Working models of working memory

Cortical-subcortical neural networks for motor learning and storing sequence memory

Motor sequence learning relies on the synergistic collaboration of multiple brain regions. However, most existing models for motor sequence learning primarily focus on functional-level analyses of sequence memory mechanisms, providing limited neurophysiological insights into how biological neural systems intrinsically encode the ordering of sequential element. Based on physiological and anatomical evidence, this study establishes a cortico-subcortical neuronal network model that differs from existing functional frameworks, emphasizing the neural mechanisms of sequence learning in the brain. The proposed model is biological plausibility and represents a potential mechanism for human sequential learning. It achieves the sequential selection and learning of elements through the cortico-basal ganglia-thalamic circuit, where the working memory function of the prefrontal cortex serves as the basis for Hebbian learning among cortical neurons, enabling the encoding of sequential order. The model successfully reproduces physiological experimental phenomena, validating its biological rationality. Furthermore, we explore the role of cholinergic interneurons in sequence learning, revealing their ability to enhance the robustness of learning. Finally, we demonstrate the model's applicability by deploying it to control a robotic arm in drawing and handwriting tasks, highlighting its adaptability to complex real-world scenarios. These biologically inspired results aim to offer a mechanistic explanation for sequence learning and memory formation in the human brain, providing valuable insights into brain-like control systems and neural networks.

Study on individual differences in visual working memory tasks based on spatiotemporal brain functional metrics and biological perspectives

Visual working memory (VWM) is a critical area of study in cognitive neuroscience, yet the neural and genetic foundations of individual differences in VWM remain unclear. This study investigates individual differences in VWM performance across four types of visual stimuli (Body, Face, Place, Tool) under 0-back and 2-back conditions by integrating gene expression data and spatiotemporal brain function metrics. First, multiple spatiotemporal brain function metrics were extracted, and Sequential Backward Selection (SBS) and Leave-One-Subject-Out Cross-Validation (LOSO-CV) linear regression were applied to predict behavioral performance under VWM conditions. Model performance was evaluated using RMSE. Next, the Working Memory Individual Differences Map (WMIDM) was constructed based on Pearson correlation coefficients between actual and predicted behavioral performance. Finally, WMIDM was integrated with Allen Human Brain Atlas (AHBA) gene expression data to explore its genetic underpinnings. Notably, the gene analysis is exploratory, providing a preliminary framework for future investigations into the molecular basis of working memory. The results demonstrated that under the 2 vs. 0-back condition, spatiotemporal metrics outperformed static metrics (rspa=0.40,q=8.9×10-28,RMSE=0.928 vs. rsta=0.28,q=2.7×10-14,RMSE=0.966). Brain regions contributing to the WMIDM were primarily located in the frontal lobe. Furthermore, genes associated with WMIDM were significantly enriched in pathways linked to intellectual disability and mental disorders, as well as related biological processes and cell types. This study highlights the neural and potential genetic foundations of individual differences in working memory through the lens of spatiotemporal multidimensional brain function and gene expression. These findings provide valuable insights for future neuroscience research and pave the way for personalized cognitive interventions.

Activation or blockade of prelimbic 5-HT4 receptors improves working memory in hemiparkinsonian rats

Working memory deficits commonly occur in Parkinson's disease. 5-hydroxytryptamine4 (5-HT4) receptors are widely distributed in the prelimbic cortex (PrL) and involved in cognition. Here we tested the effects of activation and blockade of PrL 5-HT4 receptors on working memories by T-maze rewarded alternation and Morris water maze tests in rats with unilateral 6-hydroxydopamine (6-OHDA) lesion of the medial forebrain bundle. The lesion induced working memory deficits, decreased dopamine levels in the limbic-related brain regions, changed normalized δ, high θ, α, β, low and high γ power of the PrL, and upregulated expression of PrL 5-HT4 receptor. Intra-PrL injection of 5-HT4 receptor agonist BIMU8 or antagonist GR113808 did not impact working memories in sham rats, but improved working memory deficits in the lesioned rats. Intra-PrL injection of BIMU8 or GR113808 had no effect on monoamine levels in the limbic-related brain regions or normalized low and high γ power of the PrL in sham rats. However, in the lesioned rats, intra-PrL injection of BIMU8 significantly increased dopamine and 5-HT levels in the medial prefrontal cortex, amygdala and dorsal hippocampus, while intra-PrL injection of GR113808 significantly increased dopamine levels in these brain regions and increased normalized low and high γ power of the PrL. These results suggest that 6-OHDA lesion in rats induces working memory deficits, while activation or blockade of PrL 5-HT4 receptors improves the deficits in the lesioned rats, which possibly due to the changes of monoamine levels in the limbic-related brain regions and network activity of neurons in the PrL.

Relationships and representations of brain structures, connectivity, dynamics and functions

The review explores the complex interplay between brain structures and their associated functions, presenting a diversity of hierarchical models that enhances our understanding of these relationships. Central to this approach are structure-function flow diagrams, which offer a visual representation of how specific neuroanatomical structures are linked to their functional roles. These diagrams are instrumental in mapping the intricate connections between different brain regions, providing a clearer understanding of how functions emerge from the underlying neural architecture. The study details innovative attempts to develop new functional hierarchies that integrate structural and functional data. These efforts leverage recent advancements in neuroimaging techniques such as fMRI, EEG, MEG, and PET, as well as computational models that simulate neural dynamics. By combining these approaches, the study seeks to create a more refined and dynamic hierarchy that can accommodate the brain's complexity, including its capacity for plasticity and adaptation. A significant focus is placed on the overlap of structures and functions within the brain. The manuscript acknowledges that many brain regions are multifunctional, contributing to different cognitive and behavioral processes depending on the context. This overlap highlights the need for a flexible, non-linear hierarchy that can capture the brain's intricate functional landscape. Moreover, the study examines the interdependence of these functions, emphasizing how the loss or impairment of one function can impact others. Another crucial aspect discussed is the brain's ability to compensate for functional deficits following neurological diseases or injuries. The investigation explores how the brain reorganizes itself, often through the recruitment of alternative neural pathways or the enhancement of existing ones, to maintain functionality despite structural damage. This compensatory mechanism underscores the brain's remarkable plasticity, demonstrating its ability to adapt and reconfigure itself in response to injury, thereby ensuring the continuation of essential functions. In conclusion, the study presents a system of brain functions that integrates structural, functional, and dynamic perspectives. It offers a robust framework for understanding how the brain's complex network of structures supports a wide range of cognitive and behavioral functions, with significant implications for both basic neuroscience and clinical applications.

