Mechanisms of pattern decorrelation by recurrent neuronal circuits

Development and analysis of an artificial olfactory bulb

This article presents the development of an artificial olfactory bulb (OB) using an electronic nose with thermally modulated metal-oxide sensors. Inspired by animal OBs, our approach employs thermal modulation to replicate the spatial encoding patterns of glomeruli clusters and subclusters. This new approach enhances the classification capabilities of traditional electronic noses and offers new insights for biomimetic olfaction. Molecular receptive range (MRR) analysis confirms that our artificial OB effectively mimics the glomerular distribution of animal OBs. Additionally, the incorporation of a short axon cell (SAC) network, inspired by the animal olfactory system, significantly improves lifetime sparseness and qualitative ability of the artificial OB through extensive lateral inhibition, providing a theoretical framework for enhanced olfactory performance.

Gradient-free training of recurrent neural networks using random perturbations

Recurrent neural networks (RNNs) hold immense potential for computations due to their Turing completeness and sequential processing capabilities, yet existing methods for their training encounter efficiency challenges. Backpropagation through time (BPTT), the prevailing method, extends the backpropagation (BP) algorithm by unrolling the RNN over time. However, this approach suffers from significant drawbacks, including the need to interleave forward and backward phases and store exact gradient information. Furthermore, BPTT has been shown to struggle to propagate gradient information for long sequences, leading to vanishing gradients. An alternative strategy to using gradient-based methods like BPTT involves stochastically approximating gradients through perturbation-based methods. This learning approach is exceptionally simple, necessitating only forward passes in the network and a global reinforcement signal as feedback. Despite its simplicity, the random nature of its updates typically leads to inefficient optimization, limiting its effectiveness in training neural networks. In this study, we present a new approach to perturbation-based learning in RNNs whose performance is competitive with BPTT, while maintaining the inherent advantages over gradient-based learning. To this end, we extend the recently introduced activity-based node perturbation (ANP) method to operate in the time domain, leading to more efficient learning and generalization. We subsequently conduct a range of experiments to validate our approach. Our results show similar performance, convergence time and scalability when compared to BPTT, strongly outperforming standard node perturbation and weight perturbation methods. These findings suggest that perturbation-based learning methods offer a versatile alternative to gradient-based methods for training RNNs which can be ideally suited for neuromorphic computing applications.

Common principles for odour coding across vertebrates and invertebrates

The olfactory system is an ideal and tractable system for exploring how the brain transforms sensory inputs into behaviour. The basic tasks of any olfactory system include odour detection, discrimination and categorization. The challenge for the olfactory system is to transform the high-dimensional space of olfactory stimuli into the much smaller space of perceived objects and valence that endows odours with meaning. Our current understanding of how neural circuits address this challenge has come primarily from observations of the mechanisms of the brain for processing other sensory modalities, such as vision and hearing, in which optimized deep hierarchical circuits are used to extract sensory features that vary along continuous physical dimensions. The olfactory system, by contrast, contends with an ill-defined, high-dimensional stimulus space and discrete stimuli using a circuit architecture that is shallow and parallelized. Here, we present recent observations in vertebrate and invertebrate systems that relate the statistical structure and state-dependent modulation of olfactory codes to mechanisms of perception and odour-guided behaviour.

Adult Neurogenesis Reconciles Flexibility and Stability of Olfactory Perceptual Memory

In brain regions featuring ongoing plasticity, the task of quickly encoding new information without overwriting old memories presents a significant challenge. In the rodent olfactory bulb, which is renowned for substantial structural plasticity driven by adult neurogenesis and persistent turnover of dendritic spines, we show that such plasticity is vital to overcoming this flexibility-stability dilemma. To do so, we develop a computational model for structural plasticity in the olfactory bulb and show that the maturation of adult-born neurons facilitates the abilities to learn quickly and forget slowly. Particularly important to achieve this goal are the transient enhancement of the plasticity, excitability, and susceptibility to apoptosis that characterizes young neurons. The model captures many experimental observations and makes a number of testable predictions. Overall, it identifies memory consolidation as an important role of adult neurogenesis in olfaction and exemplifies how the brain can maintain stable memories despite ongoing extensive plasticity.

Distinct information conveyed to the olfactory bulb by feedforward input from the nose and feedback from the cortex

Sensory systems are organized hierarchically, but feedback projections frequently disrupt this order. In the olfactory bulb (OB), cortical feedback projections numerically match sensory inputs. To unravel information carried by these two streams, we imaged the activity of olfactory sensory neurons (OSNs) and cortical axons in the mouse OB using calcium indicators, multiphoton microscopy, and diverse olfactory stimuli. Here, we show that odorant mixtures of increasing complexity evoke progressively denser OSN activity, yet cortical feedback activity is of similar sparsity for all stimuli. Also, representations of complex mixtures are similar in OSNs but are decorrelated in cortical axons. While OSN responses to increasing odorant concentrations exhibit a sigmoidal relationship, cortical axonal responses are complex and nonmonotonic, which can be explained by a model with activity-dependent feedback inhibition in the cortex. Our study indicates that early-stage olfactory circuits have access to local feedforward signals and global, efficiently formatted information about odor scenes through cortical feedback.

