The computational foundations of dynamic coding in working memory.

Working memory (WM) is a fundamental aspect of cognition. WM maintenance is classically thought to rely on stable patterns of neural activities. However, recent evidence shows that neural population activities during WM maintenance undergo dynamic variations before settling into a stable pattern. Although this has been difficult to explain theoretically, neural network models optimized for WM typically also exhibit such dynamics. Here, we examine stable versus dynamic coding in neural data, classical models, and task-optimized networks. We review principled mathematical reasons for why classical models do not, while task-optimized models naturally do exhibit dynamic coding. We suggest an update to our understanding of WM maintenance, in which dynamic coding is a fundamental computational feature rather than an epiphenomenon.

Response sub-additivity and variability quenching in visual cortex.

Sub-additivity and variability are ubiquitous response motifs in the primary visual cortex (V1). Response sub-additivity enables the construction of useful interpretations of the visual environment, whereas response variability indicates the factors that limit the precision with which the brain can do this. There is increasing evidence that experimental manipulations that elicit response sub-additivity often also quench response variability. Here, we provide an overview of these phenomena and suggest that they may have common origins. We discuss empirical findings and recent model-based insights into the functional operations, computational objectives and circuit mechanisms underlying V1 activity. These different modelling approaches all predict that response sub-additivity and variability quenching often co-occur. The phenomenology of these two response motifs, as well as many of the insights obtained about them in V1, generalize to other cortical areas. Thus, the connection between response sub-additivity and variability quenching may be a canonical motif across the cortex.

Optimal information loading into working memory explains dynamic coding in the prefrontal cortex.

Working memory involves the short-term maintenance of information and is critical in many tasks. The neural circuit dynamics underlying working memory remain poorly understood, with different aspects of prefrontal cortical (PFC) responses explained by different putative mechanisms. By mathematical analysis, numerical simulations, and using recordings from monkey PFC, we investigate a critical but hitherto ignored aspect of working memory dynamics: information loading. We find that, contrary to common assumptions, optimal loading of information into working memory involves inputs that are largely orthogonal, rather than similar, to the late delay activities observed during memory maintenance, naturally leading to the widely observed phenomenon of dynamic coding in PFC. Using a theoretically principled metric, we show that PFC exhibits the hallmarks of optimal information loading. We also find that optimal information loading emerges as a general dynamical strategy in task-optimized recurrent neural networks. Our theory unifies previous, seemingly conflicting theories of memory maintenance based on attractor or purely sequential dynamics and reveals a normative principle underlying dynamic coding.

Parallel functional architectures within a single dendritic tree.

The input-output transformation of individual neurons is a key building block of neural circuit dynamics. While previous models of this transformation vary widely in their complexity, they all describe the underlying functional architecture as unitary, such that each synaptic input makes a single contribution to the neuronal response. Here, we show that the input-output transformation of CA1 pyramidal cells is instead best captured by two distinct functional architectures operating in parallel. We used statistically principled methods to fit flexible, yet interpretable, models of the transformation of input spikes into the somatic "output" voltage and to automatically select among alternative functional architectures. With dendritic Na+ channels blocked, responses are accurately captured by a single static and global nonlinearity. In contrast, dendritic Na+-dependent integration requires a functional architecture with multiple dynamic nonlinearities and clustered connectivity. These two architectures incorporate distinct morphological and biophysical properties of the neuron and its synaptic organization.

The Computational and Neural Bases of Context-Dependent Learning.

Flexible behavior requires the creation, updating, and expression of memories to depend on context. While the neural underpinnings of each of these processes have been intensively studied, recent advances in computational modeling revealed a key challenge in context-dependent learning that had been largely ignored previously: Under naturalistic conditions, context is typically uncertain, necessitating contextual inference. We review a theoretical approach to formalizing context-dependent learning in the face of contextual uncertainty and the core computations it requires. We show how this approach begins to organize a large body of disparate experimental observations, from multiple levels of brain organization (including circuits, systems, and behavior) and multiple brain regions (most prominently the prefrontal cortex, the hippocampus, and motor cortices), into a coherent framework. We argue that contextual inference may also be key to understanding continual learning in the brain. This theory-driven perspective places contextual inference as a core component of learning.

Spatial uncertainty and environmental geometry in navigation.

Variations in the geometry of the environment, such as the shape and size of an enclosure, have profound effects on navigational behavior and its neural underpinning. Here, we show that these effects arise as a consequence of a single, unifying principle: to navigate efficiently, the brain must maintain and update the uncertainty about one's location. We developed an image-computable Bayesian ideal observer model of navigation, continually combining noisy visual and self-motion inputs, and a neural encoding model optimized to represent the location uncertainty computed by the ideal observer. Through mathematical analysis and numerical simulations, we show that the ideal observer accounts for a diverse range of sometimes paradoxical distortions of human homing behavior in anisotropic and deformed environments, including 'boundary tethering', and its neural encoding accounts for distortions of rodent grid cell responses under identical environmental manipulations. Our results demonstrate that spatial uncertainty plays a key role in navigation.

Contextual inference in learning and memory.

