The Structure of Hippocampal CA1 Interactions Optimizes Spatial Coding across Experience.

Although much is known about how single neurons in the hippocampus represent an animal's position, how circuit interactions contribute to spatial coding is less well understood. Using a novel statistical estimator and theoretical modeling, both developed in the framework of maximum entropy models, we reveal highly structured CA1 cell-cell interactions in male rats during open field exploration. The statistics of these interactions depend on whether the animal is in a familiar or novel environment. In both conditions the circuit interactions optimize the encoding of spatial information, but for regimes that differ in the informativeness of their spatial inputs. This structure facilitates linear decodability, making the information easy to read out by downstream circuits. Overall, our findings suggest that the efficient coding hypothesis is not only applicable to individual neuron properties in the sensory periphery, but also to neural interactions in the central brain.SIGNIFICANCE STATEMENT Local circuit interactions play a key role in neural computation and are dynamically shaped by experience. However, measuring and assessing their effects during behavior remains a challenge. Here, we combine techniques from statistical physics and machine learning to develop new tools for determining the effects of local network interactions on neural population activity. This approach reveals highly structured local interactions between hippocampal neurons, which make the neural code more precise and easier to read out by downstream circuits, across different levels of experience. More generally, the novel combination of theory and data analysis in the framework of maximum entropy models enables traditional neural coding questions to be asked in naturalistic settings.

Tunable neural networks for CT image formation.

Optimization of CT image quality typically involves balancing variance and bias. In traditional filtered back-projection, this trade-off is controlled by the filter cutoff frequency. In model-based iterative reconstruction, the regularization strength parameter often serves the same function. Deep neural networks (DNNs) typically do not provide this tunable control over output image properties. Models are often trained to minimize the expected mean squared error, which penalizes both variance and bias in image outputs but does not offer any control over the trade-off between the two. We propose a method for controlling the output image properties of neural networks with a new loss function called weighted covariance and bias (WCB). Our proposed method uses multiple noise realizations of the input images during training to allow for separate weighting matrices for the variance and bias penalty terms. Moreover, we show that tuning these weights enables targeted penalization of specific image features with spatial frequency domain penalties. To evaluate our method, we present a simulation study using digital anthropomorphic phantoms, physical simulation of CT measurements, and image formation with various algorithms. We show that the WCB loss function offers a greater degree of control over trade-offs between variance and bias, whereas mean-squared error provides only one specific image quality configuration. We also show that WCB can be used to control specific image properties including variance, bias, spatial resolution, and the noise correlation of neural network outputs. Finally, we present a method to optimize the proposed weights for a spiculated lung nodule shape discrimination task. Our results demonstrate this new image quality can control the image properties of DNN outputs and optimize image quality for task-specific performance.

Weak evidence for neural correlates of task-switching in macaque V1.

A central goal of systems neuroscience is to understand how populations of sensory neurons encode and relay information to the rest of the brain. Three key quantities of interest are 1) how mean neural activity depends on the stimulus (sensitivity), 2) how neural activity (co)varies around the mean (noise correlations), and 3) how predictive these variations are of the subject's behavior (choice probability). Previous empirical work suggests that both choice probability and noise correlations are affected by task training, with decision-related information fed back to sensory areas and aligned to neural sensitivity on a task-by-task basis. We used Utah arrays to record activity from populations of primary visual cortex (V1) neurons from two macaque monkeys that were trained to switch between two coarse orientation-discrimination tasks. Surprisingly, we find no evidence for significant trial-by-trial changes in noise covariance between tasks, nor do we find a consistent relationship between neural sensitivity and choice probability, despite recording from well-tuned task-sensitive neurons, many of which were histologically confirmed to be in supragranular V1, and despite behavioral evidence that the monkeys switched their strategy between tasks. Thus our data at best provide weak support for the hypothesis that trial-by-trial task-switching induces changes to noise correlations and choice probabilities in V1. However, our data agree with a recent finding of a single "choice axis" across tasks. They also raise the intriguing possibility that choice-related signals in early sensory areas are less indicative of task learning per se and instead reflect perceptual learning that occurs in highly overtrained subjects.NEW & NOTEWORTHY Converging evidence suggests that decision processes affect sensory neural activity, and this has informed numerous theories of neural processing. We set out to replicate and extend previous results on decision-related information and noise correlations in V1 of macaque monkeys. However, in our data, we find little evidence for a number of expected effects. Our null results therefore call attention to differences in task training, stimulus design, recording, and analysis techniques between our and prior studies.