Structure, biophysics, and circuit function of a "giant" cortical presynaptic terminal

The hippocampal mossy fiber synapse, formed between axons of dentate gyrus granule cells and dendrites of CA3 pyramidal neurons, is a key synapse in the trisynaptic circuitry of the hippocampus. Because of its comparatively large size, this synapse is accessible to direct presynaptic recording, allowing a rigorous investigation of the biophysical mechanisms of synaptic transmission and plasticity. Furthermore, because of its placement in the very center of the hippocampal memory circuit, this synapse seems to be critically involved in several higher network functions, such as learning, memory, pattern separation, and pattern completion. Recent work based on new technologies in both nanoanatomy and nanophysiology, including presynaptic patch-clamp recording, paired recording, super-resolution light microscopy, and freeze-fracture and "flash-and-freeze" electron microscopy, has provided new insights into the structure, biophysics, and network function of this intriguing synapse. This brings us one step closer to answering a fundamental question in neuroscience: how basic synaptic properties shape higher network computations.

Hopfield-like network with complementary encodings of memories

We present a Hopfield-like autoassociative network for memories representing examples of concepts. Each memory is encoded by two activity patterns with complementary properties. The first is dense and correlated across examples within concepts, and the second is sparse and exhibits no correlation among examples. The network stores each memory as a linear combination of its encodings. During retrieval, the network recovers sparse or dense patterns with a high or low activity threshold, respectively. As more memories are stored, the dense representation at low threshold shifts from examples to concepts, which are learned from accumulating common example features. Meanwhile, the sparse representation at high threshold maintains distinctions between examples due to the high capacity of sparse, decorrelated patterns. Thus, a single network can retrieve memories at both example and concept scales and perform heteroassociation between them. We obtain our results by deriving macroscopic mean-field equations that yield capacity formulas for sparse examples, dense examples, and dense concepts. We also perform simulations that verify our theoretical results and explicitly demonstrate the capabilities of the network.

Fast updating feedback from piriform cortex to the olfactory bulb relays multimodal reward contingency signals during rule-reversal

While animals readily adjust their behavior to adapt to relevant changes in the environment, the neural pathways enabling these changes remain largely unknown. Here, using multiphoton imaging, we investigated whether feedback from the piriform cortex to the olfactory bulb supports such behavioral flexibility. To this end, we engaged head-fixed mice in a multimodal rule-reversal task guided by olfactory and auditory cues. Both odor and, surprisingly, the sound cues triggered cortical bulbar feedback responses which preceded the behavioral report. Responses to the same sensory cue were strongly modulated upon changes in stimulus-reward contingency (rule reversals). The re-shaping of individual bouton responses occurred within seconds of the rule-reversal events and was correlated with changes in the behavior. Optogenetic perturbation of cortical feedback within the bulb disrupted the behavioral performance. Our results indicate that the piriform-to-olfactory bulb feedback carries reward contingency signals and is rapidly re-formatted according to changes in the behavioral context.

α2-Adrenergic modulation of Ih in adult-born granule cells in the olfactory bulb

In the olfactory bulb (OB), a large population of axon-less inhibitory interneurons, the granule cells (GCs), coordinate network activity and tune the output of principal neurons, the mitral and tufted cells (MCs), through dendrodendritic interactions. Furthermore, GCs undergo neurogenesis throughout life, providing a source of plasticity to the neural network of the OB. The function and integration of GCs in the OB are regulated by several afferent neuromodulatory signals, including noradrenaline (NA), a state-dependent neuromodulator that plays a crucial role in the regulation of cortical function and task-specific decision processes. However, the mechanisms by which NA regulates GC function are not fully understood. Here, we show that NA modulates hyperpolarization-activated currents (I_h) via the activation of α₂-adrenergic receptors (ARs) in adult-born GCs (abGCs), thus directly acting on channels that play essential roles in regulating neuronal excitability and network oscillations in the brain. This modulation affects the dendrodendritic output of GCs leading to an enhancement of lateral inhibition onto the MCs. Furthermore, we show that NA modulates subthreshold resonance in GCs, which could affect the temporal integration of abGCs. Together, these results provide a novel mechanism by which a state-dependent neuromodulator acting on I_h can regulate GC function in the OB.

Cerebellum as a kernel machine: A novel perspective on expansion recoding in granule cell layer

Sensorimotor information provided by mossy fibers (MF) is mapped to high-dimensional space by a huge number of granule cells (GrC) in the cerebellar cortex's input layer. Significant studies have demonstrated the computational advantages and primary contributor of this expansion recoding. Here, we propose a novel perspective on the expansion recoding where each GrC serve as a kernel basis function, thereby the cerebellum can operate like a kernel machine that implicitly use high dimensional (even infinite) feature spaces. We highlight that the generation of kernel basis function is indeed biologically plausible scenario, considering that the key idea of kernel machine is to memorize important input patterns. We present potential regimes for developing kernels under constrained resources and discuss the advantages and disadvantages of each regime using various simulation settings.

Structural spine plasticity: Learning and forgetting of odor-specific subnetworks in the olfactory bulb

Learning to discriminate between different sensory stimuli is essential for survival. In rodents, the olfactory bulb, which contributes to odor discrimination via pattern separation, exhibits extensive structural synaptic plasticity involving the formation and removal of synaptic spines, even in adult animals. The network connectivity resulting from this plasticity is still poorly understood. To gain insight into this connectivity we present here a computational model for the structural plasticity of the reciprocal synapses between the dominant population of excitatory principal neurons and inhibitory interneurons. It incorporates the observed modulation of spine stability by odor exposure. The model captures the striking experimental observation that the exposure to odors does not always enhance their discriminability: while training with similar odors enhanced their discriminability, training with dissimilar odors actually reduced the discriminability of the training stimuli. Strikingly, this differential learning does not require the activity-dependence of the spine stability and occurs also in a model with purely random spine dynamics in which the spine density is changed homogeneously, e.g., due to a global signal. However, the experimentally observed odor-specific reduction in the response of principal cells as a result of extended odor exposure and the concurrent disinhibition of a subset of principal cells arise only in the activity-dependent model. Moreover, this model predicts the experimentally testable recovery of odor response through weak but not through strong odor re-exposure and the forgetting of odors via exposure to interfering odors. Combined with the experimental observations, the computational model provides strong support for the prediction that odor exposure leads to the formation of odor-specific subnetworks in the olfactory bulb.

