Statistically optimal perception and learning: from behavior to neural representations

Predictive coding for the actions and emotions of others and its deficits in autism spectrum disorders

Traditionally, the neural basis of social perception has been studied by showing participants brief examples of the actions or emotions of others presented in randomized order to prevent participants from anticipating what others do and feel. This approach is optimal to isolate the importance of information flow from lower to higher cortical areas. The degree to which feedback connections and Bayesian hierarchical predictive coding contribute to how mammals process more complex social stimuli has been less explored, and will be the focus of this review. We illustrate paradigms that start to capture how participants predict the actions and emotions of others under more ecological conditions, and discuss the brain activity measurement methods suitable to reveal the importance of feedback connections in these predictions. Together, these efforts draw a richer picture of social cognition in which predictive coding and feedback connections play significant roles. We further discuss how the notion of predicting coding is influencing how we think of autism spectrum disorder.

Learning to integrate parts for whole through correlated neural variability

Neural activity in the cortex exhibits a wide range of firing variability and rich correlation structures. Studies on neural coding indicate that correlated neural variability can influence the quality of neural codes, either beneficially or adversely. However, the mechanisms by which correlated neural variability is transformed and processed across neural populations to achieve meaningful computation remain largely unclear. Here we propose a theory of covariance computation with spiking neurons which offers a unifying perspective on neural representation and computation with correlated noise. We employ a recently proposed computational framework known as the moment neural network to resolve the nonlinear coupling of correlated neural variability with a task-driven approach to constructing neural network models for performing covariance-based perceptual tasks. In particular, we demonstrate how perceptual information initially encoded entirely within the covariance of upstream neurons' spiking activity can be passed, in a near-lossless manner, to the mean firing rate of downstream neurons, which in turn can be used to inform inference. The proposed theory of covariance computation addresses an important question of how the brain extracts perceptual information from noisy sensory stimuli to generate a stable perceptual whole and indicates a more direct role that correlated variability plays in cortical information processing.

A visual representation of the hand in the resting somatomotor regions of the human brain

Hand visibility affects motor control, perception, and attention, as visual information is integrated into an internal model of somatomotor control. Spontaneous brain activity, i.e., at rest, in the absence of an active task, is correlated among somatomotor regions that are jointly activated during motor tasks. Recent studies suggest that spontaneous activity patterns not only replay task activation patterns but also maintain a model of the body's and environment's statistical regularities (priors), which may be used to predict upcoming behavior. Here, we test whether spontaneous activity in the human somatomotor cortex as measured using fMRI is modulated by visual stimuli that display hands vs. non-hand stimuli and by the use/action they represent. A multivariate pattern analysis was performed to examine the similarity between spontaneous activity patterns and task-evoked patterns to the presentation of natural hands, robot hands, gloves, or control stimuli (food). In the left somatomotor cortex, we observed a stronger (multivoxel) spatial correlation between resting state activity and natural hand picture patterns compared to other stimuli. No task-rest similarity was found in the visual cortex. Spontaneous activity patterns in somatomotor brain regions code for the visual representation of human hands and their use.

Signatures of Bayesian inference emerge from energy-efficient synapses

Biological synaptic transmission is unreliable, and this unreliability likely degrades neural circuit performance. While there are biophysical mechanisms that can increase reliability, for instance by increasing vesicle release probability, these mechanisms cost energy. We examined four such mechanisms along with the associated scaling of the energetic costs. We then embedded these energetic costs for reliability in artificial neural networks (ANNs) with trainable stochastic synapses, and trained these networks on standard image classification tasks. The resulting networks revealed a tradeoff between circuit performance and the energetic cost of synaptic reliability. Additionally, the optimised networks exhibited two testable predictions consistent with pre-existing experimental data. Specifically, synapses with lower variability tended to have (1) higher input firing rates and (2) lower learning rates. Surprisingly, these predictions also arise when synapse statistics are inferred through Bayesian inference. Indeed, we were able to find a formal, theoretical link between the performance-reliability cost tradeoff and Bayesian inference. This connection suggests two incompatible possibilities: evolution may have chanced upon a scheme for implementing Bayesian inference by optimising energy efficiency, or alternatively, energy-efficient synapses may display signatures of Bayesian inference without actually using Bayes to reason about uncertainty.

Plinko: Eliciting beliefs to build better models of statistical learning and mental model updating

Prior beliefs are central to Bayesian accounts of cognition, but many of these accounts do not directly measure priors. More specifically, initial states of belief heavily influence how new information is assumed to be utilized when updating a particular model. Despite this, prior and posterior beliefs are either inferred from sequential participant actions or elicited through impoverished means. We had participants to play a version of the game 'Plinko', to first elicit individual participant priors in a theoretically agnostic manner. Subsequent learning and updating of participant beliefs was then directly measured. We show that participants hold various priors that cluster around prototypical probability distributions that in turn influence learning. In follow-up studies, we show that participant priors are stable over time and that the ability to update beliefs is influenced by a simple environmental manipulation (i.e., a short break). These data reveal the importance of directly measuring participant beliefs rather than assuming or inferring them as has been widely done in the literature to date. The Plinko game provides a flexible and fecund means for examining statistical learning and mental model updating.

Desiderata for Normative Models of Synaptic Plasticity

Normative models of synaptic plasticity use computational rationales to arrive at predictions of behavioral and network-level adaptive phenomena. In recent years, there has been an explosion of theoretical work in this realm, but experimental confirmation remains limited. In this review, we organize work on normative plasticity models in terms of a set of desiderata that, when satisfied, are designed to ensure that a given model demonstrates a clear link between plasticity and adaptive behavior, is consistent with known biological evidence about neural plasticity and yields specific testable predictions. As a prototype, we include a detailed analysis of the REINFORCE algorithm. We also discuss how new models have begun to improve on the identified criteria and suggest avenues for further development. Overall, we provide a conceptual guide to help develop neural learning theories that are precise, powerful, and experimentally testable.