Computational methods to study information processing in neural circuits.

The brain is an information processing machine and thus naturally lends itself to be studied using computational tools based on the principles of information theory. For this reason, computational methods based on or inspired by information theory have been a cornerstone of practical and conceptual progress in neuroscience. In this Review, we address how concepts and computational tools related to information theory are spurring the development of principled theories of information processing in neural circuits and the development of influential mathematical methods for the analyses of neural population recordings. We review how these computational approaches reveal mechanisms of essential functions performed by neural circuits. These functions include efficiently encoding sensory information and facilitating the transmission of information to downstream brain areas to inform and guide behavior. Finally, we discuss how further progress and insights can be achieved, in particular by studying how competing requirements of neural encoding and readout may be optimally traded off to optimize neural information processing.

Learning shapes cortical dynamics to enhance integration of relevant sensory input.

Adaptive sensory behavior is thought to depend on processing in recurrent cortical circuits, but how dynamics in these circuits shapes the integration and transmission of sensory information is not well understood. Here, we study neural coding in recurrently connected networks of neurons driven by sensory input. We show analytically how information available in the network output varies with the alignment between feedforward input and the integrating modes of the circuit dynamics. In light of this theory, we analyzed neural population activity in the visual cortex of mice that learned to discriminate visual features. We found that over learning, slow patterns of network dynamics realigned to better integrate input relevant to the discrimination task. This realignment of network dynamics could be explained by changes in excitatory-inhibitory connectivity among neurons tuned to relevant features. These results suggest that learning tunes the temporal dynamics of cortical circuits to optimally integrate relevant sensory input.

Coordinated multiplexing of information about separate objects in visual cortex.

Sensory receptive fields are large enough that they can contain more than one perceptible stimulus. How, then, can the brain encode information about each of the stimuli that may be present at a given moment? We recently showed that when more than one stimulus is present, single neurons can fluctuate between coding one vs. the other(s) across some time period, suggesting a form of neural multiplexing of different stimuli (Caruso et al., 2018). Here, we investigate (a) whether such coding fluctuations occur in early visual cortical areas; (b) how coding fluctuations are coordinated across the neural population; and (c) how coordinated coding fluctuations depend on the parsing of stimuli into separate vs. fused objects. We found coding fluctuations do occur in macaque V1 but only when the two stimuli form separate objects. Such separate objects evoked a novel pattern of V1 spike count ('noise') correlations involving distinct distributions of positive and negative values. This bimodal correlation pattern was most pronounced among pairs of neurons showing the strongest evidence for coding fluctuations or multiplexing. Whether a given pair of neurons exhibited positive or negative correlations depended on whether the two neurons both responded better to the same object or had different object preferences. Distinct distributions of spike count correlations based on stimulus preferences were also seen in V4 for separate objects but not when two stimuli fused to form one object. These findings suggest multiple objects evoke different response dynamics than those evoked by single stimuli, lending support to the multiplexing hypothesis and suggesting a means by which information about multiple objects can be preserved despite the apparent coarseness of sensory coding.

The spatial scale of somatostatin subnetworks increases from sensory to association cortex.

Incoming signals interact with rich, ongoing population activity dynamics in cortical circuits. These intrinsic dynamics are the consequence of interactions among local excitatory and inhibitory neurons and affect inter-region communication and information coding. It is unclear whether specializations in the patterns of interactions among excitatory and inhibitory neurons underlie systematic differences in activity dynamics across the cortex. Here, in mice, we compare the functional interactions among somatostatin (SOM)-expressing inhibitory interneurons and the rest of the neural population in auditory cortex (AC), a sensory region of the cortex, and posterior parietal cortex (PPC), an association region. The spatial structure of shared variability among SOM and non-SOM neurons differs across regions: correlations decay rapidly with distance in AC but not in PPC. However, in both regions, activity of SOM neurons is more highly correlated than non-SOM neurons' activity. Our results imply both generalization and specialization in the functional structure of inhibitory subnetworks across the cortex.