A key feature of the brain is its ability to learn through the plasticity of its network. The olfactory bulb in the olfactory system is a remarkable brain area whose anatomical structure evolves substantially still in adult animals by establishing new synaptic connections and removing existing ones. We present a computational model for this process and employ it to interpret recent experimental results. By comparing the results of our model with those of a random control model we identify various experimental observations that lend strong support to the notion that the network of the olfactory bulb comprises learned, odor-specific subnetworks. Moreover, our model explains the recent observation that the learning of odors does not always improve their discriminability and provides testable predictions for the recovery of odor response after repeated odor exposure and for when the learning of new odors interferes with retaining the memory of familiar odors.

Disorder and the Neural Representation of Complex Odors

Animals smelling in the real world use a small number of receptors to sense a vast number of natural molecular mixtures, and proceed to learn arbitrary associations between odors and valences. Here, we propose how the architecture of olfactory circuits leverages disorder, diffuse sensing and redundancy in representation to meet these immense complementary challenges. First, the diffuse and disordered binding of receptors to many molecules compresses a vast but sparsely-structured odor space into a small receptor space, yielding an odor code that preserves similarity in a precise sense. Introducing any order/structure in the sensing degrades similarity preservation. Next, lateral interactions further reduce the correlation present in the low-dimensional receptor code. Finally, expansive disordered projections from the periphery to the central brain reconfigure the densely packed information into a high-dimensional representation, which contains multiple redundant subsets from which downstream neurons can learn flexible associations and valences. Moreover, introducing any order in the expansive projections degrades the ability to recall the learned associations in the presence of noise. We test our theory empirically using data from Drosophila. Our theory suggests that the neural processing of sparse but high-dimensional olfactory information differs from the other senses in its fundamental use of disorder.

Hierarchical Elemental Odor Coding for Fine Discrimination Between Enantiomer Odors or Cancer-Characteristic Odors

Odors trigger various emotional responses such as fear of predator odors, aversion to disease or cancer odors, attraction to male/female odors, and appetitive behavior to delicious food odors. Odor information processing for fine odor discrimination, however, has remained difficult to address. The olfaction and color vision share common features that G protein-coupled receptors are the remote sensors. As different orange colors can be discriminated by distinct intensity ratios of elemental colors, such as yellow and red, odors are likely perceived as multiple elemental odors hierarchically that the intensities of elemental odors are in order of dominance. For example, in a mixture of rose and fox-unique predator odors, robust rose odor alleviates the fear of mice to predator odors. Moreover, although occult blood odor is stronger than bladder cancer-characteristic odor in urine samples, sniffer mice can discriminate bladder cancer odor in occult blood-positive urine samples. In forced-choice odor discrimination tasks for pairs of enantiomers or pairs of body odors vs. cancer-induced body odor disorders, sniffer mice discriminated against learned olfactory cues in a wide range of concentrations, where correct choice rates decreased in the Fechner's law, as perceptual ambiguity increased. In this mini-review, we summarize the current knowledge of how the olfactory system encodes and hierarchically decodes multiple elemental odors to control odor-driven behaviors.

Connectivity and dynamics in the olfactory bulb

Dendrodendritic interactions between excitatory mitral cells and inhibitory granule cells in the olfactory bulb create a dense interaction network, reorganizing sensory representations of odors and, consequently, perception. Large-scale computational models are needed for revealing how the collective behavior of this network emerges from its global architecture. We propose an approach where we summarize anatomical information through dendritic geometry and density distributions which we use to calculate the connection probability between mitral and granule cells, while capturing activity patterns of each cell type in the neural dynamical systems theory of Izhikevich. In this way, we generate an efficient, anatomically and physiologically realistic large-scale model of the olfactory bulb network. Our model reproduces known connectivity between sister vs. non-sister mitral cells; measured patterns of lateral inhibition; and theta, beta, and gamma oscillations. The model in turn predicts testable relationships between network structure and several functional properties, including lateral inhibition, odor pattern decorrelation, and LFP oscillation frequency. We use the model to explore the influence of cortex on the olfactory bulb, demonstrating possible mechanisms by which cortical feedback to mitral cells or granule cells can influence bulbar activity, as well as how neurogenesis can improve bulbar decorrelation without requiring cell death. Our methodology provides a tractable tool for other researchers.

How connectivity rules and synaptic properties shape the efficacy of pattern separation in the entorhinal cortex-dentate gyrus-CA3 network

Pattern separation is a fundamental brain computation that converts small differences in input patterns into large differences in output patterns. Several synaptic mechanisms of pattern separation have been proposed, including code expansion, inhibition and plasticity; however, which of these mechanisms play a role in the entorhinal cortex (EC)-dentate gyrus (DG)-CA3 circuit, a classical pattern separation circuit, remains unclear. Here we show that a biologically realistic, full-scale EC-DG-CA3 circuit model, including granule cells (GCs) and parvalbumin-positive inhibitory interneurons (PV+-INs) in the DG, is an efficient pattern separator. Both external gamma-modulated inhibition and internal lateral inhibition mediated by PV+-INs substantially contributed to pattern separation. Both local connectivity and fast signaling at GC-PV+-IN synapses were important for maximum effectiveness. Similarly, mossy fiber synapses with conditional detonator properties contributed to pattern separation. By contrast, perforant path synapses with Hebbian synaptic plasticity and direct EC-CA3 connection shifted the network towards pattern completion. Our results demonstrate that the specific properties of cells and synapses optimize higher-order computations in biological networks and might be useful to improve the deep learning capabilities of technical networks.