Context is widely regarded as a major determinant of learning and memory across numerous domains, including classical and instrumental conditioning, episodic memory, economic decision-making, and motor learning. However, studies across these domains remain disconnected due to the lack of a unifying framework formalizing the concept of context and its role in learning. Here, we develop a unified vernacular allowing direct comparisons between different domains of contextual learning. This leads to a Bayesian model positing that context is unobserved and needs to be inferred. Contextual inference then controls the creation, expression, and updating of memories. This theoretical approach reveals two distinct components that underlie adaptation, proper and apparent learning, respectively referring to the creation and updating of memories versus time-varying adjustments in their expression. We review a number of extensions of the basic Bayesian model that allow it to account for increasingly complex forms of contextual learning.

Redistribution of the SWI/SNF Complex Dictates Coordinated Transcriptional Control over Epithelial-Mesenchymal Transition of Normal Breast Cells through TGF-ß Signaling.

Therapeutic targets in cancer cells defective for the tumor suppressor ARID1A are fundamentals of synthetic lethal strategies. However, whether modulating ARID1A function in premalignant breast epithelial cells could be exploited to reduce carcinogenic potential remains to be elucidated. In search of chromatin-modulating mechanisms activated by anti-proliferative agents in normal breast epithelial (HME-hTert) cells, we identified a distinct pattern of genome-wide H3K27 histone acetylation marks characteristic for the combined treatment by the cancer preventive rexinoid bexarotene (Bex) and carvedilol (Carv). Among these marks, several enhancers functionally linked to TGF-ß signaling were enriched for ARID1A and Brg1, subunits within the SWI/SNF chromatin-remodeling complex. The recruitment of ARID1A and Brg1 was associated with the suppression of TGFBR2, KLF4, and FoxQ1, and the induction of BMP6, while the inverse pattern ensued upon the knock-down of ARID1A. Bex+Carv treatment resulted in fewer cells expressing N-cadherin and dictated a more epithelial phenotype. However, the silencing of ARID1A expression reversed the ability of Bex and Carv to limit epithelial-mesenchymal transition. The nuclear levels of SMAD4, a canonical mediator of TGF-ß action, were more effectively suppressed by the combination than by TGF-ß. In contrast, TGF-ß treatment exceeded the ability of Bex+Carv to lower nuclear FoxQ1 levels and induced markedly higher E-cadherin positivity, indicating a target-selective antagonism of Bex+Carv to TGF-ß action. In summary, the chromatin-wide redistribution of ARID1A by Bex and Carv treatment is instrumental in the suppression of genes mediating TGF-ß signaling, and, thus, the morphologic reprogramming of normal breast epithelial cells. The concerted engagement of functionally linked targets using low toxicity clinical agents represents an attractive new approach for cancer interception.

Adaptive erasure of spurious sequences in sensory cortical circuits.

Sequential activity reflecting previously experienced temporal sequences is considered a hallmark of learning across cortical areas. However, it is unknown how cortical circuits avoid the converse problem: producing spurious sequences that are not reflecting sequences in their inputs. We develop methods to quantify and study sequentiality in neural responses. We show that recurrent circuit responses generally include spurious sequences, which are specifically prevented in circuits that obey two widely known features of cortical microcircuit organization: Dale's law and Hebbian connectivity. In particular, spike-timing-dependent plasticity in excitation-inhibition networks leads to an adaptive erasure of spurious sequences. We tested our theory in multielectrode recordings from the visual cortex of awake ferrets. Although responses to natural stimuli were largely non-sequential, responses to artificial stimuli initially included spurious sequences, which diminished over extended exposure. These results reveal an unexpected role for Hebbian experience-dependent plasticity and Dale's law in sensory cortical circuits.

Planning in the brain.

Recent breakthroughs in artificial intelligence (AI) have enabled machines to plan in tasks previously thought to be uniquely human. Meanwhile, the planning algorithms implemented by the brain itself remain largely unknown. Here, we review neural and behavioral data in sequential decision-making tasks that elucidate the ways in which the brain does-and does not-plan. To systematically review available biological data, we create a taxonomy of planning algorithms by summarizing the relevant design choices for such algorithms in AI. Across species, recording techniques, and task paradigms, we find converging evidence that the brain represents future states consistent with a class of planning algorithms within our taxonomy-focused, depth-limited, and serial. However, we argue that current data are insufficient for addressing more detailed algorithmic questions. We propose a new approach leveraging AI advances to drive experiments that can adjudicate between competing candidate algorithms.

Contextual inference underlies the learning of sensorimotor repertoires.

ASBTRACT: Humans spend a lifetime learning, storing and refining a repertoire of motor memories. For example, through experience, we become proficient at manipulating a large range of objects with distinct dynamical properties. However, it is unknown what principle underlies how our continuous stream of sensorimotor experience is segmented into separate memories and how we adapt and use this growing repertoire. Here we develop a theory of motor learning based on the key principle that memory creation, updating and expression are all controlled by a single computation-contextual inference. Our theory reveals that adaptation can arise both by creating and updating memories (proper learning) and by changing how existing memories are differentially expressed (apparent learning). This insight enables us to account for key features of motor learning that had no unified explanation: spontaneous recovery1, savings2, anterograde interference3, how environmental consistency affects learning rate4,5 and the distinction between explicit and implicit learning6. Critically, our theory also predicts new phenomena-evoked recovery and context-dependent single-trial learning-which we confirm experimentally. These results suggest that contextual inference, rather than classical single-context mechanisms1,4,7-9, is the key principle underlying how a diverse set of experiences is reflected in our motor behaviour.