Single trial Bayesian inference by population vector readout in the barn owl's sound localization system

Bayesian models have proven effective in characterizing perception, behavior, and neural encoding across diverse species and systems. The neural implementation of Bayesian inference in the barn owl's sound localization system and behavior has been previously explained by a non-uniform population code model. This model specifies the neural population activity pattern required for a population vector readout to match the optimal Bayesian estimate. While prior analyses focused on trial-averaged comparisons of model predictions with behavior and single-neuron responses, it remains unknown whether this model can accurately approximate Bayesian inference on single trials under varying sensory reliability, a fundamental condition for natural perception and behavior. In this study, we utilized mathematical analysis and simulations to demonstrate that decoding a non-uniform population code via a population vector readout approximates the Bayesian estimate on single trials for varying sensory reliabilities. Our findings provide additional support for the non-uniform population code model as a viable explanation for the barn owl's sound localization pathway and behavior.

Pattern completion and disruption characterize contextual modulation in the visual cortex

Vision is fundamentally context-dependent, with neuronal responses influenced not just by local features but also by surrounding contextual information. In the visual cortex, studies using simple grating stimuli indicate that congruent stimuli - where the center and surround share the same orientation - are more inhibitory than when orientations are orthogonal, potentially serving redundancy reduction and predictive coding. Understanding these center-surround interactions in relation to natural image statistics is challenging due to the high dimensionality of the stimulus space, yet crucial for deciphering the neuronal code of real-world sensory processing. Utilizing large-scale recordings from mouse V1, we trained convolutional neural networks (CNNs) to predict and synthesize surround patterns that either optimally suppressed or enhanced responses to center stimuli, confirmed by in vivo experiments. Contrary to the notion that congruent stimuli are suppressive, we found that surrounds that completed patterns based on natural image statistics were facilitatory, while disruptive surrounds were suppressive. Applying our CNN image synthesis method in macaque V1, we discovered that pattern completion within the near surround occurred more frequently with excitatory than with inhibitory surrounds, suggesting that our results in mice are conserved in macaques. Further, experiments and model analyses confirmed previous studies reporting the opposite effect with grating stimuli in both species. Using the MICrONS functional connectomics dataset, we observed that neurons with similar feature selectivity formed excitatory connections regardless of their receptive field overlap, aligning with the pattern completion phenomenon observed for excitatory surrounds. Finally, our empirical results emerged in a normative model of perception implementing Bayesian inference, where neuronal responses are modulated by prior knowledge of natural scene statistics. In summary, our findings identify a novel relationship between contextual information and natural scene statistics and provide evidence for a role of contextual modulation in hierarchical inference.

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.

Neural representation in active inference: Using generative models to interact with-and understand-the lived world

This paper considers neural representation through the lens of active inference, a normative framework for understanding brain function. It delves into how living organisms employ generative models to minimize the discrepancy between predictions and observations (as scored with variational free energy). The ensuing analysis suggests that the brain learns generative models to navigate the world adaptively, not (or not solely) to understand it. Different living organisms may possess an array of generative models, spanning from those that support action-perception cycles to those that underwrite planning and imagination; namely, from explicit models that entail variables for predicting concurrent sensations, like objects, faces, or people-to action-oriented models that predict action outcomes. It then elucidates how generative models and belief dynamics might link to neural representation and the implications of different types of generative models for understanding an agent's cognitive capabilities in relation to its ecological niche. The paper concludes with open questions regarding the evolution of generative models and the development of advanced cognitive abilities-and the gradual transition from pragmatic to detached neural representations. The analysis on offer foregrounds the diverse roles that generative models play in cognitive processes and the evolution of neural representation.

Probing the Structure and Functional Properties of the Dropout-Induced Correlated Variability in Convolutional Neural Networks

Computational neuroscience studies have shown that the structure of neural variability to an unchanged stimulus affects the amount of information encoded. Some artificial deep neural networks, such as those with Monte Carlo dropout layers, also have variable responses when the input is fixed. However, the structure of the trial-by-trial neural covariance in neural networks with dropout has not been studied, and its role in decoding accuracy is unknown. We studied the above questions in a convolutional neural network model with dropout in both the training and testing phases. We found that trial-by-trial correlation between neurons (i.e., noise correlation) is positive and low dimensional. Neurons that are close in a feature map have larger noise correlation. These properties are surprisingly similar to the findings in the visual cortex. We further analyzed the alignment of the main axes of the covariance matrix. We found that different images share a common trial-by-trial noise covariance subspace, and they are aligned with the global signal covariance. This evidence that the noise covariance is aligned with signal covariance suggests that noise covariance in dropout neural networks reduces network accuracy, which we further verified directly with a trial-shuffling procedure commonly used in neuroscience. These findings highlight a previously overlooked aspect of dropout layers that can affect network performance. Such dropout networks could also potentially be a computational model of neural variability.

Tracking subjects' strategies in behavioural choice experiments at trial resolution

Investigating how, when, and what subjects learn during decision-making tasks requires tracking their choice strategies on a trial-by-trial basis. Here, we present a simple but effective probabilistic approach to tracking choice strategies at trial resolution using Bayesian evidence accumulation. We show this approach identifies both successful learning and the exploratory strategies used in decision tasks performed by humans, non-human primates, rats, and synthetic agents. Both when subjects learn and when rules change the exploratory strategies of win-stay and lose-shift, often considered complementary, are consistently used independently. Indeed, we find the use of lose-shift is strong evidence that subjects have latently learnt the salient features of a new rewarded rule. Our approach can be extended to any discrete choice strategy, and its low computational cost is ideally suited for real-time analysis and closed-loop control.