Principles of odor coding in vertebrates and artificial chemosensory systems

The biological olfactory system is the sensory system responsible for the detection of the chemical composition of the environment. Several attempts to mimic biological olfactory systems have led to various artificial olfactory systems using different technical approaches. Here we provide a parallel description of biological olfactory systems and their technical counterparts. We start with a presentation of the input to the systems, the stimuli, and treat the interface between the external world and the environment where receptor neurons or artificial chemosensors reside. We then delineate the functions of receptor neurons and chemosensors as well as their overall I-O relationships. Up to this point, our account of the systems goes along similar lines. The next processing steps differ considerably: while in biology the processing step following the receptor neurons is the "integration" and "processing" of receptor neuron outputs in the olfactory bulb, this step has various realizations in electronic noses. For a long period of time, the signal processing stages beyond the olfactory bulb, i.e., the higher olfactory centers were little studied. Only recently there has been a marked growth of studies tackling the information processing in these centers. In electronic noses, a third stage of processing has virtually never been considered. In this review, we provide an up-to-date overview of the current knowledge of both fields and, for the first time, attempt to tie them together. We hope it will be a breeding ground for better information, communication, and data exchange between very related but so far little connected fields.

Ion-channel regulation of response decorrelation in a heterogeneous multi-scale model of the dentate gyrus

Heterogeneities in biological neural circuits manifest in afferent connectivity as well as in local-circuit components such as neuronal excitability, neural structure and local synaptic strengths. The expression of adult neurogenesis in the dentate gyrus (DG) amplifies local-circuit heterogeneities and guides heterogeneities in afferent connectivity. How do neurons and their networks endowed with these distinct forms of heterogeneities respond to perturbations to individual ion channels, which are known to change under several physiological and pathophysiological conditions? We sequentially traversed the ion channels-neurons-network scales and assessed the impact of eliminating individual ion channels on conductance-based neuronal and network models endowed with disparate local-circuit and afferent heterogeneities. We found that many ion channels differentially contributed to specific neuronal or network measurements, and the elimination of any given ion channel altered several functional measurements. We then quantified the impact of ion-channel elimination on response decorrelation, a well-established metric to assess the ability of neurons in a network to convey complementary information, in DG networks endowed with different forms of heterogeneities. Notably, we found that networks constructed with structurally immature neurons exhibited functional robustness, manifesting as minimal changes in response decorrelation in the face of ion-channel elimination. Importantly, the average change in output correlation was dependent on the eliminated ion channel but invariant to input correlation. Our analyses suggest that neurogenesis-driven structural heterogeneities could assist the DG network in providing functional resilience to molecular perturbations.

•

Perturbations at one scale result in a cascading impact on physiology across scales.

•

Heterogeneous multi-scale models used to assess the impact of ion-channel deletion.

•

Mapping of structural components to functional outcomes is many-to-many.

•

Differential & variable impact of ion channel deletion on response decorrelation.

•

Neurogenesis-induced structural heterogeneity confers resilience to perturbations.

Effect of Interglomerular Inhibitory Networks on Olfactory Bulb Odor Representations

Lateral inhibition is a fundamental feature of circuits that process sensory information. In the mammalian olfactory system, inhibitory interneurons called short axon cells (SACs) comprise the first network mediating lateral inhibition between glomeruli, the functional units of early olfactory coding and processing. The connectivity of this network and its impact on odor representations is not well understood. To explore this question, we constructed a computational model of the interglomerular inhibitory network using detailed characterizations of SAC morphologies taken from mouse olfactory bulb (OB). We then examined how this network transformed glomerular patterns of odorant-evoked sensory input (taken from previously-published datasets) as a function of the selectivity of interglomerular inhibition. We examined three connectivity schemes: selective (each glomerulus connects to few others with heterogeneous strength), nonselective (glomeruli connect to most others with heterogenous strength), or global (glomeruli connect to all others with equal strength). We found that both selective and nonselective interglomerular networks could mediate heterogeneous patterns of inhibition across glomeruli when driven by realistic sensory input patterns, but that global inhibitory networks were unable to produce input-output transformations that matched experimental data and were poor mediators of intensity-dependent gain control. We further found that networks whose interglomerular connectivities were tuned by sensory input profile decorrelated odor representations moreeffectively. These results suggest that, despite their multiglomerular innervation patterns, SACs are capable of mediating odorant-specific patterns of inhibition between glomeruli that could, theoretically, be tuned by experience or evolution to optimize discrimination of particular odorants.SIGNIFICANCE STATEMENT Lateral inhibition is a key feature of circuitry in many sensory systems including vision, audition, and olfaction. We investigate how lateral inhibitory networks mediated by short axon cells (SACs) in the mouse olfactory bulb (OB) might shape odor representations as a function of their interglomerular connectivity. Using a computational model of interglomerular connectivity derived from experimental data, we find that SAC networks, despite their broad innervation patterns, can mediate heterogeneous patterns of inhibition across glomeruli, and that the canonical model of global inhibition does not generate experimentally observed responses to stimuli. In addition, inhibitory connections tuned by input statistics yield enhanced decorrelation of similar input patterns. These results elucidate how the organization of inhibition between neural elements may affect computations.