Sequential transmission of task-relevant information in cortical neuronal networks.

Cortical processing of task-relevant information enables recognition of behaviorally meaningful sensory events. It is unclear how task-related information is represented within cortical networks by the activity of individual neurons and their functional interactions. Here, we use two-photon imaging to record neuronal activity from the primary auditory cortex of mice during a pure-tone discrimination task. We find that a subset of neurons transiently encode sensory information used to inform behavioral choice. Using Granger causality analysis, we show that these neurons form functional networks in which information transmits sequentially. Network structures differ for target versus non-target tones, encode behavioral choice, and differ between correct versus incorrect behavioral choices. Correct behavioral choices are associated with shorter communication timescales, larger functional correlations, and greater information redundancy. In summary, specialized neurons in primary auditory cortex integrate task-related information and form functional networks whose structures encode both sensory input and behavioral choice.

A general decoding strategy explains the relationship between behavior and correlated variability.

Improvements in perception are frequently accompanied by decreases in correlated variability in sensory cortex. This relationship is puzzling because overall changes in correlated variability should minimally affect optimal information coding. We hypothesize that this relationship arises because instead of using optimal strategies for decoding the specific stimuli at hand, observers prioritize generality: a single set of neuronal weights to decode any stimuli. We tested this using a combination of multineuron recordings in the visual cortex of behaving rhesus monkeys and a cortical circuit model. We found that general decoders optimized for broad rather than narrow sets of visual stimuli better matched the animals' decoding strategy, and that their performance was more related to the magnitude of correlated variability. In conclusion, the inverse relationship between perceptual performance and correlated variability can be explained by observers using a general decoding strategy, capable of decoding neuronal responses to the variety of stimuli encountered in natural vision.

Noise Correlations for Faster and More Robust Learning.

Distributed population codes are ubiquitous in the brain and pose a challenge to downstream neurons that must learn an appropriate readout. Here we explore the possibility that this learning problem is simplified through inductive biases implemented by stimulus-independent noise correlations that constrain learning to task-relevant dimensions. We test this idea in a set of neural networks that learn to perform a perceptual discrimination task. Correlations among similarly tuned units were manipulated independently of an overall population signal-to-noise ratio to test how the format of stored information affects learning. Higher noise correlations among similarly tuned units led to faster and more robust learning, favoring homogenous weights assigned to neurons within a functionally similar pool, and could emerge through Hebbian learning. When multiple discriminations were learned simultaneously, noise correlations across relevant feature dimensions sped learning, whereas those across irrelevant feature dimensions slowed it. Our results complement the existing theory on noise correlations by demonstrating that when such correlations are produced without significant degradation of the signal-to-noise ratio, they can improve the speed of readout learning by constraining it to appropriate dimensions.SIGNIFICANCE STATEMENT Positive noise correlations between similarly tuned neurons theoretically reduce the representational capacity of the brain, yet they are commonly observed, emerge dynamically in complex tasks, and persist even in well-trained animals. Here we show that such correlations, when embedded in a neural population with a fixed signal-to-noise ratio, can improve the speed and robustness with which an appropriate readout is learned. In a simple discrimination task such correlations can emerge naturally through Hebbian learning. In more complex tasks that require multiple discriminations, correlations between neurons that similarly encode the task-relevant feature improve learning by constraining it to the appropriate task dimension.

Direct extraction of signal and noise correlations from two-photon calcium imaging of ensemble neuronal activity.

Neuronal activity correlations are key to understanding how populations of neurons collectively encode information. While two-photon calcium imaging has created a unique opportunity to record the activity of large populations of neurons, existing methods for inferring correlations from these data face several challenges. First, the observations of spiking activity produced by two-photon imaging are temporally blurred and noisy. Secondly, even if the spiking data were perfectly recovered via deconvolution, inferring network-level features from binary spiking data is a challenging task due to the non-linear relation of neuronal spiking to endogenous and exogenous inputs. In this work, we propose a methodology to explicitly model and directly estimate signal and noise correlations from two-photon fluorescence observations, without requiring intermediate spike deconvolution. We provide theoretical guarantees on the performance of the proposed estimator and demonstrate its utility through applications to simulated and experimentally recorded data from the mouse auditory cortex.