The autocorrelated Bayesian sampler: A rational process for probability judgments, estimates, confidence intervals, choices, confidence judgments, and response times

Normative models of decision-making that optimally transform noisy (sensory) information into categorical decisions qualitatively mismatch human behavior. Indeed, leading computational models have only achieved high empirical corroboration by adding task-specific assumptions that deviate from normative principles. In response, we offer a Bayesian approach that implicitly produces a posterior distribution of possible answers (hypotheses) in response to sensory information. But we assume that the brain has no direct access to this posterior, but can only sample hypotheses according to their posterior probabilities. Accordingly, we argue that the primary problem of normative concern in decision-making is integrating stochastic hypotheses, rather than stochastic sensory information, to make categorical decisions. This implies that human response variability arises mainly from posterior sampling rather than sensory noise. Because human hypothesis generation is serially correlated, hypothesis samples will be autocorrelated. Guided by this new problem formulation, we develop a new process, the Autocorrelated Bayesian Sampler (ABS), which grounds autocorrelated hypothesis generation in a sophisticated sampling algorithm. The ABS provides a single mechanism that qualitatively explains many empirical effects of probability judgments, estimates, confidence intervals, choice, confidence judgments, response times, and their relationships. Our analysis demonstrates the unifying power of a perspective shift in the exploration of normative models. It also exemplifies the proposal that the "Bayesian brain" operates using samples not probabilities, and that variability in human behavior may primarily reflect computational rather than sensory noise. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Bounded contribution of human early visual cortex to the topographic anisotropy in spatial extent perception

To interact successfully with objects, it is crucial to accurately perceive their spatial extent, an enclosed region they occupy in space. Although the topographic representation of space in the early visual cortex (EVC) has been favored as a neural correlate of spatial extent perception, its exact nature and contribution to perception remain unclear. Here, we inspect the topographic representations of human individuals' EVC and perception in terms of how much their anisotropy is influenced by the orientation (co-axiality) and radial position (radiality) of stimuli. We report that while the anisotropy is influenced by both factors, its direction is primarily determined by radiality in EVC but by co-axiality in perception. Despite this mismatch, the individual differences in both radial and co-axial anisotropy are substantially shared between EVC and perception. Our findings suggest that spatial extent perception builds on EVC's spatial representation but requires an additional mechanism to transform its topographic bias.

Neural correlates tracking different aspects of the emerging representation of novel visual categories

Current studies investigating electroencephalogram correlates associated with categorization of sensory stimuli (P300 event-related potential, alpha event-related desynchronization, theta event-related synchronization) typically use an oddball paradigm with few, familiar, highly distinct stimuli providing limited insight about the aspects of categorization (e.g. difficulty, membership, uncertainty) that the correlates are linked to. Using a more complex task, we investigated whether such more specific links could be established between correlates and learning and how these links change during the emergence of new categories. In our study, participants learned to categorize novel stimuli varying continuously on multiple integral feature dimensions, while electroencephalogram was recorded from the beginning of the learning process. While there was no significant P300 event-related potential modulation, both alpha event-related desynchronization and theta event-related synchronization followed a characteristic trajectory in proportion with the gradual acquisition of the two categories. Moreover, the two correlates were modulated by different aspects of categorization, alpha event-related desynchronization by the difficulty of the task, whereas the magnitude of theta -related synchronization by the identity and possibly the strength of category membership. Thus, neural signals commonly related to categorization are appropriate for tracking both the dynamic emergence of internal representation of categories, and different meaningful aspects of the categorization process.

Metacognition and Confidence: A Review and Synthesis

Determining the psychological, computational, and neural bases of confidence and uncertainty holds promise for understanding foundational aspects of human metacognition. While a neuroscience of confidence has focused on the mechanisms underpinning subpersonal phenomena such as representations of uncertainty in the visual or motor system, metacognition research has been concerned with personal-level beliefs and knowledge about self-performance. I provide a road map for bridging this divide by focusing on a particular class of confidence computation: propositional confidence in one's own (hypothetical) decisions or actions. Propositional confidence is informed by the observer's models of the world and their cognitive system, which may be more or less accurate-thus explaining why metacognitive judgments are inferential and sometimes diverge from task performance. Disparate findings on the neural basis of uncertainty and performance monitoring are integrated into a common framework, and a new understanding of the locus of action of metacognitive interventions is developed.

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

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

Decentralized Neural Circuits of Multisensory Information Integration in the Brain

The brain combines multisensory inputs together to obtain a complete and reliable description of the world. Recent experiments suggest that several interconnected multisensory brain areas are simultaneously involved to integrate multisensory information. It was unknown how these mutually connected multisensory areas achieve multisensory integration. To answer this question, using biologically plausible neural circuit models we developed a decentralized system for information integration that comprises multiple interconnected multisensory brain areas. Through studying an example of integrating visual and vestibular cues to infer heading direction, we show that such a decentralized system is well consistent with experimental observations. In particular, we demonstrate that this decentralized system can optimally integrate information by implementing sampling-based Bayesian inference. The Poisson variability of spike generation provides appropriate variability to drive sampling, and the interconnections between multisensory areas store the correlation prior between multisensory stimuli. The decentralized system predicts that optimally integrated information emerges locally from the dynamics of the communication between brain areas and sheds new light on the interpretation of the connectivity between multisensory brain areas.

Bayesian encoding and decoding as distinct perspectives on neural coding

The Bayesian brain hypothesis is one of the most influential ideas in neuroscience. However, unstated differences in how Bayesian ideas are operationalized make it difficult to draw general conclusions about how Bayesian computations map onto neural circuits. Here, we identify one such unstated difference: some theories ask how neural circuits could recover information about the world from sensory neural activity (Bayesian decoding), whereas others ask how neural circuits could implement inference in an internal model (Bayesian encoding). These two approaches require profoundly different assumptions and lead to different interpretations of empirical data. We contrast them in terms of motivations, empirical support and relationship to neural data. We also use a simple model to argue that encoding and decoding models are complementary rather than competing. Appreciating the distinction between Bayesian encoding and Bayesian decoding will help to organize future work and enable stronger empirical tests about the nature of inference in the brain.