Longitudinal excitation-inhibition balance altered by sex and APOE-ε4

Neuronal hyperexcitation affects memory and neural processing across the Alzheimer's disease (AD) cognitive continuum. Levetiracetam, an antiepileptic, shows promise in improving cognitive impairment by restoring the neural excitation/inhibition balance in AD patients. We previously identified a hyper-excitable phenotype in cognitively unimpaired female APOE-ε4 carriers relative to male counterparts cross-sectionally. This sex difference lacks longitudinal validation; however, clarifying the vulnerability of female ε4-carriers could better inform antiepileptic treatment efficacy. Here, we investigated this sex-by-ε4 interaction using a longitudinal design. We used resting-state fMRI and diffusion tensor imaging collected longitudinally from 106 participants who were cognitively unimpaired for at least one scan event but may have been assessed to have clinical dementia ratings corresponding to early mild cognitive impairment over time. By including scan events where participants transitioned to mild cognitive impairment, we modeled the trajectory of the whole-brain excitation-inhibition ratio throughout the preclinical cognitively healthy continuum and extended to early impairment. A linear mixed model revealed a significant three-way interaction among sex, ε4-status, and time, with female ε4-carriers showing a significant hyper-excitable trajectory. These findings suggest a possible pathway for preventative therapy targeting preclinical hyperexcitation in female ε4-carriers.

Reactivating and reorganizing activity-silent working memory: two distinct mechanisms underlying pinging the brain

Recent studies have proposed that visual information in working memory (WM) can be maintained in an activity-silent state and reactivated by task-irrelevant high-contrast visual impulses ("ping"). Although pinging the brain has become a popular tool for exploring activity-silent WM, its underlying mechanisms remain unclear. In the current study, we directly compared the neural reactivation effects and behavioral consequences of spatial-nonmatching and spatial-matching pings to distinguish the noise-reduction and target-interaction hypotheses of pinging the brain. Initially, in an electroencephalogram study, our neural decoding results showed that spatial-nonmatching pings reactivated activity-silent WM transiently without changing the original WM representations or recall performance. Conversely, spatial-matching pings reactivated activity-silent WM more durably and further reorganized WM information by decreasing neural representations' dynamics. Notably, only the reactivation strength of spatial-matching pings correlated with recall performance and was modulated by the location of memorized items, with neural reactivation occurring only when both items and pings were presented horizontally. Consistently, in a follow-up behavioral study, we found that only spatial-matching, horizontal pings impaired recall performance compared to no ping. Together, our results demonstrated two distinct mechanisms underlying pinging the brain, highlighting the critical role of the ping's context (i.e. spatial information) in reactivating and reorganizing activity-silent WM.

Prefrontal working memory activity slots support sequence memory similar to hippocampal long-term memory position recall

Prefrontal cortex and medial temporal lobe information processing might not be that different after all. In this issue of Neuron, Whittington et al.1 show that prefrontal cortex working memory slot activity enables sequence memorizing similar to hippocampal long-term memory. Here, this approach is outlined and its implications are discussed.

Cognitive Control

Humans and other primates have a remarkable ability to perform a wide range of tasks and behaviors, even novel ones, in order to achieve their goals. Further, they are able to shift flexibly among these behaviors as the contexts demand. Cognitive control is the function at the base of this remarkable behavioral generativity and flexibility. The present review provides a survey of current research on cognitive control focusing on two of its primary features within a control systems framework: (a) the ability to select new behaviors based on context and (b) the ability to monitor ongoing behavior and adjust accordingly. Throughout, the review places an emphasis on how differences in the content and structure of task representations affect these core features of cognitive control.

Robust working memory in a two-dimensional continuous attractor network

Continuous bump attractor networks (CANs) have been widely used in the past to explain the phenomenology of working memory (WM) tasks in which continuous-valued information has to be maintained to guide future behavior. Standard CAN models suffer from two major limitations: the stereotyped shape of the bump attractor does not reflect differences in the representational quality of WM items and the recurrent connections within the network require a biologically unrealistic level of fine tuning. We address both challenges in a two-dimensional (2D) network model formalized by two coupled neural field equations of Amari type. It combines the lateral-inhibition-type connectivity of classical CANs with a locally balanced excitatory and inhibitory feedback loop. We first use a radially symmetric connectivity to analyze the existence, stability and bifurcation structure of 2D bumps representing the conjunctive WM of two input dimensions. To address the quality of WM content, we show in model simulations that the bump amplitude reflects the temporal integration of bottom-up and top-down evidence for a specific combination of input features. This includes the network capacity to transform a stable subthreshold memory trace of a weak input into a high fidelity memory representation by an unspecific cue given retrospectively during WM maintenance. To address the fine-tuning problem, we test numerically different perturbations of the assumed radial symmetry of the connectivity function including random spatial fluctuations in the connection strength. Different to the behavior of standard CAN models, the bump does not drift in representational space but remains stationary at the input position.

Human stem cell-based models to study synaptic dysfunction and cognition in schizophrenia: A narrative review

Cognitive impairment is the strongest predictor of functional outcomes in schizophrenia and is hypothesized to result from synaptic dysfunction. However, targeting synaptic plasticity and cognitive deficits in patients remains a significant clinical challenge. A comprehensive understanding of synaptic plasticity and the molecular basis of learning and memory in a disease context can provide specific targets for the development of novel therapeutics targeting cognitive impairments in schizophrenia. Here, we describe the role of synaptic plasticity in cognition, summarize evidence for synaptic dysfunction in schizophrenia and demonstrate the use of patient derived induced-pluripotent stem cells for studying synaptic plasticity in vitro. Lastly, we discuss current advances and future technologies for bridging basic science research of synaptic dysfunction with clinical and translational research that can be used to predict treatment response and develop novel therapeutics.

Maintenance and transformation of representational formats during working memory prioritization

Visual working memory depends on both material-specific brain areas in the ventral visual stream (VVS) that support the maintenance of stimulus representations and on regions in the prefrontal cortex (PFC) that control these representations. How executive control prioritizes working memory contents and whether this affects their representational formats remains an open question, however. Here, we analyzed intracranial EEG (iEEG) recordings in epilepsy patients with electrodes in VVS and PFC who performed a multi-item working memory task involving a retro-cue. We employed Representational Similarity Analysis (RSA) with various Deep Neural Network (DNN) architectures to investigate the representational format of prioritized VWM content. While recurrent DNN representations matched PFC representations in the beta band (15-29 Hz) following the retro-cue, they corresponded to VVS representations in a lower frequency range (3-14 Hz) towards the end of the maintenance period. Our findings highlight the distinct coding schemes and representational formats of prioritized content in VVS and PFC.

Single-pulse Transcranial Magnetic Stimulation Affects Working-memory Performance via Posterior Beta-band Oscillations

A single pulse of TMS (spTMS) during the delay period of a double serial retrocuing working-memory task can briefly rescue decodability of an unprioritized memory item (UMI). This physiological phenomenon, which is paralleled in behavior by involuntary retrieval of the UMI, is carried by the beta frequency band, implicating beta-band dynamics in priority coding in working memory. We decomposed EEG data from 12 participants performing double serial retrocuing with concurrent delivery of spTMS using Spatially distributed PhAse Coupling Extraction. This procedure decomposes the scalp-level signal into a set of discrete coupled oscillators, each with a component strength that can vary over time. The decomposition revealed a diversity of low-frequency components, a subset of them strengthening with the onset of the task, and the majority declining in strength across the trial, as well as within each delay period. Results with spTMS revealed no evidence that it works by activating previously "silent" sources; instead, it had the effect of modulating ongoing activity, specifically by exaggerating the within-delay decrease in strength of posterior beta components. Furthermore, the magnitude of the effect of spTMS on the loading strength of a posterior beta component correlated with the disruptive effect of spTMS on performance, a pattern also seen when analyses were restricted to trials with "UMI-lure" memory probes. Rather than reflecting the "activation" of a putatively "activity silent" UMI, these results implicate beta-band dynamics in a mechanism that distinguishes prioritized from unprioritized, and suggest that the effect of spTMS is to disrupt this code.