Enhanced Discriminative Abilities of Auditory Cortex Neurons for Pup Calls Despite Reduced Evoked Responses in C57BL/6 Mother Mice.

A fundamental task for the auditory system is to process communication sounds according to their behavioral significance. In many mammalian species, pup calls became more significant for mothers than other conspecific and heterospecific communication sounds. To study the cortical consequences of motherhood on the processing of communication sounds, we recorded neuronal responses in the primary auditory cortex of virgin and mother C57BL/6 mice which had similar ABR thresholds. In mothers, the evoked firing rate in response to pure tones was decreased and the frequency receptive fields were narrower. The responses to pup and adult calls were also reduced but the amount of mutual information (MI) per spike about the pup call's identity was increased in mother mice. The response latency to pup and adult calls was significantly shorter in mothers. Despite similarly decreased responses to guinea pig whistles, the response latency, and the MI per spike did not differ between virgins and mothers for these heterospecific vocalizations. Noise correlations between cortical recordings were decreased in mothers, suggesting that the firing rate of distant neurons was more independent from each other. Together, these results indicate that in the most commonly used mouse strain for behavioral studies, the discrimination of pup calls by auditory cortex neurons is more efficient during motherhood.

Pairwise Synchrony and Correlations Depend on the Structure of the Population Code in Visual Cortex.

In visual areas of primates, neurons activate in parallel while the animal is engaged in a behavioral task. In this study, we examine the structure of the population code while the animal performs delayed match-to-sample tasks on complex natural images. The macaque monkeys visualized two consecutive stimuli that were either the same or different, while being recorded with laminar arrays across the cortical depth in cortical areas V1 and V4. We decode correct choice behavior from neural populations of simultaneously recorded units. Utilizing decoding weights, we divide neurons into most informative and less informative and show that most informative neurons in V4, but not in V1, are more strongly synchronized, coupled, and correlated than less informative neurons. Because neurons are divided into two coding pools according to their coding preference, in V4, but not in V1, spiking synchrony, coupling, and correlations within the coding pool are stronger than across coding pools.

Inference of synaptic connectivity and external variability in neural microcircuits.

A major goal in neuroscience is to estimate neural connectivity from large scale extracellular recordings of neural activity in vivo. This is challenging in part because any such activity is modulated by the unmeasured external synaptic input to the network, known as the common input problem. Many different measures of functional connectivity have been proposed in the literature, but their direct relationship to synaptic connectivity is often assumed or ignored. For in vivo data, measurements of this relationship would require a knowledge of ground truth connectivity, which is nearly always unavailable. Instead, many studies use in silico simulations as benchmarks for investigation, but such approaches necessarily rely upon a variety of simplifying assumptions about the simulated network and can depend on numerous simulation parameters. We combine neuronal network simulations, mathematical analysis, and calcium imaging data to address the question of when and how functional connectivity, synaptic connectivity, and latent external input variability can be untangled. We show numerically and analytically that, even though the precision matrix of recorded spiking activity does not uniquely determine synaptic connectivity, it is in practice often closely related to synaptic connectivity. This relation becomes more pronounced when the spatial structure of neuronal variability is jointly considered.

Information-Limiting Correlations in Large Neural Populations.

Understanding the neural code requires understanding how populations of neurons code information. Theoretical models predict that information may be limited by correlated noise in large neural populations. Nevertheless, analyses based on tens of neurons have failed to find evidence of saturation. Moreover, some studies have shown that noise correlations can be very small, and therefore may not affect information coding. To determine whether information-limiting correlations exist, we implanted eight Utah arrays in prefrontal cortex (PFC; area 46) of two male macaque monkeys, recording >500 neurons simultaneously. We estimated information in PFC about saccades as a function of ensemble size. Noise correlations were, on average, small (∼10-3). However, information scaled strongly sublinearly with ensemble size. After shuffling trials, destroying noise correlations, information was a linear function of ensemble size. Thus, we provide evidence for the existence of information-limiting noise correlations in large populations of PFC neurons.SIGNIFICANCE STATEMENT Recent theoretical work has shown that even small correlations can limit information if they are "differential correlations," which are difficult to measure directly. However, they can be detected through decoding analyses on recordings from a large number of neurons over a large number of trials. We have achieved both by collecting neural activity in dorsal-lateral prefrontal cortex of macaques using eight microelectrode arrays (768 electrodes), from which we were able to compute accurate information estimates. We show, for the first time, strong evidence for information-limiting correlations. Despite pairwise correlations being small (on the order of 10-3), they affect information coding in populations on the order of 100 s of neurons.