Sampling-based Bayesian inference in recurrent circuits of stochastic spiking neurons

Two facts about cortex are widely accepted: neuronal responses show large spiking variability with near Poisson statistics and cortical circuits feature abundant recurrent connections between neurons. How these spiking and circuit properties combine to support sensory representation and information processing is not well understood. We build a theoretical framework showing that these two ubiquitous features of cortex combine to produce optimal sampling-based Bayesian inference. Recurrent connections store an internal model of the external world, and Poissonian variability of spike responses drives flexible sampling from the posterior stimulus distributions obtained by combining feedforward and recurrent neuronal inputs. We illustrate how this framework for sampling-based inference can be used by cortex to represent latent multivariate stimuli organized either hierarchically or in parallel. A neural signature of such network sampling are internally generated differential correlations whose amplitude is determined by the prior stored in the circuit, which provides an experimentally testable prediction for our framework.

Studying the neural representations of uncertainty

The study of the brain's representations of uncertainty is a central topic in neuroscience. Unlike most quantities of which the neural representation is studied, uncertainty is a property of an observer's beliefs about the world, which poses specific methodological challenges. We analyze how the literature on the neural representations of uncertainty addresses those challenges and distinguish between 'code-driven' and 'correlational' approaches. Code-driven approaches make assumptions about the neural code for representing world states and the associated uncertainty. By contrast, correlational approaches search for relationships between uncertainty and neural activity without constraints on the neural representation of the world state that this uncertainty accompanies. To compare these two approaches, we apply several criteria for neural representations: sensitivity, specificity, invariance and functionality. Our analysis reveals that the two approaches lead to different but complementary findings, shaping new research questions and guiding future experiments.

Neural Circuits for Fast Poisson Compressed Sensing in the Olfactory Bulb

Within a single sniff, the mammalian olfactory system can decode the identity and concentration of odorants wafted on turbulent plumes of air. Yet, it must do so given access only to the noisy, dimensionally-reduced representation of the odor world provided by olfactory receptor neurons. As a result, the olfactory system must solve a compressed sensing problem, relying on the fact that only a handful of the millions of possible odorants are present in a given scene. Inspired by this principle, past works have proposed normative compressed sensing models for olfactory decoding. However, these models have not captured the unique anatomy and physiology of the olfactory bulb, nor have they shown that sensing can be achieved within the 100-millisecond timescale of a single sniff. Here, we propose a rate-based Poisson compressed sensing circuit model for the olfactory bulb. This model maps onto the neuron classes of the olfactory bulb, and recapitulates salient features of their connectivity and physiology. For circuit sizes comparable to the human olfactory bulb, we show that this model can accurately detect tens of odors within the timescale of a single sniff. We also show that this model can perform Bayesian posterior sampling for accurate uncertainty estimation. Fast inference is possible only if the geometry of the neural code is chosen to match receptor properties, yielding a distributed neural code that is not axis-aligned to individual odor identities. Our results illustrate how normative modeling can help us map function onto specific neural circuits to generate new hypotheses.

Surprise and novelty in the brain

Notions of surprise and novelty have been used in various experimental and theoretical studies across multiple brain areas and species. However, 'surprise' and 'novelty' refer to different quantities in different studies, which raises concerns about whether these studies indeed relate to the same functionalities and mechanisms in the brain. Here, we address these concerns through a systematic investigation of how different aspects of surprise and novelty relate to different brain functions and physiological signals. We review recent classifications of definitions proposed for surprise and novelty along with links to experimental observations. We show that computational modeling and quantifiable definitions enable novel interpretations of previous findings and form a foundation for future theoretical and experimental studies.

Neural correlates of valence-dependent belief and value updating during uncertainty reduction: An fNIRS study

Selective use of new information is crucial for adaptive decision-making. Combining a gamble bidding task with assessing cortical responses using functional near-infrared spectroscopy (fNIRS), we investigated potential effects of information valence on behavioral and neural processes of belief and value updating during uncertainty reduction in young adults. By modeling changes in the participants' expressed subjective values using a Bayesian model, we dissociated processes of (i) updating beliefs about statistical properties of the gamble, (ii) updating values of a gamble based on new information about its winning probabilities, as well as (iii) expectancy violation. The results showed that participants used new information to update their beliefs and values about the gambles in a quasi-optimal manner, as reflected in the selective updating only in situations with reducible uncertainty. Furthermore, their updating was valence-dependent: information indicating an increase in winning probability was underweighted, whereas information about a decrease in winning probability was updated in good agreement with predictions of the Bayesian decision theory. Results of model-based and moderation analyses showed that this valence-dependent asymmetry was associated with a distinct contribution of expectancy violation, besides belief updating, to value updating after experiencing new positive information regarding winning probabilities. In line with the behavioral results, we replicated previous findings showing involvements of frontoparietal brain regions in the different components of updating. Furthermore, this study provided novel results suggesting a valence-dependent recruitment of brain regions. Individuals with stronger oxyhemoglobin responses during value updating was more in line with predictions of the Bayesian model while integrating new information that indicates an increase in winning probability. Taken together, this study provides first results showing expectancy violation as a contributing factor to sub-optimal valence-dependent updating during uncertainty reduction and suggests limitations of normative Bayesian decision theory.

Desiderata for normative models of synaptic plasticity

Normative models of synaptic plasticity use a combination of mathematics and computational simulations to arrive at predictions of behavioral and network-level adaptive phenomena. In recent years, there has been an explosion of theoretical work on these models, but experimental confirmation is relatively limited. In this review, we organize work on normative plasticity models in terms of a set of desiderata which, when satisfied, are designed to guarantee that a model has a clear link between plasticity and adaptive behavior, consistency with known biological evidence about neural plasticity, and specific testable predictions. We then discuss how new models have begun to improve on these criteria and suggest avenues for further development. As prototypes, we provide detailed analyses of two specific models - REINFORCE and the Wake-Sleep algorithm. We provide a conceptual guide to help develop neural learning theories that are precise, powerful, and experimentally testable.