Task-specific topology of brain networks supporting working memory and inhibition

Network neuroscience explores the brain's connectome, demonstrating that dynamic neural networks support cognitive functions. This study investigates how distinct cognitive abilities-working memory and cognitive inhibitory control-are supported by unique brain network configurations constructed by estimating whole-brain networks using mutual information. The study involved 195 participants who completed the Sternberg Item Recognition task and Flanker tasks while undergoing electroencephalography recording. A mixed-effects linear model analyzed the influence of network metrics on cognitive performance, considering individual differences and task-specific dynamics. The findings indicate that working memory and cognitive inhibitory control are associated with different network attributes, with working memory relying on distributed networks and cognitive inhibitory control on more segregated ones. Our analysis suggests that both strong and weak connections contribute to cognitive processes, with weak connections potentially leading to a more stable and support networks of memory and cognitive inhibitory control. The findings indirectly support the network neuroscience theory of intelligence, suggesting different functional topology of networks inherent to various cognitive functions. Nevertheless, we propose that understanding individual variations in cognitive abilities requires recognizing both shared and unique processes within the brain's network dynamics.

Bridging flexible goal-directed cognition and consciousness: The Goal-Aligning Representation Internal Manipulation theory

Goal-directed manipulation of internal representations is a key element of human flexible behaviour, while consciousness is commonly associated with higher-order cognition and human flexibility. Current perspectives have only partially linked these processes, thus preventing a clear understanding of how they jointly generate flexible cognition and behaviour. Moreover, these limitations prevent an effective exploitation of this knowledge for technological scopes. We propose a new theoretical perspective that extends our 'three-component theory of flexible cognition' toward higher-order cognition and consciousness, based on the systematic integration of key concepts from Cognitive Neuroscience and AI/Robotics. The theory proposes that the function of conscious processes is to support the alignment of representations with multi-level goals. This higher alignment leads to more flexible and effective behaviours. We analyse here our previous model of goal-directed flexible cognition (validated with more than 20 human populations) as a starting GARIM-inspired model. By bridging the main theories of consciousness and goal-directed behaviour, the theory has relevant implications for scientific and technological fields. In particular, it contributes to developing new experimental tasks and interpreting clinical evidence. Finally, it indicates directions for improving machine learning and robotics systems and for informing real-world applications (e.g., in digital-twin healthcare and roboethics).

Representation and computation in visual working memory

The ability to sustain internal representations of the sensory environment beyond immediate perception is a fundamental requirement of cognitive processing. In recent years, debates regarding the capacity and fidelity of the working memory (WM) system have advanced our understanding of the nature of these representations. In particular, there is growing recognition that WM representations are not merely imperfect copies of a perceived object or event. New experimental tools have revealed that observers possess richer information about the uncertainty in their memories and take advantage of environmental regularities to use limited memory resources optimally. Meanwhile, computational models of visuospatial WM formulated at different levels of implementation have converged on common principles relating capacity to variability and uncertainty. Here we review recent research on human WM from a computational perspective, including the neural mechanisms that support it.

Unified control of temporal and spatial scales of sensorimotor behavior through neuromodulation of short-term synaptic plasticity

Neuromodulators have been shown to alter the temporal profile of short-term synaptic plasticity (STP); however, the computational function of this neuromodulation remains unexplored. Here, we propose that the neuromodulation of STP provides a general mechanism to scale neural dynamics and motor outputs in time and space. We trained recurrent neural networks that incorporated STP to produce complex motor trajectories-handwritten digits-with different temporal (speed) and spatial (size) scales. Neuromodulation of STP produced temporal and spatial scaling of the learned dynamics and enhanced temporal or spatial generalization compared to standard training of the synaptic weights in the absence of STP. The model also accounted for the results of two experimental studies involving flexible sensorimotor timing. Neuromodulation of STP provides a unified and biologically plausible mechanism to control the temporal and spatial scales of neural dynamics and sensorimotor behaviors.

Unifying network model links recency and central tendency biases in working memory

The central tendency bias, or contraction bias, is a phenomenon where the judgment of the magnitude of items held in working memory appears to be biased toward the average of past observations. It is assumed to be an optimal strategy by the brain and commonly thought of as an expression of the brain's ability to learn the statistical structure of sensory input. On the other hand, recency biases such as serial dependence are also commonly observed and are thought to reflect the content of working memory. Recent results from an auditory delayed comparison task in rats suggest that both biases may be more related than previously thought: when the posterior parietal cortex (PPC) was silenced, both short-term and contraction biases were reduced. By proposing a model of the circuit that may be involved in generating the behavior, we show that a volatile working memory content susceptible to shifting to the past sensory experience - producing short-term sensory history biases - naturally leads to contraction bias. The errors, occurring at the level of individual trials, are sampled from the full distribution of the stimuli and are not due to a gradual shift of the memory toward the sensory distribution's mean. Our results are consistent with a broad set of behavioral findings and provide predictions of performance across different stimulus distributions and timings, delay intervals, as well as neuronal dynamics in putative working memory areas. Finally, we validate our model by performing a set of human psychophysics experiments of an auditory parametric working memory task.

Complex Regulation of Protocadherin Epigenetics on Aging-Related Brain Health

Life expectancy continues to increase in the high-income world due to advances in medical care; however, quality of life declines with increasing age due to normal aging processes. Current research suggests that various aspects of aging are genetically modulated and thus may be slowed via genetic modification. Here, we show evidence for epigenetic modulation of the aging process in the brain from over 1800 individuals as part of the Framingham Heart Study. We investigated the methylation of genes in the protocadherin (PCDH) clusters, including the alpha (PCHDA), beta (PCDHB), and gamma (PCDHG) clusters. Reduced PCDHG, elevated PCDHA, and elevated PCDHB methylation levels were associated with substantial reductions in the rate of decline of regional white matter volume as well as certain cognitive skills, independent of overall accelerated or retarded aging as estimated by a DNA clock. These results are likely due to the different effects of the expression of genes in the alpha, beta, and gamma PCHD clusters and suggest that experience-based aging processes related to a decline in regional brain volume and select cognitive skills may be slowed via targeted epigenetic modifications.