Effect of Circuit Structure on Odor Representation in the Insect Olfactory System

In neuroscience, the structure of a circuit has often been used to intuit function—an inversion of Louis Kahn’s famous dictum, “Form follows function” (Kristan and Katz, 2006). However, different brain networks may use different network architectures to solve the same problem. The olfactory circuits of two insects, the locust, Schistocerca americana, and the fruit fly, Drosophila melanogaster, serve the same function—to identify and discriminate odors. The neural circuitry that achieves this shows marked structural differences. Projection neurons (PNs) in the antennal lobe innervate Kenyon cells (KCs) of the mushroom body. In locust, each KC receives inputs from ∼50% of PNs, a scheme that maximizes the difference between inputs to any two of ∼50,000 KCs. In contrast, in Drosophila, this number is only 5% and appears suboptimal. Using a computational model of the olfactory system, we show that the activity of KCs is sufficiently high-dimensional that it can separate similar odors regardless of the divergence of PN–KC connections. However, when temporal patterning encodes odor attributes, dense connectivity outperforms sparse connections. Increased separability comes at the cost of reliability. The disadvantage of sparse connectivity can be mitigated by incorporating other aspects of circuit architecture seen in Drosophila. Our simulations predict that Drosophila and locust circuits lie at different ends of a continuum where the Drosophila gives up on the ability to resolve similar odors to generalize across varying environments, while the locust separates odor representations but risks misclassifying noisy variants of the same odor.

Effect of Stimulus-Dependent Spike Timing on Population Coding of Sound Location in the Owl's Auditory Midbrain

In the auditory system, the spectrotemporal structure of acoustic signals determines the temporal pattern of spikes. Here, we investigated this effect in neurons of the barn owl's auditory midbrain (Tyto furcata) that are selective for auditory space and whether it can influence the coding of sound direction. We found that in the nucleus where neurons first become selective to combinations of sound localization cues, reproducibility of spike trains across repeated trials of identical sounds, a metric of across-trial temporal fidelity of spiking patterns evoked by a stimulus, was maximal at the sound direction that elicited the highest firing rate. We then tested the hypothesis that this stimulus-dependent patterning resulted in rate co-modulation of cells with similar frequency and spatial selectivity, driving stimulus-dependent synchrony of population responses. Tetrodes were used to simultaneously record multiple nearby units in the optic tectum (OT), where auditory space is topographically represented. While spiking of neurons in OT showed lower reproducibility across trials compared with upstream nuclei, spike-time synchrony between nearby OT neurons was highest for sounds at their preferred direction. A model of the midbrain circuit explained the relationship between stimulus-dependent reproducibility and synchrony, and demonstrated that this effect can improve the decoding of sound location from the OT output. Thus, stimulus-dependent spiking patterns in the auditory midbrain can have an effect on spatial coding. This study reports a functional connection between spike patterning elicited by spectrotemporal features of a sound and the coding of its location.

Quantitative properties of a feedback circuit predict frequency-dependent pattern separation

Feedback inhibitory motifs are thought to be important for pattern separation across species. How feedback circuits may implement pattern separation of biologically plausible, temporally structured input in mammals is, however, poorly understood. We have quantitatively determined key properties of netfeedback inhibition in the mouse dentate gyrus, a region critically involved in pattern separation. Feedback inhibition is recruited steeply with a low dynamic range (0% to 4% of active GCs), and with a non-uniform spatial profile. Additionally, net feedback inhibition shows frequency-dependent facilitation, driven by strongly facilitating mossy fiber inputs. Computational analyses show a significant contribution of the feedback circuit to pattern separation of theta modulated inputs, even within individual theta cycles. Moreover, pattern separation was selectively boosted at gamma frequencies, in particular for highly similar inputs. This effect was highly robust, suggesting that frequency-dependent pattern separation is a key feature of the feedback inhibitory microcircuit.

You can probably recall where you left your car this morning without too much trouble. But assuming you use the same busy parking lot every day, can you remember which space you parked in yesterday? Or the day before that? Most people find this difficult not because they cannot remember what happened two or three days ago, but because it requires distinguishing between very similar memories. The car, the parking lot, and the time of day were the same on each occasion. So how do you remember where you parked this morning?

This ability to distinguish between memories of similar events depends on a brain region called the hippocampus. A subregion of the hippocampus called the dentate gyrus generates different patterns of activity in response to events that are similar but distinct. This process is called pattern separation, and it helps ensure that you do not look for your car in yesterday’s parking space.

Pattern separation in the dentate gyrus is thought to involve a form of negative feedback called feedback inhibition, a phenomenon where the output of a process acts to limit or stop the same process. To test this idea, Braganza et al. studied feedback inhibition in the dentate gyrus of mice, before building a computer model simulating the inhibition process and supplying the model with two types of realistic input. The first consisted of low-frequency theta brainwaves, which occur, for instance, in the dentate gyrus when animals explore their environment. The second consisted of higher frequency gamma brainwaves, which occur, for example, when animals experience something new.

Testing the model showed that feedback inhibition contributes to pattern separation with both theta and gamma inputs. However, pattern separation is stronger with gamma input. This suggests that high frequency brainwaves in the hippocampus could help animals distinguish new events from old ones by promoting pattern separation.

Various brain disorders, including Alzheimer’s disease, schizophrenia and epilepsy, involve changes in the dentate gyrus and altered brain rhythms. The current findings could help reveal how these changes contribute to memory impairments and to a reduced ability to distinguish similar experiences.