Long- and short-term history effects in a spiking network model of statistical learning

The statistical structure of the environment is often important when making decisions. There are multiple theories of how the brain represents statistical structure. One such theory states that neural activity spontaneously samples from probability distributions. In other words, the network spends more time in states which encode high-probability stimuli. Starting from the neural assembly, increasingly thought of to be the building block for computation in the brain, we focus on how arbitrary prior knowledge about the external world can both be learned and spontaneously recollected. We present a model based upon learning the inverse of the cumulative distribution function. Learning is entirely unsupervised using biophysical neurons and biologically plausible learning rules. We show how this prior knowledge can then be accessed to compute expectations and signal surprise in downstream networks. Sensory history effects emerge from the model as a consequence of ongoing learning.

Chronic exposure to odors at naturally occurring concentrations triggers limited plasticity in early stages of Drosophila olfactory processing

In insects and mammals, olfactory experience in early life alters olfactory behavior and function in later life. In the vinegar fly Drosophila, flies chronically exposed to a high concentration of a monomolecular odor exhibit reduced behavioral aversion to the familiar odor when it is reencountered. This change in olfactory behavior has been attributed to selective decreases in the sensitivity of second-order olfactory projection neurons (PNs) in the antennal lobe that respond to the overrepresented odor. However, since odorant compounds do not occur at similarly high concentrations in natural sources, the role of odor experience-dependent plasticity in natural environments is unclear. Here, we investigated olfactory plasticity in the antennal lobe of flies chronically exposed to odors at concentrations that are typically encountered in natural odor sources. These stimuli were chosen to each strongly and selectively excite a single class of primary olfactory receptor neuron (ORN), thus facilitating a rigorous assessment of the selectivity of olfactory plasticity for PNs directly excited by overrepresented stimuli. Unexpectedly, we found that chronic exposure to three such odors did not result in decreased PN sensitivity but rather mildly increased responses to weak stimuli in most PN types. Odor-evoked PN activity in response to stronger stimuli was mostly unaffected by odor experience. When present, plasticity was observed broadly in multiple PN types and thus was not selective for PNs receiving direct input from the chronically active ORNs. We further investigated the DL5 olfactory coding channel and found that chronic odor-mediated excitation of its input ORNs did not affect PN intrinsic properties, local inhibitory innervation, ORN responses or ORN-PN synaptic strength; however, broad-acting lateral excitation evoked by some odors was increased. These results show that PN odor coding is only mildly affected by strong persistent activation of a single olfactory input, highlighting the stability of early stages of insect olfactory processing to significant perturbations in the sensory environment.

Rewards weaken cross-modal inhibition of return with visual targets

Previous studies have shown that rewards weaken visual inhibition of return (IOR). However, the specific mechanisms underlying the influence of rewards on cross-modal IOR remain unclear. Based on the Posner exogenous cue-target paradigm, the present study was conducted to investigate the effect of rewards on exogenous spatial cross-modal IOR in both visual cue with auditory target (VA) and auditory cue with visual target (AV) conditions. The results showed the following: in the AV condition, the IOR effect size in the high-reward condition was significantly lower than that in the low-reward condition. However, in the VA condition, there was no significant IOR in either the high- or low-reward condition and there was no significant difference between the two conditions. In other words, the use of rewards modulated exogenous spatial cross-modal IOR with visual targets; specifically, high rewards may have weakened IOR in the AV condition. Taken together, our study extended the effect of rewards on IOR to cross-modal attention conditions and demonstrated for the first time that higher motivation among individuals under high-reward conditions weakened the cross-modal IOR with visual targets. Moreover, the present study provided evidence for future research on the relationship between reward and attention.

The cost of aiming for the best answers: Inconsistent perception

The laws of physics and mathematics describe the world we live in as internally consistent. As these rules provide a very effective description, and our interaction with the world is also very effective, it seems self-evident that our perception follows these laws. As a result, when trying to explain imperfections in perception, we tend to impose consistency and introduce concepts such as deformations of visual space. In this review, we provide numerous examples that show that in many situations we perceive related attributes to have inconsistent values. We discuss how our tendency to assume consistency leads to erroneous conclusions on how we process sensory information. We propose that perception is not about creating a consistent internal representation of the outside world, but about answering specific questions about the outside world. As the information used to answer a question is specific for that question, this naturally leads to inconsistencies in perception and to an apparent dissociation between some perceptual judgments and related actions.

A characterization of the neural representation of confidence during probabilistic learning

Learning in a stochastic and changing environment is a difficult task. Models of learning typically postulate that observations that deviate from the learned predictions are surprising and used to update those predictions. Bayesian accounts further posit the existence of a confidence-weighting mechanism: learning should be modulated by the confidence level that accompanies those predictions. However, the neural bases of this confidence are much less known than the ones of surprise. Here, we used a dynamic probability learning task and high-field MRI to identify putative cortical regions involved in the representation of confidence about predictions during human learning. We devised a stringent test based on the conjunction of four criteria. We localized several regions in parietal and frontal cortices whose activity is sensitive to the confidence of an ideal observer, specifically so with respect to potential confounds (surprise and predictability), and in a way that is invariant to which item is predicted. We also tested for functionality in two ways. First, we localized regions whose activity patterns at the subject level showed an effect of both confidence and surprise in qualitative agreement with the confidence-weighting principle. Second, we found neural representations of ideal confidence that also accounted for subjective confidence. Taken together, those results identify a set of cortical regions potentially implicated in the confidence-weighting of learning.

Motor invariants in action execution and perception

The nervous system is sensitive to statistical regularities of the external world and forms internal models of these regularities to predict environmental dynamics. Given the inherently social nature of human behavior, being capable of building reliable predictive models of others' actions may be essential for successful interaction. While social prediction might seem to be a daunting task, the study of human motor control has accumulated ample evidence that our movements follow a series of kinematic invariants, which can be used by observers to reduce their uncertainty during social exchanges. Here, we provide an overview of the most salient regularities that shape biological motion, examine the role of these invariants in recognizing others' actions, and speculate that anchoring socially-relevant perceptual decisions to such kinematic invariants provides a key computational advantage for inferring conspecifics' goals and intentions.