Building a realistic, scalable memory model with independent engrams using a homeostatic mechanism

Memory formation is usually associated with Hebbian learning and synaptic plasticity, which changes the synaptic strengths but omits structural changes. A recent study suggests that structural plasticity can also lead to silent memory engrams, reproducing a conditioned learning paradigm with neuron ensembles. However, this study is limited by its way of synapse formation, enabling the formation of only one memory engram. Overcoming this, our model allows the formation of many engrams simultaneously while retaining high neurophysiological accuracy, e.g., as found in cortical columns. We achieve this by substituting the random synapse formation with the Model of Structural Plasticity. As a homeostatic model, neurons regulate their activity by growing and pruning synaptic elements based on their current activity. Utilizing synapse formation based on the Euclidean distance between the neurons with a scalable algorithm allows us to easily simulate 4 million neurons with 343 memory engrams. These engrams do not interfere with one another by default, yet we can change the simulation parameters to form long-reaching associations. Our model's analysis shows that homeostatic engram formation requires a certain spatiotemporal order of events. It predicts that synaptic pruning precedes and enables synaptic engram formation and that it does not occur as a mere compensatory response to enduring synapse potentiation as in Hebbian plasticity with synaptic scaling. Our model paves the way for simulations addressing further inquiries, ranging from memory chains and hierarchies to complex memory systems comprising areas with different learning mechanisms.

Maintenance of color memoranda in activity-quiescent working memory states: Evidence from impulse perturbation

In the present study, we used an impulse perturbation method to probe working memory maintenance of colors in neurally active and activity-quiescent states, focusing on a set of pre-registered analyses. We analyzed the electroencephalograph (EEG) data of 30 participants who completed a delayed match-to-sample working memory task, in which one of the two items that were presented was retro-cued as task relevant. The analyses revealed that both cued and uncued colors were decodable from impulse-evoked activity, the latter in contrast to previous reports of working memory for orientation gratings. Decoding of colors from oscillations in the alpha band showed that cued items could be decoded therein whereas uncued items could not. Overall, the outcomes suggest that subtle differences exist between the representation of colors, and that of stimuli with spatial properties, but the present results also demonstrate that regardless of their specific neural state, both are accessible through visual impulse perturbation.

Neurobiological Causal Models of Language Processing

The language faculty is physically realized in the neurobiological infrastructure of the human brain. Despite significant efforts, an integrated understanding of this system remains a formidable challenge. What is missing from most theoretical accounts is a specification of the neural mechanisms that implement language function. Computational models that have been put forward generally lack an explicit neurobiological foundation. We propose a neurobiologically informed causal modeling approach which offers a framework for how to bridge this gap. A neurobiological causal model is a mechanistic description of language processing that is grounded in, and constrained by, the characteristics of the neurobiological substrate. It intends to model the generators of language behavior at the level of implementational causality. We describe key features and neurobiological component parts from which causal models can be built and provide guidelines on how to implement them in model simulations. Then we outline how this approach can shed new light on the core computational machinery for language, the long-term storage of words in the mental lexicon and combinatorial processing in sentence comprehension. In contrast to cognitive theories of behavior, causal models are formulated in the "machine language" of neurobiology which is universal to human cognition. We argue that neurobiological causal modeling should be pursued in addition to existing approaches. Eventually, this approach will allow us to develop an explicit computational neurobiology of language.

Aperiodic EEG and 7T MRSI evidence for maturation of E/I balance supporting the development of working memory through adolescence

Adolescence has been hypothesized to be a critical period for the development of human association cortex and higher-order cognition. A defining feature of critical period development is a shift in the excitation: inhibition (E/I) balance of neural circuitry, however how changes in E/I may enhance cortical circuit function to support maturational improvements in cognitive capacities is not known. Harnessing ultra-high field 7 T MR spectroscopy and EEG in a large, longitudinal cohort of youth (N = 164, ages 10-32 years old, 347 neuroimaging sessions), we delineate biologically specific associations between age-related changes in excitatory glutamate and inhibitory GABA neurotransmitters and EEG-derived measures of aperiodic neural activity reflective of E/I balance in prefrontal association cortex. Specifically, we find that developmental increases in E/I balance reflected in glutamate:GABA balance are linked to changes in E/I balance assessed by the suppression of prefrontal aperiodic activity, which in turn facilitates robust improvements in working memory. These findings indicate a role for E/I-engendered changes in prefrontal signaling mechanisms in the maturation of cognitive maintenance. More broadly, this multi-modal imaging study provides evidence that human association cortex undergoes physiological changes consistent with critical period plasticity during adolescence.

The "psychiatric" neuron: the psychic neuron of the cerebral cortex, revisited

Nearly 25 years ago, Dr. Patricia Goldman-Rakic published her review paper, "The 'Psychic' Neuron of the Cerebral Cortex," outlining the circuit-level dynamics, neurotransmitter systems, and behavioral correlates of pyramidal neurons in the cerebral cortex, particularly as they relate to working memory. In the decades since the release of this paper, the existing literature and our understanding of the pyramidal neuron have increased tremendously, and research is still underway to better characterize the role of the pyramidal neuron in both healthy and psychiatric disease states. In this review, we revisit Dr. Goldman-Rakic's characterization of the pyramidal neuron, focusing on the pyramidal neurons of the prefrontal cortex (PFC) and their role in working memory. Specifically, we examine the role of PFC pyramidal neurons in the intersection of working memory and social function and describe how deficits in working memory may actually underlie the pathophysiology of social dysfunction in psychiatric disease states. We briefly describe the cortico-cortical and corticothalamic connections between the PFC and non-PFC brain regions, as well the microcircuit dynamics of the pyramidal neuron and interneurons, and the role of both these macro- and microcircuits in the maintenance of the excitatory/inhibitory balance of the cerebral cortex for working memory function. Finally, we discuss the consequences to working memory when pyramidal neurons and their circuits are dysfunctional, emphasizing the resulting social deficits in psychiatric disease states with known working memory dysfunction.

Sources of richness and ineffability for phenomenally conscious states

Conscious states-state that there is something it is like to be in-seem both rich or full of detail and ineffable or hard to fully describe or recall. The problem of ineffability, in particular, is a longstanding issue in philosophy that partly motivates the explanatory gap: the belief that consciousness cannot be reduced to underlying physical processes. Here, we provide an information theoretic dynamical systems perspective on the richness and ineffability of consciousness. In our framework, the richness of conscious experience corresponds to the amount of information in a conscious state and ineffability corresponds to the amount of information lost at different stages of processing. We describe how attractor dynamics in working memory would induce impoverished recollections of our original experiences, how the discrete symbolic nature of language is insufficient for describing the rich and high-dimensional structure of experiences, and how similarity in the cognitive function of two individuals relates to improved communicability of their experiences to each other. While our model may not settle all questions relating to the explanatory gap, it makes progress toward a fully physicalist explanation of the richness and ineffability of conscious experience-two important aspects that seem to be part of what makes qualitative character so puzzling.

