriented (mnemonic) and externally oriented (perceptual) brain areas. Using functional magnetic resonance imaging, we observed robust inverted (negative) retinotopic coding in category-selective memory areas at the cortical apex, which is functionally linked to the classic (positive) retinotopic coding in category-selective perceptual areas in high-level visual cortex. These functionally linked retinotopic populations in mnemonic and perceptual areas exhibit spatially specific opponent responses during both bottom-up perception and top-down recall, suggesting that these areas are interlocked in a mutually inhibitory dynamic. These results show that retinotopic coding structures interactions between perceptual and mnemonic neural systems, providing a scaffold for their dynamic interaction.

How does the primate brain combine generative and discriminative computations in vision?

Vision is widely understood as an inference problem. However, two contrasting conceptions of the inference process have each been influential in research on biological vision as well as the engineering of machine vision. The first emphasizes bottom-up signal flow, describing vision as a largely feedforward, discriminative inference process that filters and transforms the visual information to remove irrelevant variation and represent behaviorally relevant information in a format suitable for downstream functions of cognition and behavioral control. In this conception, vision is driven by the sensory data, and perception is direct because the processing proceeds from the data to the latent variables of interest. The notion of "inference" in this conception is that of the engineering literature on neural networks, where feedforward convolutional neural networks processing images are said to perform inference. The alternative conception is that of vision as an inference process in Helmholtz's sense, where the sensory evidence is evaluated in the context of a generative model of the causal processes that give rise to it. In this conception, vision inverts a generative model through an interrogation of the sensory evidence in a process often thought to involve top-down predictions of sensory data to evaluate the likelihood of alternative hypotheses. The authors include scientists rooted in roughly equal numbers in each of the conceptions and motivated to overcome what might be a false dichotomy between them and engage the other perspective in the realm of theory and experiment. The primate brain employs an unknown algorithm that may combine the advantages of both conceptions. We explain and clarify the terminology, review the key empirical evidence, and propose an empirical research program that transcends the dichotomy and sets the stage for revealing the mysterious hybrid algorithm of primate vision.

Hippocampal mechanisms resolve competition in memory and perception

Behaving adaptively requires selection of relevant memories and sensations and suppression of competing ones. We hypothesized that these mechanisms are linked, such that hippocampal computations that resolve competition in memory also shape the precision of sensory representations to guide selective attention. We leveraged f MRI-based pattern similarity, receptive field modeling, and eye tracking to test this hypothesis in humans performing a memory-dependent visual search task. In the hippocampus, differentiation of competing memories predicted the precision of memory-guided eye movements. In visual cortex, preparatory coding of remembered target locations predicted search successes, whereas preparatory coding of competing locations predicted search failures due to interference. These effects were linked: stronger hippocampal memory differentiation was associated with lower competitor activation in visual cortex, yielding more precise preparatory representations. These results demonstrate a role for memory differentiation in shaping the precision of sensory representations, highlighting links between mechanisms that overcome competition in memory and perception.

Effect of virtual reality-based exercise and physical exercise on adolescents with overweight and obesity: study protocol for a randomised controlled trial

INTRODUCTION

Obesity is a complex and multifactorial disease that has affected many adolescents in recent decades. Clinical practice guidelines recommend exercise as the key treatment option for adolescents with overweight and obesity. However, the effects of virtual reality (VR) exercise on the physical and brain health of adolescents with overweight and obese remain unclear. This study aims to evaluate the effects of physical and VR exercises on physical and brain outcomes and explore the differences in benefits between them. Moreover, we will apply a multiomics analysis to investigate the mechanism underlying the effects of physical and VR exercises on adolescents with overweight and obesity.

METHODS AND ANALYSIS

This randomised controlled clinical trial will include 220 adolescents with overweight and obesity aged between 11 and 17 years. The participants will be randomised into five groups after screening. Participants in the exercise groups will perform an exercise programme by adding physical or VR table tennis or soccer classes to routine physical education classes in schools three times a week for 8 weeks. Participants in the control group will maintain their usual physical activity. The primary outcome will be the change in body fat mass measured using bioelectrical impedance analysis. The secondary outcomes will include changes in other physical health-related parameters, brain health-related parameters and multiomics variables.

ETHICS AND DISSEMINATION

This study was approved by the Ethics Committee of Shanghai Sixth People's Hospital and registered in the Chinese Clinical Trial Registry. Dissemination of the findings will include peer-reviewed publications, conference presentations and media releases.

TRIAL REGISTRATION NUMBER

ChiCTR2300068786.