The Effects of Population Tuning and Trial-by-Trial Variability on Information Encoding and Behavior.

Identifying the features of population responses that are relevant to the amount of information encoded by neuronal populations is a crucial step toward understanding population coding. Statistical features, such as tuning properties, individual and shared response variability, and global activity modulations, could all affect the amount of information encoded and modulate behavioral performance. We show that two features in particular affect information: the modulation of population responses across conditions (population signal) and the inverse population covariability along the modulation axis (projected precision). We demonstrate that fluctuations of these two quantities are correlated with fluctuations of behavioral performance in various tasks and brain regions consistently across 4 monkeys (1 female and 1 male Macaca mulatta; and 2 male Macaca fascicularis). In contrast, fluctuations in mean correlations among neurons and global activity have negligible or inconsistent effects on the amount of information encoded and behavioral performance. We also show that differential correlations reduce the amount of information encoded in finite populations by reducing projected precision. Our results are consistent with predictions of a model that optimally decodes population responses to produce behavior.SIGNIFICANCE STATEMENT The last two or three decades of research have seen hot debates about what features of population tuning and trial-by-trial variability influence the information carried by a population of neurons, with some camps arguing, for instance, that mean pairwise correlations or global fluctuations are important while other camps report opposite results. In this study, we identify the most important features of neural population responses that determine the amount of encoded information and behavioral performance by combining analytic calculations with a novel nonparametric method that allows us to isolate the effects of different statistical features. We tested our hypothesis on 4 macaques, three decision-making tasks, and two brain areas. The predictions of our theory were in agreement with the experimental data.

Correlation Transfer by Layer 5 Cortical Neurons Under Recreated Synaptic Inputs In Vitro.

Correlated electrical activity in neurons is a prominent characteristic of cortical microcircuits. Despite a growing amount of evidence concerning both spike-count and subthreshold membrane potential pairwise correlations, little is known about how different types of cortical neurons convert correlated inputs into correlated outputs. We studied pyramidal neurons and two classes of GABAergic interneurons of layer 5 in neocortical brain slices obtained from rats of both sexes, and we stimulated them with biophysically realistic correlated inputs, generated using dynamic clamp. We found that the physiological differences between cell types manifested unique features in their capacity to transfer correlated inputs. We used linear response theory and computational modeling to gain clear insights into how cellular properties determine both the gain and timescale of correlation transfer, thus tying single-cell features with network interactions. Our results provide further ground for the functionally distinct roles played by various types of neuronal cells in the cortical microcircuit.SIGNIFICANCE STATEMENT No matter how we probe the brain, we find correlated neuronal activity over a variety of spatial and temporal scales. For the cerebral cortex, significant evidence has accumulated on trial-to-trial covariability in synaptic inputs activation, subthreshold membrane potential fluctuations, and output spike trains. Although we do not yet fully understand their origin and whether they are detrimental or beneficial for information processing, we believe that clarifying how correlations emerge is pivotal for understanding large-scale neuronal network dynamics and computation. Here, we report quantitative differences between excitatory and inhibitory cells, as they relay input correlations into output correlations. We explain this heterogeneity by simple biophysical models and provide the most experimentally validated test of a theory for the emergence of correlations.

Leveraging Nonhuman Primate Multisensory Neurons and Circuits in Assessing Consciousness Theory.