Inferring neural dynamics of memory during naturalistic social communication

Memory processes in complex behaviors like social communication require forming representations of the past that grow with time. The neural mechanisms that support such continually growing memory remain unknown. We address this gap in the context of fly courtship, a natural social behavior involving the production and perception of long, complex song sequences. To study female memory for male song history in unrestrained courtship, we present 'Natural Continuation' (NC)-a general, simulation-based model comparison procedure to evaluate candidate neural codes for complex stimuli using naturalistic behavioral data. Applying NC to fly courtship revealed strong evidence for an adaptive population mechanism for how female auditory neural dynamics could convert long song histories into a rich mnemonic format. Song temporal patterning is continually transformed by heterogeneous nonlinear adaptation dynamics, then integrated into persistent activity, enabling common neural mechanisms to retain continuously unfolding information over long periods and yielding state-of-the-art predictions of female courtship behavior. At a population level this coding model produces multi-dimensional advection-diffusion-like responses that separate songs over a continuum of timescales and can be linearly transformed into flexible output signals, illustrating its potential to create a generic, scalable mnemonic format for extended input signals poised to drive complex behavioral responses. This work thus shows how naturalistic behavior can directly inform neural population coding models, revealing here a novel process for memory formation.

The GluN2C/D-specific positive allosteric modulator CIQ rescues delay-induced working memory deficits in mice

Working memory is of short duration and is, therefore, particularly sensitive to time delays. Moreover, NMDA receptors are significantly involved in working memory. In the present study, we tested whether a commonly used measure of working memory, spontaneous alternation in the Y-maze, is sensitive to time delays and, if so, whether impairments due to time-delay can be rescued by treatment with CIQ, a positive allosteric modulator of the GluN2C/D subunits of NMDA receptor. Our results indicate that the effects of time delay do depend on the performance of the individual mice under basal condition. Those mice that performed well under basal conditions showed impaired spontaneous alternations when tested with a 45-s delay. Treatment with CIQ resulted in an improvement of spontaneous alternations, regardless of delay, sex, or basal performance. On the one hand, our study shows that repeated measures of individual behavior can better control the effects of confounding factors such as time delays. On the other hand, our study also highlights the potential of GluN2C/D-specific positive allosteric modulators in the treatment of human disorders associated with working memory deficits, such as schizophrenia.

Competitive learning to generate sparse representations for associative memory

One of the most well established brain principles, Hebbian learning, has led to the theoretical concept of neural assemblies. Based on it, many interesting brain theories have spawned. Palm's work implements this concept through multiple binary Willshaw associative memories, in a model that not only has a wide cognitive explanatory power but also makes neuroscientific predictions. Yet, Willshaw's associative memory can only achieve top capacity when the stored vectors are extremely sparse (number of active bits can grow logarithmically with the vector's length). This strict requirement makes it difficult to apply any model that uses this associative memory, like Palm's, to real data. Hence the fact that most works apply the memory to optimal randomly generated codes that do not represent any information. This issue creates the need for encoders that can take real data, and produce sparse representations - a problem which is also raised following Barlow's efficient coding principle. In this work, we propose a biologically-constrained network that encodes images into codes that are suitable for Willshaw's associative memory. The network is organized into groups of neurons that specialize on local receptive fields, and learn through a competitive scheme. After conducting auto- and hetero-association experiments on two visual data sets, we can conclude that our network not only beats sparse coding baselines, but also that it comes close to the performance achieved using optimal random codes.

Natural brain state change with E/I balance shifting toward inhibition is associated with vigilance impairment

The delicate balance between cortical excitation and inhibition (E/I) plays a pivotal role in brain state changes. While previous studies have associated cortical hyperexcitability with brain state changes induced by sleep deprivation, whether cortical hypoexcitability is also linked to brain state changes and, if so, how it could affect cognitive performance remain unknown. Here, we address these questions by examining the brain state change occurring after meals, i.e., postprandial somnolence, and comparing it with that induced by sleep deprivation. By analyzing features representing network excitability based on electroencephalogram (EEG) signals, we confirmed cortical hyperexcitability under sleep deprivation but revealed hypoexcitability under postprandial somnolence. In addition, we found that both sleep deprivation and postprandial somnolence adversely affected the level of vigilance. These results indicate that cortical E/I balance toward inhibition is associated with brain state changes, and deviation from the balanced state, regardless of its direction, could impair cognitive performance.

Artificial Neural Network Model with Astrocyte-Driven Short-Term Memory

In this study, we introduce an innovative hybrid artificial neural network model incorporating astrocyte-driven short-term memory. The model combines a convolutional neural network with dynamic models of short-term synaptic plasticity and astrocytic modulation of synaptic transmission. The model's performance was evaluated using simulated data from visual change detection experiments conducted on mice. Comparisons were made between the proposed model, a recurrent neural network simulating short-term memory based on sustained neural activity, and a feedforward neural network with short-term synaptic depression (STPNet) trained to achieve the same performance level as the mice. The results revealed that incorporating astrocytic modulation of synaptic transmission enhanced the model's performance.

An oscillatory mechanism for multi-level storage in short-term memory

Oscillatory activity is commonly observed during the maintenance of information in short-term memory, but its role remains unclear. Non-oscillatory models of short-term memory storage are able to encode stimulus identity through their spatial patterns of activity, but are typically limited to either an all-or-none representation of stimulus amplitude or exhibit a biologically implausible exact-tuning condition. Here we demonstrate a simple mechanism by which oscillatory input enables a circuit to generate persistent or sequential activity that encodes information not only in the spatial pattern of activity, but also in the amplitude of activity. This is accomplished through a phase-locking phenomenon that permits many different amplitudes of persistent activity to be stored without requiring exact tuning of model parameters. Altogether, this work proposes a class of models for the storage of information in working memory, a potential role for brain oscillations, and a dynamical mechanism for maintaining multi-stable neural representations.

Pinging the brain to reveal the hidden attentional priority map using encephalography

Attention has been usefully thought of as organized in priority maps - putative maps of space where attentional priority is weighted across spatial regions in a winner-take-all competition for attentional deployment. Recent work has highlighted the influence of past experiences on the weighting of spatial priority - called selection history. Aside from being distinct from more well-studied, top-down forms of attentional enhancement, little is known about the neural substrates of history-mediated attentional priority. Using a task known to induce statistical learning of target distributions, in an EEG study we demonstrate that this otherwise invisible, latent attentional priority map can be visualized during the intertrial period using a 'pinging' technique in conjunction with multivariate pattern analyses. Our findings not only offer a method of visualizing the history-mediated attentional priority map, but also shed light on the underlying mechanisms allowing our past experiences to influence future behavior.