The testing effect for visual materials depends on preexisting knowledge

Remembering facilitates future remembering. This benefit of practicing by active retrieval, as compared to more passive relearning, is known as the testing effect and is one of the most robust findings in the memory literature. It has typically been assessed using verbal materials such as word pairs, sentences, or educational texts. We here investigate if memory for visual materials equally benefits from retrieval-mediated learning. Based on cognitive and neuroscientific theories, we hypothesize that testing effects will be limited to meaningful visual images that can be related to preexisting knowledge. In a series of four experiments, we systematically varied the type of material (meaningless "squiggle" shapes vs. meaningful object images) and the format of the test used to probe memory (a visually driven alternative forced-choice test vs. a remember/know recognition test). Within each experiment, we assessed the effects of practice type (retrieval or restudy) and the delay of the final test (immediate vs. 1 week) on the resulting practice benefits. Abstract shapes never showed a significant testing benefit, irrespective of test format. Meaningful object images did benefit from testing, particularly at long delays, and with a test format probing the recollective component of recognition memory. Together, our results indicate that retrieval can facilitate the recollection of visual images when they represent meaningful semantic units. This pattern of results is predicted by cognitive and neurobiologically motivated theories proposing that retrieval's benefits emerge through spreading activation in semantic networks, producing more easily accessible and longer-lasting memory traces. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Neural population dynamics of human working memory

The activity of neurons in macaque prefrontal cortex (PFC) persists during working memory (WM) delays, providing a mechanism for memory.1,2,3,4,5,6,7,8,9,10,11 Although theory,11,12 including formal network models,13,14 assumes that WM codes are stable over time, PFC neurons exhibit dynamics inconsistent with these assumptions.15,16,17,18,19 Recently, multivariate reanalyses revealed the coexistence of both stable and dynamic WM codes in macaque PFC.20,21,22,23 Human EEG studies also suggest that WM might contain dynamics.24,25 Nonetheless, how WM dynamics vary across the cortical hierarchy and which factors drive dynamics remain unknown. To elucidate WM dynamics in humans, we decoded WM content from fMRI responses across multiple cortical visual field maps.26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48 We found coexisting stable and dynamic neural representations of WM during a memory-guided saccade task. Geometric analyses of neural subspaces revealed that early visual cortex exhibited stronger dynamics than high-level visual and frontoparietal cortex. Leveraging models of population receptive fields, we visualized and made the neural dynamics interpretable. We found that during WM delays, V1 population initially encoded a narrowly tuned bump of activation centered on the peripheral memory target. Remarkably, this bump then spread inward toward foveal locations, forming a vector along the trajectory of the forthcoming memory-guided saccade. In other words, the neural code transformed into an abstraction of the stimulus more proximal to memory-guided behavior. Therefore, theories of WM must consider both sensory features and their task-relevant abstractions because changes in the format of memoranda naturally drive neural dynamics.

Tasks and their role in visual neuroscience

Vision is widely used as a model system to gain insights into how sensory inputs are processed and interpreted by the brain. Historically, careful quantification and control of visual stimuli have served as the backbone of visual neuroscience. There has been less emphasis, however, on how an observer's task influences the processing of sensory inputs. Motivated by diverse observations of task-dependent activity in the visual system, we propose a framework for thinking about tasks, their role in sensory processing, and how we might formally incorporate tasks into our models of vision.

Category-specific and category-general neural codes of recognition memory in the ventral visual pathway

Researchers have identified category-specific brain regions, such as the fusiform face area (FFA) and parahippocampal place area (PPA) in the ventral visual pathway, which respond preferentially to one particular category of visual objects. In addition to their category-specific role in visual object identification and categorization, regions in the ventral visual pathway play critical roles in recognition memory. Nevertheless, it is not clear whether the contributions of those brain regions to recognition memory are category-specific or category-general. To address this question, the present study adopted a subsequent memory paradigm and multivariate pattern analysis (MVPA) to explore category-specific and category-general neural codes of recognition memory in the visual pathway. The results revealed that the right FFA and the bilateral PPA showed category-specific neural patterns supporting recognition memory of faces and scenes, respectively. In contrast, the lateral occipital cortex seemed to carry category-general neural codes of recognition memory. These results provide neuroimaging evidence for category-specific and category-general neural mechanisms of recognition memory in the ventral visual pathway.

Remembrance of things perceived: Adding thalamocortical function to artificial neural networks

Recent research has illuminated the complexity and importance of the thalamocortical system but it has been difficult to identify what computational functions it performs. Meanwhile, deep-learning artificial neural networks (ANNs) based on bio-inspired models of purely cortical circuits have achieved surprising success solving sophisticated cognitive problems associated historically with human intelligence. Nevertheless, the limitations and shortcomings of artificial intelligence (AI) based on such ANNs are becoming increasingly clear. This review considers how the addition of thalamocortical connectivity and its putative functions related to cortical attention might address some of those shortcomings. Such bio-inspired models are now providing both testable theories of biological cognition and improved AI technology, much of which is happening outside the usual academic venues.

Transformative neural representations support long-term episodic memory

Memory is often conceived as a dynamic process that involves substantial transformations of mental representations. However, the neural mechanisms underlying these transformations and their role in memory formation and retrieval have only started to be elucidated. Combining intracranial EEG recordings with deep neural network models, we provide a detailed picture of the representational transformations from encoding to short-term memory maintenance and long-term memory retrieval that underlie successful episodic memory. We observed substantial representational transformations during encoding. Critically, more pronounced semantic representational formats predicted better subsequent long-term memory, and this effect was mediated by more consistent item-specific representations across encoding events. The representations were further transformed right after stimulus offset, and the representations during long-term memory retrieval were more similar to those during short-term maintenance than during encoding. Our results suggest that memory representations pass through multiple stages of transformations to achieve successful long-term memory formation and recall.