Plasticity in olfactory bulb circuits

Olfaction is crucial for animal survival and human well-being. The olfactory bulb is the obligatory input station for olfactory information. In contrast to the traditional view as a static relay station, recent evidence indicates that the olfactory bulb dynamically processes olfactory information in an experience-dependent and context-dependent manner. Here, we review recent studies on experience-dependent plasticity of the main circuit components within the olfactory bulb of rodents. We argue that the olfactory bulb plasticity allows optimal representations of behaviorally-relevant odors in the continuously changing olfactory environment.

Optimality of sparse olfactory representations is not affected by network plasticity

The neural representation of a stimulus is repeatedly transformed as it moves from the sensory periphery to deeper layers of the nervous system. Sparsening transformations are thought to increase the separation between similar representations, encode stimuli with great specificity, maximize storage capacity of associative memories, and provide an energy efficient instantiation of information in neural circuits. In the insect olfactory system, odors are initially represented in the periphery as a combinatorial code with relatively simple temporal dynamics. Subsequently, in the antennal lobe this representation is transformed into a dense and complex spatiotemporal activity pattern. Next, in the mushroom body Kenyon cells (KCs), the representation is dramatically sparsened. Finally, in mushroom body output neurons (MBONs), the representation takes on a new dense spatiotemporal format. Here, we develop a computational model to simulate this chain of olfactory processing from the receptor neurons to MBONs. We demonstrate that representations of similar odorants are maximally separated, measured by the distance between the corresponding MBON activity vectors, when KC responses are sparse. Sparseness is maintained across variations in odor concentration by adjusting the feedback inhibition that KCs receive from an inhibitory neuron, the Giant GABAergic neuron. Different odor concentrations require different strength and timing of feedback inhibition for optimal processing. Importantly, as observed in vivo, the KC–MBON synapse is highly plastic, and, therefore, changes in synaptic strength after learning can change the balance of excitation and inhibition, potentially leading to changes in the distance between MBON activity vectors of two odorants for the same level of KC population sparseness. Thus, what is an optimal degree of sparseness before odor learning, could be rendered sub–optimal post learning. Here, we show, however, that synaptic weight changes caused by spike timing dependent plasticity increase the distance between the odor representations from the perspective of MBONs. A level of sparseness that was optimal before learning remains optimal post-learning.

Kenyon cells (KCs) of the mushroom body represent odors as a sparse code. When viewed from the perspective of follower neurons, mushroom body output neurons (MBONs) reveal an optimal level of coding sparseness that maximally separates the representations of odors. However, the KC–MBON synapse is highly plastic and may be potentiated or depressed by odor–driven experience that could, in turn, disrupt the optimality formed by pre–synaptic circuits. Contrary to this expectation, we show that synaptic plasticity based on spike timing of pre- and postsynaptic neurons improves the ability of the system to distinguish between the representations of similar odors while preserving the optimality determined by pre–synaptic circuits.

Whitening of odor representations by the wiring diagram of the olfactory bulb

Neuronal computations underlying higher brain functions depend on synaptic interactions among specific neurons. A mechanistic understanding of such computations requires wiring diagrams of neuronal networks. In this study, we examined how the olfactory bulb (OB) performs 'whitening', a fundamental computation that decorrelates activity patterns and supports their classification by memory networks. We measured odor-evoked activity in the OB of a zebrafish larva and subsequently reconstructed the complete wiring diagram by volumetric electron microscopy. The resulting functional connectome revealed an over-representation of multisynaptic connectivity motifs that mediate reciprocal inhibition between neurons with similar tuning. This connectivity suppressed redundant responses and was necessary and sufficient to reproduce whitening in simulations. Whitening of odor representations is therefore mediated by higher-order structure in the wiring diagram that is adapted to natural input patterns.

Mutual Inhibition with Few Inhibitory Cells via Nonlinear Inhibitory Synaptic Interaction

In computational neural network models, neurons are usually allowed to excite some and inhibit other neurons, depending on the weight of their synaptic connections. The traditional way to transform such networks into networks that obey Dale's law (i.e., a neuron can either excite or inhibit) is to accompany each excitatory neuron with an inhibitory one through which inhibitory signals are mediated. However, this requires an equal number of excitatory and inhibitory neurons, whereas a realistic number of inhibitory neurons is much smaller. In this letter, we propose a model of nonlinear interaction of inhibitory synapses on dendritic compartments of excitatory neurons that allows the excitatory neurons to mediate inhibitory signals through a subset of the inhibitory population. With this construction, the number of required inhibitory neurons can be reduced tremendously.

Mosaic representations of odors in the input and output layers of the mouse olfactory bulb

The elementary stimulus features encoded by the olfactory system remain poorly understood. We examined the relationship between 1,666 physical-chemical descriptors of odors and the activity of olfactory bulb inputs and outputs in awake mice. Glomerular and mitral and tufted cell responses were sparse and locally heterogeneous, with only a weak dependence of their positions on physical-chemical properties. Odor features represented by ensembles of mitral and tufted cells were overlapping but distinct from those represented in glomeruli, which is consistent with an extensive interplay between feedforward and feedback inputs to the bulb. This reformatting was well described as a rotation in odor space. The physical-chemical descriptors accounted for a small fraction in response variance, and the similarity of odors in the physical-chemical space was a poor predictor of similarity in neuronal representations. Our results suggest that commonly used physical-chemical properties are not systematically represented in bulbar activity and encourage further searches for better descriptors of odor space.