Both the global neuronal workspace (GNW) and integrated information theory (IIT) posit that highly complex and interconnected networks engender perceptual awareness. GNW specifies that activity recruiting frontoparietal networks will elicit a subjective experience, whereas IIT is more concerned with the functional architecture of networks than with activity within it. Here, we argue that according to IIT mathematics, circuits converging on integrative versus convergent yet non-integrative neurons should support a greater degree of consciousness. We test this hypothesis by analyzing a dataset of neuronal responses collected simultaneously from primary somatosensory cortex (S1) and ventral premotor cortex (vPM) in nonhuman primates presented with auditory, tactile, and audio-tactile stimuli as they are progressively anesthetized with propofol. We first describe the multisensory (audio-tactile) characteristics of S1 and vPM neurons (mean and dispersion tendencies, as well as noise-correlations), and functionally label these neurons as convergent or integrative according to their spiking responses. Then, we characterize how these different pools of neurons behave as a function of consciousness. At odds with the IIT mathematics, results suggest that convergent neurons more readily exhibit properties of consciousness (neural complexity and noise correlation) and are more impacted during the loss of consciousness than integrative neurons. Last, we provide support for the GNW by showing that neural ignition (i.e., same trial coactivation of S1 and vPM) was more frequent in conscious than unconscious states. Overall, we contrast GNW and IIT within the same single-unit activity dataset, and support the GNW.SIGNIFICANCE STATEMENT A number of prominent theories of consciousness exist, and a number of these share strong commonalities, such as the central role they ascribe to integration. Despite the important and far reaching consequences developing a better understanding of consciousness promises to bring, for instance in diagnosing disorders of consciousness (e.g., coma, vegetative-state, locked-in syndrome), these theories are seldom tested via invasive techniques (with high signal-to-noise ratios), and never directly confronted within a single dataset. Here, we first derive concrete and testable predictions from the global neuronal workspace and integrated information theory of consciousness. Then, we put these to the test by functionally labeling specific neurons as either convergent or integrative nodes, and examining the response of these neurons during anesthetic-induced loss of consciousness.

Local and Global Influences of Visual Spatial Selection and Locomotion in Mouse Primary Visual Cortex.

Sensory selection and movement locally and globally modulate neural responses in seemingly similar ways. For example, locomotion enhances visual responses in mouse primary visual cortex (V1), resembling the effects of spatial attention on primate visual cortical activity. However, interactions between these local and global mechanisms and the resulting effects on perceptual behavior remain largely unknown. Here, we describe a novel mouse visual spatial selection task in which animals either monitor one of two locations for a contrast change ("selective mice") or monitor both ("non-selective mice") and can run at will. Selective mice perform well only when their selected stimulus changes, giving rise to local electrophysiological changes in the corresponding hemisphere of V1 including decreased noise correlations and increased visual information. Non-selective mice perform well when either stimulus changes, giving rise to global changes across both hemispheres of V1. During locomotion, selective mice have worse behavioral performance, increased noise correlations in V1, and decreased visual information, while non-selective mice have decreased noise correlations in V1 but no change in performance or visual information. Our findings demonstrate that mice can locally or globally enhance visual information, but the interaction of the global effect of locomotion with local selection impairs behavioral performance. Moving forward, this mouse model will facilitate future studies of local and global sensory modulatory mechanisms and their effects on behavior.

Assessing the Impacts of Correlated Variability with Dissociated Timescales.

Despite the profound influence on coding capacity of sensory neurons, the measurements of noise correlations have been inconsistent. This is, possibly, because nonstationarity, i.e., drifting baselines, engendered the spurious long-term correlations even if no actual short-term correlation existed. Although attempts to separate them have been made previously, they were ad hoc for specific cases or computationally too demanding. Here we proposed an information-geometric method to unbiasedly estimate pure short-term noise correlations irrespective of the background brain activities without demanding computational resources. First, the benchmark simulations demonstrated that the proposed estimator is more accurate and computationally efficient than the conventional correlograms and the residual correlations with Kalman filters or moving averages of length three or more, while the best moving average of length two coincided with the propose method regarding correlation estimates. Next, we analyzed the cat V1 neural responses to demonstrate that the statistical test accompanying the proposed method combined with the existing nonstationarity test enabled us to dissociate short-term and long-term noise correlations. When we excluded the spurious noise correlations of purely long-term nature, only a small fraction of neuron pairs showed significant short-term correlations, possibly reconciling the previous inconsistent observations on existence of significant noise correlations. The decoding accuracy was slightly improved by the short-term correlations. Although the long-term correlations deteriorated the generalizability, the generalizability was recovered by the decoder with trend removal, suggesting that brains could overcome nonstationarity. Thus, the proposed method enables us to elucidate the impacts of short-term and long-term noise correlations in a dissociated manner.