Neuropsychological Function in Traumatic Brain Injury and the Influence of Chronic Pain

Cognitive dysfunction, pain, and psychological morbidity all present unique challenges to those living with traumatic brain injury (TBI). In this study we examined (a) the impact of pain across domains of attention, memory, and executive function, and (b) the relationships between pain and depression, anxiety, and post-traumatic stress disorder (PTSD) in persons with chronic TBI. Our sample included 86 participants with a TBI and chronic pain (n = 26), patients with TBI and no chronic pain (n = 23), and a pain-free control group without TBI (n = 37). Participants visited the laboratory and completed a comprehensive battery of neuropsychological tests as part of a structured interview. Multivariate analysis of covariance using education as a covariate, failed to detect a significant group difference for neuropsychological composite scores of attention, memory, and executive function (p = .165). A follow-up analysis using multiple one-way analysis of variance (ANOVA) was conducted for individual measures of executive function. Post-hoc testing indicated that those in both TBI groups preformed significantly worse on measures of semantic fluency when compared to controls (p < 0.001, ηρ2 = .16). Additionally, multiple ANOVAs indicated that those with TBI and pain scored significantly worse across all psychological assessments (p < .001). We also found significant associations between measures of pain and most psychological symptoms. A follow-up stepwise linear regression among those in the TBI pain group indicated that post concussive complaints, pain severity, and neuropathic pain symptoms differentially contributed to symptoms of depression, anxiety, and PTSD. These findings suggest deficits in verbal fluency among those living with chronic TBI, with results also reinforcing the multidimensional nature of pain and its psychological significance in this population.

Rumination burdens the updating of working memory

Working memory is a vital, but capacity-limited, cognitive instrument that requires frequent updating as our goals and environment change. Individuals diagnosed with depression have a reduced capacity compared with the general population, as they have a propensity to fixate on negative information, even when it is not relevant for the task at hand. Here we investigated how characteristics of psychiatric illnesses, such as rumination, affect a person's ability to efficiently update emotional information in mind. We used both neutral and negative pictures of scenes in a working memory updating task that required participants to occasionally replace items held in mind during a brief delay period. Participants were presented with a probe item at the end of each trial and asked to report whether that item was in their current memory set. Responses were slowest and least accurate for images that had been replaced (i.e., "lures"), indicating there was some difficulty in successfully updating working memory in this paradigm. Participants who have both a high propensity to ruminate and a low working memory capacity were significantly more likely to false alarm to these lures. While emotional valence did not impact accuracy for these participants, their false alarms were faster for negative stimuli compared with neutral stimuli, indicating that task-irrelevant emotional information was more difficult to remove from working memory. These results demonstrate how rumination impairs goal-directed behavior by obscuring the boundary between relevant and irrelevant information in working memory.

An Evaluation of the Structure of Attention in Adolescence

This study evaluated the factor structure of attention, considering internal and external attention, and juxtaposed alongside processing speed (PS) and working memory (WM). We expected the hypothesized model to fit better than unitary or method factors. We included 27 measures with 212 Hispanic middle schoolers from Spanish-speaking backgrounds, where a substantial proportion were at risk for learning difficulties. Confirmatory factor analytic models separated factors of PS and WM, but the final model did not align with theoretical predictions; rather only measurement factors emerged. Findings extend and refine our understanding of the structure of attention in adolescents.

Impulse perturbation reveals cross-modal access to sensory working memory through learned associations

We investigated if learned associations between visual and auditory stimuli can afford full cross-modal access to working memory. Previous research using the impulse perturbation technique has shown that cross-modal access to working memory is one-sided; visual impulses reveal both auditory and visual memoranda, but auditory impulses do not seem to reveal visual memoranda (Wolff et al., 2020b). Our participants first learned to associate six auditory pure tones with six visual orientation gratings. Next, a delayed match-to-sample task for the orientations was completed, while EEG was recorded. Orientation memories were recalled either via their learned auditory counterpart, or were visually presented. We then decoded the orientation memories from the EEG responses to both auditory and visual impulses presented during the memory delay. Working memory content could always be decoded from visual impulses. Importantly, through recall of the learned associations, the auditory impulse also evoked a decodable response from the visual WM network, providing evidence for full cross-modal access. We also observed that after a brief initial dynamic period, the representational codes of the memory items generalized across time, as well as between perceptual maintenance and long-term recall conditions. Our results thus demonstrate that accessing learned associations in long-term memory provides a cross-modal pathway to working memory that seems to be based on a common coding scheme.

Predicted utility modulates working memory fidelity in the brain

The predicted utility of information stored in working memory (WM) is hypothesized to influence the strategic allocation of WM resources. Prior work has shown that when information is prioritized, it is remembered with greater precision relative to other remembered items. However, these paradigms often complicate interpretation of the effects of predicted utility on item fidelity due to a concurrent memory load. Likewise, no fMRI studies have examined whether the predicted utility of an item modulates fidelity in the neural representation of items during the memory delay without a concurrent load. In the current study, we used fMRI to investigate whether predicted utility influences fidelity of WM representations in the brain. Using a generative model multivoxel analysis approach to estimate the quality of remembered representations across predicted utility conditions, we observed that items with greater predicted utility are maintained in memory with greater fidelity, even when they are the only item being maintained. Further, we found that this pattern follows a parametric relationship where more predicted utility corresponded to greater fidelity. These precision differences could not be accounted for based on a redistribution of resources among already-remembered items. Rather, we interpret these results in terms of a gating mechanism that allows for pre-allocation of resources based on predicted value alone. This evidence supports a theoretical distinction between resource allocation that occurs as a result of load and resource pre-allocation that occurs as a result of predicted utility.

Neural population dynamics of computing with synaptic modulations

In addition to long-timescale rewiring, synapses in the brain are subject to significant modulation that occurs at faster timescales that endow the brain with additional means of processing information. Despite this, models of the brain like recurrent neural networks (RNNs) often have their weights frozen after training, relying on an internal state stored in neuronal activity to hold task-relevant information. In this work, we study the computational potential and resulting dynamics of a network that relies solely on synapse modulation during inference to process task-relevant information, the multi-plasticity network (MPN). Since the MPN has no recurrent connections, this allows us to study the computational capabilities and dynamical behavior contributed by synapses modulations alone. The generality of the MPN allows for our results to apply to synaptic modulation mechanisms ranging from short-term synaptic plasticity (STSP) to slower modulations such as spike-time dependent plasticity (STDP). We thoroughly examine the neural population dynamics of the MPN trained on integration-based tasks and compare it to known RNN dynamics, finding the two to have fundamentally different attractor structure. We find said differences in dynamics allow the MPN to outperform its RNN counterparts on several neuroscience-relevant tests. Training the MPN across a battery of neuroscience tasks, we find its computational capabilities in such settings is comparable to networks that compute with recurrent connections. Altogether, we believe this work demonstrates the computational possibilities of computing with synaptic modulations and highlights important motifs of these computations so that they can be identified in brain-like systems.

Working memory depends on the excitatory-inhibitory balance in neuron-astrocyte network

Previous studies have shown that astrocytes are involved in information processing and working memory (WM) in the central nervous system. Here, the neuron-astrocyte network model with biological properties is built to study the effects of excitatory-inhibitory balance and neural network structures on WM tasks. It is found that the performance metrics of WM tasks under the scale-free network are higher than other network structures, and the WM task can be successfully completed when the proportion of excitatory neurons in the network exceeds 30%. There exists an optimal region for the proportion of excitatory neurons and synaptic weight that the memory performance metrics of the WM tasks are higher. The multi-item WM task shows that the spatial calcium patterns for different items overlap significantly in the astrocyte network, which is consistent with the formation of cognitive memory in the brain. Moreover, complex image tasks show that cued recall can significantly reduce systematic noise and maintain the stability of the WM tasks. The results may contribute to understand the mechanisms of WM formation and provide some inspirations into the dynamic storage and recall of memory.