Disparate forms of heterogeneities and interactions among them drive channel decorrelation in the dentate gyrus: Degeneracy and dominance

The ability of a neuronal population to effectuate channel decorrelation, which is one form of response decorrelation, has been identified as an essential prelude to efficient neural encoding. To what extent are diverse forms of local and afferent heterogeneities essential in accomplishing channel decorrelation in the dentate gyrus (DG)? Here, we incrementally incorporated four distinct forms of biological heterogeneities into conductance-based network models of the DG and systematically delineate their relative contributions to channel decorrelation. First, to effectively incorporate intrinsic heterogeneities, we built physiologically validated heterogeneous populations of granule (GC) and basket cells (BC) through independent stochastic search algorithms spanning exhaustive parametric spaces. These stochastic search algorithms, which were independently constrained by experimentally determined ion channels and by neurophysiological signatures, revealed cellular-scale degeneracy in the DG. Specifically, in GC and BC populations, disparate parametric combinations yielded similar physiological signatures, with underlying parameters exhibiting significant variability and weak pair-wise correlations. Second, we introduced synaptic heterogeneities through randomization of local synaptic strengths. Third, in including adult neurogenesis, we subjected the valid model populations to randomized structural plasticity and matched neuronal excitability to electrophysiological data. We assessed networks comprising different combinations of these three local heterogeneities with identical or heterogeneous afferent inputs from the entorhinal cortex. We found that the three forms of local heterogeneities were independently and synergistically capable of mediating significant channel decorrelation when the network was driven by identical afferent inputs. However, when we incorporated afferent heterogeneities into the network to account for the divergence in DG afferent connectivity, the impact of all three forms of local heterogeneities was significantly suppressed by the dominant role of afferent heterogeneities in mediating channel decorrelation. Our results unveil a unique convergence of cellular- and network-scale degeneracy in the emergence of channel decorrelation in the DG, whereby disparate forms of local and afferent heterogeneities could synergistically drive input discriminability.

Re-evaluating Circuit Mechanisms Underlying Pattern Separation

When animals interact with complex environments, their neural circuits must separate overlapping patterns of activity that represent sensory and motor information. Pattern separation is thought to be a key function of several brain regions, including the cerebellar cortex, insect mushroom body, and dentate gyrus. However, recent findings have questioned long-held ideas on how these circuits perform this fundamental computation. Here, we re-evaluate the functional and structural mechanisms underlying pattern separation. We argue that the dimensionality of the space available for population codes representing sensory and motor information provides a common framework for understanding pattern separation. We then discuss how these three circuits use different strategies to separate activity patterns and facilitate associative learning in the presence of trial-to-trial variability.

Top-down inputs drive neuronal network rewiring and context-enhanced sensory processing in olfaction

Much of the computational power of the mammalian brain arises from its extensive top-down projections. To enable neuron-specific information processing these projections have to be precisely targeted. How such a specific connectivity emerges and what functions it supports is still poorly understood. We addressed these questions in silico in the context of the profound structural plasticity of the olfactory system. At the core of this plasticity are the granule cells of the olfactory bulb, which integrate bottom-up sensory inputs and top-down inputs delivered by vast top-down projections from cortical and other brain areas. We developed a biophysically supported computational model for the rewiring of the top-down projections and the intra-bulbar network via adult neurogenesis. The model captures various previous physiological and behavioral observations and makes specific predictions for the cortico-bulbar network connectivity that is learned by odor exposure and environmental contexts. Specifically, it predicts that-after learning-the granule-cell receptive fields with respect to sensory and with respect to cortical inputs are highly correlated. This enables cortical cells that respond to a learned odor to enact disynaptic inhibitory control specifically of bulbar principal cells that respond to that odor. For this the reciprocal nature of the granule cell synapses with the principal cells is essential. Functionally, the model predicts context-enhanced stimulus discrimination in cluttered environments ('olfactory cocktail parties') and the ability of the system to adapt to its tasks by rapidly switching between different odor-processing modes. These predictions are experimentally testable. At the same time they provide guidance for future experiments aimed at unraveling the cortico-bulbar connectivity.

Parvalbumin+ interneurons obey unique connectivity rules and establish a powerful lateral-inhibition microcircuit in dentate gyrus

Parvalbumin-positive (PV⁺) GABAergic interneurons in hippocampal microcircuits are thought to play a key role in several higher network functions, such as feedforward and feedback inhibition, network oscillations, and pattern separation. Fast lateral inhibition mediated by GABAergic interneurons may implement a winner-takes-all mechanism in the hippocampal input layer. However, it is not clear whether the functional connectivity rules of granule cells (GCs) and interneurons in the dentate gyrus are consistent with such a mechanism. Using simultaneous patch-clamp recordings from up to seven GCs and up to four PV⁺ interneurons in the dentate gyrus, we find that connectivity is structured in space, synapse-specific, and enriched in specific disynaptic motifs. In contrast to the neocortex, lateral inhibition in the dentate gyrus (in which a GC inhibits neighboring GCs via a PV⁺ interneuron) is ~ 10-times more abundant than recurrent inhibition (in which a GC inhibits itself). Thus, unique connectivity rules may enable the dentate gyrus to perform specific higher-order computations.

GABAergic interneurons are known to provide inhibition to allow computational function of neuronal network. Here, Espinoza and colleagues show that connectivity of granule cells and interneurons in the dentate gyrus of mouse hippocampus are consistent with the circuit architecture capable of performing a winners-take-all mechanism.