Flexible sensory-motor mapping rules manifest in correlated variability of stimulus and action codes across the brain

Humans and non-human primates can flexibly switch between different arbitrary mappings from sensation to action to solve a cognitive task. It has remained unknown how the brain implements such flexible sensory-motor mapping rules. Here, we uncovered a dynamic reconfiguration of task-specific correlated variability between sensory and motor brain regions. Human participants switched between two rules for reporting visual orientation judgments during fMRI recordings. Rule switches were either signaled explicitly or inferred by the participants from ambiguous cues. We used behavioral modeling to reconstruct the time course of their belief about the active rule. In both contexts, the patterns of correlations between ongoing fluctuations in stimulus- and action-selective activity across visual- and action-related brain regions tracked participants' belief about the active rule. The rule-specific correlation patterns broke down around the time of behavioral errors. We conclude that internal beliefs about task state are instantiated in brain-wide, selective patterns of correlated variability.

Children and adults rely on different heuristics for estimation of durations

Time is a uniquely human yet culturally ubiquitous concept acquired over childhood and provides an underlying dimension for episodic memory and estimating durations. Because time, unlike distance, lacks a sensory representation, we hypothesized that subjects at different ages attribute different meanings to it when comparing durations; pre-kindergarten children compare the density of events, while adults use the concept of observer-independent absolute time. We asked groups of pre-kindergarteners, school-age children, and adults to compare the durations of an "eventful" and "uneventful" video, both 1-minute long but durations unknown to subjects. In addition, participants were asked to express the durations of both videos non-verbally with simple hand gestures. Statistical analysis has revealed highly polarized temporal biases in each group, where pre-kindergarteners estimated the duration of the eventful video as "longer." In contrast, the school-age group of children and adults claimed the same about the uneventful video. The tendency to represent temporal durations with a horizontal hand gesture was evident among all three groups, with an increasing prevalence with age. These results support the hypothesis that pre-kindergarten-age children use heuristics to estimate time, and they convert from availability to sampling heuristics between pre-kindergarten and school age.

Size tuning of neural response variability in laminar circuits of macaque primary visual cortex

A defining feature of the cortex is its laminar organization, which is likely critical for cortical information processing. For example, visual stimuli of different size evoke distinct patterns of laminar activity. Visual information processing is also influenced by the response variability of individual neurons and the degree to which this variability is correlated among neurons. To elucidate laminar processing, we studied how neural response variability across the layers of macaque primary visual cortex is modulated by visual stimulus size. Our laminar recordings revealed that single neuron response variability and the shared variability among neurons are tuned for stimulus size, and this size-tuning is layer-dependent. In all layers, stimulation of the receptive field (RF) reduced single neuron variability, and the shared variability among neurons, relative to their pre-stimulus values. As the stimulus was enlarged beyond the RF, both single neuron and shared variability increased in supragranular layers, but either did not change or decreased in other layers. Surprisingly, we also found that small visual stimuli could increase variability relative to baseline values. Our results suggest multiple circuits and mechanisms as the source of variability in different layers and call for the development of new models of neural response variability.

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.

Where is the error? Hierarchical predictive coding through dendritic error computation

Top-down feedback in cortex is critical for guiding sensory processing, which has prominently been formalized in the theory of hierarchical predictive coding (hPC). However, experimental evidence for error units, which are central to the theory, is inconclusive and it remains unclear how hPC can be implemented with spiking neurons. To address this, we connect hPC to existing work on efficient coding in balanced networks with lateral inhibition and predictive computation at apical dendrites. Together, this work points to an efficient implementation of hPC with spiking neurons, where prediction errors are computed not in separate units, but locally in dendritic compartments. We then discuss the correspondence of this model to experimentally observed connectivity patterns, plasticity, and dynamics in cortex.

Efficient coding theory of dynamic attentional modulation

Activity of sensory neurons is driven not only by external stimuli but also by feedback signals from higher brain areas. Attention is one particularly important internal signal whose presumed role is to modulate sensory representations such that they only encode information currently relevant to the organism at minimal cost. This hypothesis has, however, not yet been expressed in a normative computational framework. Here, by building on normative principles of probabilistic inference and efficient coding, we developed a model of dynamic population coding in the visual cortex. By continuously adapting the sensory code to changing demands of the perceptual observer, an attention-like modulation emerges. This modulation can dramatically reduce the amount of neural activity without deteriorating the accuracy of task-specific inferences. Our results suggest that a range of seemingly disparate cortical phenomena such as intrinsic gain modulation, attention-related tuning modulation, and response variability could be manifestations of the same underlying principles, which combine efficient sensory coding with optimal probabilistic inference in dynamic environments.

The effects of base rate neglect on sequential belief updating and real-world beliefs

Base-rate neglect is a pervasive bias in judgment that is conceptualized as underweighting of prior information and can have serious consequences in real-world scenarios. This bias is thought to reflect variability in inferential processes but empirical support for a cohesive theory of base-rate neglect with sufficient explanatory power to account for longer-term and real-world beliefs is lacking. A Bayesian formalization of base-rate neglect in the context of sequential belief updating predicts that belief trajectories should exhibit dynamic patterns of dependence on the order in which evidence is presented and its consistency with prior beliefs. To test this, we developed a novel 'urn-and-beads' task that systematically manipulated the order of colored bead sequences and elicited beliefs via an incentive-compatible procedure. Our results in two independent online studies confirmed the predictions of the sequential base-rate neglect model: people exhibited beliefs that are more influenced by recent evidence and by evidence inconsistent with prior beliefs. We further found support for a noisy-sampling inference model whereby base-rate neglect results from rational discounting of noisy internal representations of prior beliefs. Finally, we found that model-derived indices of base-rate neglect-including noisier prior representation-correlated with propensity for unusual beliefs outside the laboratory. Our work supports the relevance of Bayesian accounts of sequential base-rate neglect to real-world beliefs and hints at strategies to minimize deleterious consequences of this pervasive bias.