Robust and brain-like working memory through short-term synaptic plasticity

Working memory has long been thought to arise from sustained spiking/attractor dynamics. However, recent work has suggested that short-term synaptic plasticity (STSP) may help maintain attractor states over gaps in time with little or no spiking. To determine if STSP endows additional functional advantages, we trained artificial recurrent neural networks (RNNs) with and without STSP to perform an object working memory task. We found that RNNs with and without STSP were able to maintain memories despite distractors presented in the middle of the memory delay. However, RNNs with STSP showed activity that was similar to that seen in the cortex of a non-human primate (NHP) performing the same task. By contrast, RNNs without STSP showed activity that was less brain-like. Further, RNNs with STSP were more robust to network degradation than RNNs without STSP. These results show that STSP can not only help maintain working memories, it also makes neural networks more robust and brain-like.

Network instability dynamics drive a transient bursting period in the developing hippocampus in vivo

Spontaneous correlated activity is a universal hallmark of immature neural circuits. However, the cellular dynamics and intrinsic mechanisms underlying network burstiness in the intact developing brain are largely unknown. Here, we use two-photon Ca2+ imaging to comprehensively map the developmental trajectories of spontaneous network activity in the hippocampal area CA1 of mice in vivo. We unexpectedly find that network burstiness peaks after the developmental emergence of effective synaptic inhibition in the second postnatal week. We demonstrate that the enhanced network burstiness reflects an increased functional coupling of individual neurons to local population activity. However, pairwise neuronal correlations are low, and network bursts (NBs) recruit CA1 pyramidal cells in a virtually random manner. Using a dynamic systems modeling approach, we reconcile these experimental findings and identify network bi-stability as a potential regime underlying network burstiness at this age. Our analyses reveal an important role of synaptic input characteristics and network instability dynamics for NB generation. Collectively, our data suggest a mechanism, whereby developing CA1 performs extensive input-discrimination learning prior to the onset of environmental exploration.

Reactivating ordinal position information from auditory sequence memory in human brains

Retaining a sequence of events in their order is a core ability of many cognitive functions, such as speech recognition, movement control, and episodic memory. Although content representations have been widely studied in working memory (WM), little is known about how ordinal position information of an auditory sequence is retained in the human brain as well as its coding characteristics. In fact, there is still a lack of an efficient approach to directly accessing the stored ordinal position code during WM retention. Here, 31 participants performed an auditory sequence WM task with their brain activities recorded using electroencephalography (EEG). We developed new triggering events that could successfully reactivate neural representations of ordinal position during the delay period. Importantly, the ordinal position reactivation is further related to recognition behavior, confirming its indexing of WM storage. Furthermore, the ordinal position code displays an intriguing “stable-dynamic” format, i.e. undergoing the same dynamic neutral trajectory in the multivariate neural space during both encoding and retention (whenever reactivated). Overall, our results provide an effective approach to accessing the behaviorally-relevant ordinal position information in auditory sequence WM and reveal its new temporal characteristics.

Multiple forms of working memory emerge from synapse-astrocyte interactions in a neuron-glia network model

Persistent activity in populations of neurons, time-varying activity across a neural population, or activity-silent mechanisms carried out by hidden internal states of the neural population have been proposed as different mechanisms of working memory (WM). Whether these mechanisms could be mutually exclusive or occur in the same neuronal circuit remains, however, elusive, and so do their biophysical underpinnings. While WM is traditionally regarded to depend purely on neuronal mechanisms, cortical networks also include astrocytes that can modulate neural activity. We propose and investigate a network model that includes both neurons and glia and show that glia-synapse interactions can lead to multiple stable states of synaptic transmission. Depending on parameters, these interactions can lead in turn to distinct patterns of network activity that can serve as substrates for WM.

Taste cues elicit prolonged modulation of feeding behavior in Drosophila

Taste cues regulate immediate feeding behavior, but their ability to modulate future behavior has been less well studied. Pairing one taste with another can modulate subsequent feeding responses through associative learning, but this requires simultaneous exposure to both stimuli. We investigated whether exposure to one taste modulates future responses to other tastes even when they do not overlap in time. Using Drosophila, we found that brief exposure to sugar enhanced future feeding responses, whereas bitter exposure suppressed them. This modulation relies on neural pathways distinct from those that acutely regulate feeding or mediate learning-dependent changes. Sensory neuron activity was required not only during initial taste exposure but also afterward, suggesting that ongoing sensory activity may maintain experience-dependent changes in downstream circuits. Thus, the brain stores a memory of each taste stimulus after it disappears, enabling animals to integrate information as they sequentially sample different taste cues that signal local food quality.

Simulations of working memory spiking networks driven by short-term plasticity

Working Memory (WM) is a cognitive mechanism that enables temporary holding and manipulation of information in the human brain. This mechanism is mainly characterized by a neuronal activity during which neuron populations are able to maintain an enhanced spiking activity after being triggered by a short external cue. In this study, we implement, using the NEST simulator, a spiking neural network model in which the WM activity is sustained by a mechanism of short-term synaptic facilitation related to presynaptic calcium kinetics. The model, which is characterized by leaky integrate-and-fire neurons with exponential postsynaptic currents, is able to autonomously show an activity regime in which the memory information can be stored in a synaptic form as a result of synaptic facilitation, with spiking activity functional to facilitation maintenance. The network is able to simultaneously keep multiple memories by showing an alternated synchronous activity which preserves the synaptic facilitation within the neuron populations holding memory information. The results shown in this study confirm that a WM mechanism can be sustained by synaptic facilitation.

Priority-based transformations of stimulus representation in visual working memory

How does the brain prioritize among the contents of working memory (WM) to appropriately guide behavior? Previous work, employing inverted encoding modeling (IEM) of electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) datasets, has shown that unprioritized memory items (UMI) are actively represented in the brain, but in a "flipped", or opposite, format compared to prioritized memory items (PMI). To acquire independent evidence for such a priority-based representational transformation, and to explore underlying mechanisms, we trained recurrent neural networks (RNNs) with a long short-term memory (LSTM) architecture to perform a 2-back WM task. Visualization of LSTM hidden layer activity using Principal Component Analysis (PCA) confirmed that stimulus representations undergo a representational transformation-consistent with a flip-while transitioning from the functional status of UMI to PMI. Demixed (d)PCA of the same data identified two representational trajectories, one each within a UMI subspace and a PMI subspace, both undergoing a reversal of stimulus coding axes. dPCA of data from an EEG dataset also provided evidence for priority-based transformations of the representational code, albeit with some differences. This type of transformation could allow for retention of unprioritized information in WM while preventing it from interfering with concurrent behavior. The results from this initial exploration suggest that the algorithmic details of how this transformation is carried out by RNNs, versus by the human brain, may differ.