Representations of uncertainty: where art thou?

Perception is often described as probabilistic inference requiring an internal representation of uncertainty. However, it is unknown whether uncertainty is represented in a task-dependent manner, solely at the level of decisions, or in a fully Bayesian manner, across the entire perceptual pathway. To address this question, we first codify and evaluate the possible strategies the brain might use to represent uncertainty, and highlight the normative advantages of fully Bayesian representations. In such representations, uncertainty information is explicitly represented at all stages of processing, including early sensory areas, allowing for flexible and efficient computations in a wide variety of situations. Next, we critically review neural and behavioral evidence about the representation of uncertainty in the brain agreeing with fully Bayesian representations. We argue that sufficient behavioral evidence for fully Bayesian representations is lacking and suggest experimental approaches for demonstrating the existence of multivariate posterior distributions along the perceptual pathway.

Cortical-like dynamics in recurrent circuits optimized for sampling-based probabilistic inference.

Sensory cortices display a suite of ubiquitous dynamical features, such as ongoing noise variability, transient overshoots and oscillations, that have so far escaped a common, principled theoretical account. We developed a unifying model for these phenomena by training a recurrent excitatory-inhibitory neural circuit model of a visual cortical hypercolumn to perform sampling-based probabilistic inference. The optimized network displayed several key biological properties, including divisive normalization and stimulus-modulated noise variability, inhibition-dominated transients at stimulus onset and strong gamma oscillations. These dynamical features had distinct functional roles in speeding up inferences and made predictions that we confirmed in novel analyses of recordings from awake monkeys. Our results suggest that the basic motifs of cortical dynamics emerge as a consequence of the efficient implementation of the same computational function-fast sampling-based inference-and predict further properties of these motifs that can be tested in future experiments.

Representational untangling by the firing rate nonlinearity in V1 simple cells.

An important computational goal of the visual system is 'representational untangling' (RU): representing increasingly complex features of visual scenes in an easily decodable format. RU is typically assumed to be achieved in high-level visual cortices via several stages of cortical processing. Here we show, using a canonical population coding model, that RU of low-level orientation information is already performed at the first cortical stage of visual processing, but not before that, by a fundamental cellular-level property: the thresholded firing rate nonlinearity of simple cells in the primary visual cortex (V1). We identified specific, experimentally measurable parameters that determined the optimal firing threshold for RU and found that the thresholds of V1 simple cells extracted from in vivo recordings in awake behaving mice were near optimal. These results suggest that information re-formatting, rather than maximisation, may already be a relevant computational goal for the early visual system.

Unimodal statistical learning produces multimodal object-like representations.

The concept of objects is fundamental to cognition and is defined by a consistent set of sensory properties and physical affordances. Although it is unknown how the abstract concept of an object emerges, most accounts assume that visual or haptic boundaries are crucial in this process. Here, we tested an alternative hypothesis that boundaries are not essential but simply reflect a more fundamental principle: consistent visual or haptic statistical properties. Using a novel visuo-haptic statistical learning paradigm, we familiarised participants with objects defined solely by across-scene statistics provided either visually or through physical interactions. We then tested them on both a visual familiarity and a haptic pulling task, thus measuring both within-modality learning and across-modality generalisation. Participants showed strong within-modality learning and 'zero-shot' across-modality generalisation which were highly correlated. Our results demonstrate that humans can segment scenes into objects, without any explicit boundary cues, using purely statistical information.

Global and Multiplexed Dendritic Computations under In Vivo-like Conditions.

Dendrites integrate inputs nonlinearly, but it is unclear how these nonlinearities contribute to the overall input-output transformation of single neurons. We developed statistically principled methods using a hierarchical cascade of linear-nonlinear subunits (hLN) to model the dynamically evolving somatic response of neurons receiving complex, in vivo-like spatiotemporal synaptic input patterns. We used the hLN to predict the somatic membrane potential of an in vivo-validated detailed biophysical model of a L2/3 pyramidal cell. Linear input integration with a single global dendritic nonlinearity achieved above 90% prediction accuracy. A novel hLN motif, input multiplexing into parallel processing channels, could improve predictions as much as conventionally used additional layers of local nonlinearities. We obtained similar results in two other cell types. This approach provides a data-driven characterization of a key component of cortical circuit computations: the input-output transformation of neurons during in vivo-like conditions.

The Redemption of Noise: Inference with Neural Populations.

In 2006, Ma et al. (Nat. Neurosci. 1006;9:1432-1438) presented an elegant theory for how populations of neurons might represent uncertainty to perform Bayesian inference. Critically, according to this theory, neural variability is no longer a nuisance, but rather a vital part of how the brain encodes probability distributions and performs computations with them.