Calibrating Vision: Concepts and Questions

The idea that visual coding and perception are shaped by experience and adjust to changes in the environment or the observer is universally recognized as a cornerstone of visual processing, yet the functions and processes mediating these calibrations remain in many ways poorly understood. In this article we review a number of facets and issues surrounding the general notion of calibration, with a focus on plasticity within the encoding and representational stages of visual processing. These include how many types of calibrations there are - and how we decide; how plasticity for encoding is intertwined with other principles of sensory coding; how it is instantiated at the level of the dynamic networks mediating vision; how it varies with development or between individuals; and the factors that may limit the form or degree of the adjustments. Our goal is to give a small glimpse of an enormous and fundamental dimension of vision, and to point to some of the unresolved questions in our understanding of how and why ongoing calibrations are a pervasive and essential element of vision.

Sampling motion trajectories during hippocampal theta sequences

Efficient planning in complex environments requires that uncertainty associated with current inferences and possible consequences of forthcoming actions is represented. Representation of uncertainty has been established in sensory systems during simple perceptual decision making tasks but it remains unclear if complex cognitive computations such as planning and navigation are also supported by probabilistic neural representations. Here, we capitalized on gradually changing uncertainty along planned motion trajectories during hippocampal theta sequences to capture signatures of uncertainty representation in population responses. In contrast with prominent theories, we found no evidence of encoding parameters of probability distributions in the momentary population activity recorded in an open-field navigation task in rats. Instead, uncertainty was encoded sequentially by sampling motion trajectories randomly and efficiently in subsequent theta cycles from the distribution of potential trajectories. Our analysis is the first to demonstrate that the hippocampus is well equipped to contribute to optimal planning by representing uncertainty.

Classification of brain states that predicts future performance in visual tasks based on co-integration analysis of EEG data

Electroencephalogram (EEG) is a popular tool for studying brain activity. Numerous statistical techniques exist to enhance understanding of the complex dynamics underlying the EEG recordings. Inferring the functional network connectivity between EEG channels is of interest, and non-parametric inference methods are typically applied. We propose a fully parametric model-based approach via cointegration analysis. It not only estimates the network but also provides further insight through cointegration vectors, which characterize equilibrium states, and the corresponding loadings, which describe the mechanism of how the EEG dynamics is drawn to the equilibrium. We outline the estimation procedure in the context of EEG data, which faces specific challenges compared with the common econometric problems, for which cointegration analysis was originally conceived. In particular, the dimension is higher, typically around 64; there is usually access to repeated trials; and the data are artificially linearly dependent through the normalization done in EEG recordings. Finally, we illustrate the method on EEG data from a visual task experiment and show how brain states identified via cointegration analysis can be utilized in further investigations of determinants playing roles in sensory identifications.

Hierarchical amplitude modulation structures and rhythm patterns: Comparing Western musical genres, song, and nature sounds to Babytalk

Statistical learning of physical stimulus characteristics is important for the development of cognitive systems like language and music. Rhythm patterns are a core component of both systems, and rhythm is key to language acquisition by infants. Accordingly, the physical stimulus characteristics that yield speech rhythm in "Babytalk" may also describe the hierarchical rhythmic relationships that characterize human music and song. Computational modelling of the amplitude envelope of "Babytalk" (infant-directed speech, IDS) using a demodulation approach (Spectral-Amplitude Modulation Phase Hierarchy model, S-AMPH) can describe these characteristics. S-AMPH modelling of Babytalk has shown previously that bands of amplitude modulations (AMs) at different temporal rates and their phase relations help to create its structured inherent rhythms. Additionally, S-AMPH modelling of children's nursery rhymes shows that different rhythm patterns (trochaic, iambic, dactylic) depend on the phase relations between AM bands centred on ~2 Hz and ~5 Hz. The importance of these AM phase relations was confirmed via a second demodulation approach (PAD, Probabilistic Amplitude Demodulation). Here we apply both S-AMPH and PAD to demodulate the amplitude envelopes of Western musical genres and songs. Quasi-rhythmic and non-human sounds found in nature (birdsong, rain, wind) were utilized for control analyses. We expected that the physical stimulus characteristics in human music and song from an AM perspective would match those of IDS. Given prior speech-based analyses, we also expected that AM cycles derived from the modelling may identify musical units like crotchets, quavers and demi-quavers. Both models revealed an hierarchically-nested AM modulation structure for music and song, but not nature sounds. This AM modulation structure for music and song matched IDS. Both models also generated systematic AM cycles yielding musical units like crotchets and quavers. Both music and language are created by humans and shaped by culture. Acoustic rhythm in IDS and music appears to depend on many of the same physical characteristics, facilitating learning.

Statistical learning of distractor co-occurrences facilitates visual search

Visual search is facilitated by knowledge of the relationship between the target and the distractors, including both where the target is likely to be among the distractors and how it differs from the distractors. Whether the statistical structure among distractors themselves, unrelated to target properties, facilitates search is less well understood. Here, we assessed the benefit of distractor structure using novel shapes whose relationship to each other was learned implicitly during visual search. Participants searched for target items in arrays of shapes that comprised either four pairs of co-occurring distractor shapes (structured scenes) or eight distractor shapes randomly partitioned into four pairs on each trial (unstructured scenes). Across five online experiments (N = 1,140), we found that after a period of search training, participants were more efficient when searching for targets in structured than unstructured scenes. This structure benefit emerged independently of whether the position of the shapes within each pair was fixed or variable and despite participants having no explicit knowledge of the structured pairs they had seen. These results show that implicitly learned co-occurrence statistics between distractor shapes increases search efficiency. Increased efficiency in the rejection of regularly co-occurring distractors may contribute to the efficiency of visual search in natural scenes, where such regularities are abundant.