Biologically Plausible Training Mechanisms for Self-Supervised Learning in Deep Networks

We develop biologically plausible training mechanisms for self-supervised learning (SSL) in deep networks. Specifically, by biologically plausible training we mean (i) all updates of weights are based on current activities of pre-synaptic units and current, or activity retrieved from short term memory of post synaptic units, including at the top-most error computing layer, (ii) complex computations such as normalization, inner products and division are avoided, (iii) asymmetric connections between units, and (iv) most learning is carried out in an unsupervised manner. SSL with a contrastive loss satisfies the third condition as it does not require labeled data and it introduces robustness to observed perturbations of objects, which occur naturally as objects or observers move in 3D and with variable lighting over time. We propose a contrastive hinge based loss whose error involves simple local computations satisfying (ii), as opposed to the standard contrastive losses employed in the literature, which do not lend themselves easily to implementation in a network architecture due to complex computations involving ratios and inner products. Furthermore, we show that learning can be performed with one of two more plausible alternatives to backpropagation that satisfy conditions (i) and (ii). The first is difference target propagation (DTP), which trains network parameters using target-based local losses and employs a Hebbian learning rule, thus overcoming the biologically implausible symmetric weight problem in backpropagation. The second is layer-wise learning, where each layer is directly connected to a layer computing the loss error. The layers are either updated sequentially in a greedy fashion (GLL) or in random order (RLL), and each training stage involves a single hidden layer network. Backpropagation through one layer needed for each such network can either be altered with fixed random feedback weights (RF) or using updated random feedback weights (URF) as in Amity's study 2019. Both methods represent alternatives to the symmetric weight issue of backpropagation. By training convolutional neural networks (CNNs) with SSL and DTP, GLL or RLL, we find that our proposed framework achieves comparable performance to standard BP learning downstream linear classifier evaluation of the learned embeddings.

Alterations in Neural Networks During Working Memory Encoding Related to Cognitive Impairment in Temporal Lobe Epilepsy

Objective: The aim of the current study was to investigate the alterations in the neural networks of patients with temporal lobe epilepsy (TLE) during working memory (WM) encoding.

Methods: Patients with TLE (n = 52) and healthy volunteers (n = 35) completed a WM task, during which 34-channel electroencephalogram signals were recorded. The neural networks during WM encoding were calculated in TLE patients with (TLE-WM) and without (TLE-N) WM deficits.

Results: Functional connectivity strength decreased, and the theta network was altered in the TLE-WM group, although no significant differences in clinical features were observed between the TLE-N and TLE-WM groups.

Conclusions: Not all patients with TLE present with cognitive impairments and alterations in the theta network were identified in TLE patients with functional cognitive deficits.

Significance: The theta network may represent a sensitive measure of cognitive impairment and could predict cognitive outcomes among patients with TLE.

Trial-to-Trial Variability of Spiking Delay Activity in Prefrontal Cortex Constrains Burst-Coding Models of Working Memory

A hallmark neuronal correlate of working memory (WM) is stimulus-selective spiking activity of neurons in PFC during mnemonic delays. These observations have motivated an influential computational modeling framework in which WM is supported by persistent activity. Recently, this framework has been challenged by arguments that observed persistent activity may be an artifact of trial-averaging, which potentially masks high variability of delay activity at the single-trial level. In an alternative scenario, WM delay activity could be encoded in bursts of selective neuronal firing which occur intermittently across trials. However, this alternative proposal has not been tested on single-neuron spike-train data. Here, we developed a framework for addressing this issue by characterizing the trial-to-trial variability of neuronal spiking quantified by Fano factor (FF). By building a doubly stochastic Poisson spiking model, we first demonstrated that the burst-coding proposal implies a significant increase in FF positively correlated with firing rate, and thus loss of stability across trials during the delay. Simulation of spiking cortical circuit WM models further confirmed that FF is a sensitive measure that can well dissociate distinct WM mechanisms. We then tested these predictions on datasets of single-neuron recordings from macaque PFC during three WM tasks. In sharp contrast to the burst-coding model predictions, we only found a small fraction of neurons showing increased WM-dependent burstiness, and stability across trials during delay was strengthened in empirical data. Therefore, reduced trial-to-trial variability during delay provides strong constraints on the contribution of single-neuron intermittent bursting to WM maintenance.SIGNIFICANCE STATEMENT There are diverging classes of theoretical models explaining how information is maintained in working memory by cortical circuits. In an influential model class, neurons exhibit persistent elevated memorandum-selective firing, whereas a recently developed class of burst-coding models suggests that persistent activity is an artifact of trial-averaging, and spiking is sparse in each single trial, subserved by brief intermittent bursts. However, this alternative picture has not been characterized or tested on empirical spike-train data. Here we combine mathematical analysis, computational model simulation, and experimental data analysis to test empirically these two classes of models and show that the trial-to-trial variability of empirical spike trains is not consistent with burst coding. These findings provide constraints for theoretical models of working memory.

Adaptation supports short-term memory in a visual change detection task

The maintenance of short-term memories is critical for survival in a dynamically changing world. Previous studies suggest that this memory can be stored in the form of persistent neural activity or using a synaptic mechanism, such as with short-term plasticity. Here, we compare the predictions of these two mechanisms to neural and behavioral measurements in a visual change detection task. Mice were trained to respond to changes in a repeated sequence of natural images while neural activity was recorded using two-photon calcium imaging. We also trained two types of artificial neural networks on the same change detection task as the mice. Following fixed pre-processing using a pretrained convolutional neural network, either a recurrent neural network (RNN) or a feedforward neural network with short-term synaptic depression (STPNet) was trained to the same level of performance as the mice. While both networks are able to learn the task, the STPNet model contains units whose activity are more similar to the in vivo data and produces errors which are more similar to the mice. When images are omitted, an unexpected perturbation which was absent during training, mice often do not respond to the omission but are more likely to respond to the subsequent image. Unlike the RNN model, STPNet produces a similar pattern of behavior. These results suggest that simple neural adaptation mechanisms may serve as an important bottom-up memory signal in this task, which can be used by downstream areas in the decision-making process.

Animals have to adapt to environments with rich dynamics and maintain multiple types of memories. In this study, we focus on a visual change detection task in mice which requires short-term memory. Learning which features need to be maintained in short-term memory can be realized in a recurrent neural network by changing connections in the network, resulting in memory maintenance through persistent activity. However, in biological networks, a large diversity of time-dependent intrinsic mechanisms are also available. As an alternative to persistent neural activity, we find that learning to make use of internal adapting dynamics better matches both the observed neural activity and behavior of animals in this simple task. The presence of a large diversity of temporal traces could be one of the reasons for the diversity of cells observed. We believe that both learning to keep representations of relevant stimuli in persistent activity and learning to make use of intrinsic time-dependent mechanisms exist, and their relative use will be dependent on the exact task.