Fractional neural sampling as a theory of spatiotemporal probabilistic computations in neural circuits

A range of perceptual and cognitive processes have been characterized from the perspective of probabilistic representations and inference. To understand the neural circuit mechanism underlying these probabilistic computations, we develop a theory based on complex spatiotemporal dynamics of neural population activity. We first implement and explore this theory in a biophysically realistic, spiking neural circuit. Population activity patterns emerging from the circuit capture realistic variability or fluctuations of neural dynamics both in time and in space. These activity patterns implement a type of probabilistic computations that we name fractional neural sampling (FNS). We further develop a mathematical model to reveal the algorithmic nature of FNS and its computational advantages for representing multimodal distributions, a major challenge faced by existing theories. We demonstrate that FNS provides a unified account of a diversity of experimental observations of neural spatiotemporal dynamics and perceptual processes such as visual perception inference, and that FNS makes experimentally testable predictions.

Statistical Learning in Vision

Vision and learning have long been considered to be two areas of research linked only distantly. However, recent developments in vision research have changed the conceptual definition of vision from a signal-evaluating process to a goal-oriented interpreting process, and this shift binds learning, together with the resulting internal representations, intimately to vision. In this review, we consider various types of learning (perceptual, statistical, and rule/abstract) associated with vision in the past decades and argue that they represent differently specialized versions of the fundamental learning process, which must be captured in its entirety when applied to complex visual processes. We show why the generalized version of statistical learning can provide the appropriate setup for such a unified treatment of learning in vision, what computational framework best accommodates this kind of statistical learning, and what plausible neural scheme could feasibly implement this framework. Finally, we list the challenges that the field of statistical learning faces in fulfilling the promise of being the right vehicle for advancing our understanding of vision in its entirety. Expected final online publication date for the Annual Review of Vision Science, Volume 8 is September 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Drifting neuronal representations: Bug or feature?

The brain displays a remarkable ability to sustain stable memories, allowing animals to execute precise behaviors or recall stimulus associations years after they were first learned. Yet, recent long-term recording experiments have revealed that single-neuron representations continuously change over time, contravening the classical assumption that learned features remain static. How do unstable neural codes support robust perception, memories, and actions? Here, we review recent experimental evidence for such representational drift across brain areas, as well as dissections of its functional characteristics and underlying mechanisms. We emphasize theoretical proposals for how drift need not only be a form of noise for which the brain must compensate. Rather, it can emerge from computationally beneficial mechanisms in hierarchical networks performing robust probabilistic computations.

Towards a cross-level understanding of Bayesian inference in the brain

Perception emerges from unconscious probabilistic inference, which guides behaviour in our ubiquitously uncertain environment. Bayesian decision theory is a prominent computational model that describes how people make rational decisions using noisy and ambiguous sensory observations. However, critical questions have been raised about the validity of the Bayesian framework in explaining the mental process of inference. Firstly, some natural behaviours deviate from Bayesian optimum. Secondly, the neural mechanisms that support Bayesian computations in the brain are yet to be understood. Taking Marr's cross level approach, we review the recent progress made in addressing these challenges. We first review studies that combined behavioural paradigms and modelling approaches to explain both optimal and suboptimal behaviours. Next, we evaluate the theoretical advances and the current evidence for ecologically feasible algorithms and neural implementations in the brain, which may enable probabilistic inference. We argue that this cross-level approach is necessary for the worthwhile pursuit to uncover mechanistic accounts of human behaviour.

Variability of regional glucose metabolism and the topology of functional networks in the human brain

The brain consumes the most energy per relative mass amongst the organs in the human body. Theoretical and empirical studies have shown that behavioral processes are relatively inexpensive metabolically, and that most energy goes to maintaining the status quo, i.e., the balance of cell membranes' resting potentials and subthreshold spontaneous activity. Spontaneous activity fluctuates across brain regions in a correlated fashion that defines multi-scale hierarchical networks called resting-state networks (RSNs). Different regions of the brain display different metabolic consumption, but the relationship between regional brain metabolism and RSNs is still under investigation. Here, we examine the variability of glucose metabolism across brain regions, measured with the relative standard uptake value (SUVR) using ¹⁸F-FDG PET, and the topology of RSNs, measured through graph analysis applied to fMRI resting-state functional connectivity (FC). We found a moderate linear relationship between the strength (STR) of pairwise regional FC and metabolism. Moreover, the linear correlation between SUVR and STR grew stronger as we considered more connected regions (hubs). Regions connecting different RSNs, or connector hubs, showed higher SUVR than regions connecting nodes within the same RSN, or provincial hubs. Our results show that functional connections as probed by fMRI are related to glucose metabolism, especially in a system of provincial and connector hubs.

Conflict and competition between model-based and model-free control

A large literature has accumulated suggesting that human and animal decision making is driven by at least two systems, and that important functions of these systems can be captured by reinforcement learning algorithms. The "model-free" system caches and uses stimulus-value or stimulus-response associations, and the "model-based" system implements more flexible planning using a model of the world. However, it is not clear how the two systems interact during deliberation and how a single decision emerges from this process, especially when they disagree. Most previous work has assumed that while the systems operate in parallel, they do so independently, and they combine linearly to influence decisions. Using an integrated reinforcement learning/drift-diffusion model, we tested the hypothesis that the two systems interact in a non-linear fashion similar to other situations with cognitive conflict. We differentiated two forms of conflict: action conflict, a binary state representing whether the systems disagreed on the best action, and value conflict, a continuous measure of the extent to which the two systems disagreed on the difference in value between the available options. We found that decisions with greater value conflict were characterized by reduced model-based control and increased caution both with and without action conflict. Action conflict itself (the binary state) acted in the opposite direction, although its effects were less prominent. We also found that between-system conflict was highly correlated with within-system conflict, and although it is less clear a priori why the latter might influence the strength of each system above its standard linear contribution, we could not rule it out. Our work highlights the importance of non-linear conflict effects, and provides new constraints for more detailed process models of decision making. It also presents new avenues to explore with relation to disorders of compulsivity, where an imbalance between systems has been implicated.