Sparse synaptic connectivity is required for decorrelation and pattern separation in feedforward networks

Pattern separation is a fundamental function of the brain. The divergent feedforward networks thought to underlie this computation are widespread, yet exhibit remarkably similar sparse synaptic connectivity. Marr-Albus theory postulates that such networks separate overlapping activity patterns by mapping them onto larger numbers of sparsely active neurons. But spatial correlations in synaptic input and those introduced by network connectivity are likely to compromise performance. To investigate the structural and functional determinants of pattern separation we built models of the cerebellar input layer with spatially correlated input patterns, and systematically varied their synaptic connectivity. Performance was quantified by the learning speed of a classifier trained on either the input or output patterns. Our results show that sparse synaptic connectivity is essential for separating spatially correlated input patterns over a wide range of network activity, and that expansion and correlations, rather than sparse activity, are the major determinants of pattern separation.

Input decorrelation, expansion recoding and sparse activity have been proposed to separate overlapping activity patterns in feedforward networks. Here the authors use reduced and detailed spiking models to elucidate how synaptic connectivity affects the contribution of these mechanisms to pattern separation in cerebellar cortex.

Distinct Correlation Structure Supporting a Rate-Code for Sound Localization in the Owl's Auditory Forebrain

While a topographic map of auditory space exists in the vertebrate midbrain, it is absent in the forebrain. Yet, both brain regions are implicated in sound localization. The heterogeneous spatial tuning of adjacent sites in the forebrain compared to the midbrain reflects different underlying circuitries, which is expected to affect the correlation structure, i.e., signal (similarity of tuning) and noise (trial-by-trial variability) correlations. Recent studies have drawn attention to the impact of response correlations on the information readout from a neural population. We thus analyzed the correlation structure in midbrain and forebrain regions of the barn owl’s auditory system. Tetrodes were used to record in the midbrain and two forebrain regions, Field L and the downstream auditory arcopallium (AAr), in anesthetized owls. Nearby neurons in the midbrain showed high signal and noise correlations (R_NCs), consistent with shared inputs. As previously reported, Field L was arranged in random clusters of similarly tuned neurons. Interestingly, AAr neurons displayed homogeneous monotonic azimuth tuning, while response variability of nearby neurons was significantly less correlated than the midbrain. Using a decoding approach, we demonstrate that low R_NC in AAr restricts the potentially detrimental effect it can have on information, assuming a rate code proposed for mammalian sound localization. This study harnesses the power of correlation structure analysis to investigate the coding of auditory space. Our findings demonstrate distinct correlation structures in the auditory midbrain and forebrain, which would be beneficial for a rate-code framework for sound localization in the nontopographic forebrain representation of auditory space.

Optimal Degrees of Synaptic Connectivity

Synaptic connectivity varies widely across neuronal types. Cerebellar granule cells receive five orders of magnitude fewer inputs than the Purkinje cells they innervate, and cerebellum-like circuits, including the insect mushroom body, also exhibit large divergences in connectivity. In contrast, the number of inputs per neuron in cerebral cortex is more uniform and large. We investigate how the dimension of a representation formed by a population of neurons depends on how many inputs each neuron receives and what this implies for learning associations. Our theory predicts that the dimensions of the cerebellar granule-cell and Drosophila Kenyon-cell representations are maximized at degrees of synaptic connectivity that match those observed anatomically, showing that sparse connectivity is sometimes superior to dense connectivity. When input synapses are subject to supervised plasticity, however, dense wiring becomes advantageous, suggesting that the type of plasticity exhibited by a set of synapses is a major determinant of connection density.

Inhibition decorrelates visual feature representations in the inner retina

The retina extracts visual features for transmission to the brain. Different types of bipolar cell split the photoreceptor input into parallel channels and provide the excitatory drive for downstream visual circuits. Anatomically and genetically, mouse bipolar cell types have been described at great detail, but a similarly deep understanding of their functional diversity is lacking. By imaging light-driven glutamate release from more than 13,000 bipolar cell axon terminals in the intact retina, we here show that bipolar cell functional diversity is generated by the interplay of dendritic excitatory inputs and axonal inhibitory inputs. The resultant centre and surround components of bipolar cell receptive fields interact to decorrelate bipolar cell output in the spatial and temporal domain. Our findings highlight the importance of inhibitory circuits in generating functionally diverse excitatory pathways and suggest that decorrelation of parallel visual pathways begins already at the second synapse of the mouse visual system.

Power-law statistics and universal scaling in the absence of criticality

Critical states are sometimes identified experimentally through power-law statistics or universal scaling functions. We show here that such features naturally emerge from networks in self-sustained irregular regimes away from criticality. In these regimes, statistical physics theory of large interacting systems predict a regime where the nodes have independent and identically distributed dynamics. We thus investigated the statistics of a system in which units are replaced by independent stochastic surrogates and found the same power-law statistics, indicating that these are not sufficient to establish criticality. We rather suggest that these are universal features of large-scale networks when considered macroscopically. These results put caution on the interpretation of scaling laws found in nature.

A Low-Correlation Resting State of the Striatum during Cortical Avalanches and Its Role in Movement Suppression

During quiet resting behavior, involuntary movements are suppressed. Such movement control is attributed to cortico-basal ganglia loops, yet population dynamics within these loops during resting and their relation to involuntary movements are not well characterized. Here, we show by recording cortical and striatal ongoing population activity in awake rats during quiet resting that intrastriatal inhibition maintains a low-correlation striatal resting state in the presence of cortical neuronal avalanches. Involuntary movements arise from disturbed striatal resting activity through two different population dynamics. Nonselectively reducing intrastriatal γ-aminobutyric acid (GABA) receptor-A inhibition synchronizes striatal dynamics, leading to involuntary movements at low rate. In contrast, reducing striatal interneuron (IN)-mediated inhibition maintains decorrelation and induces intermittent involuntary movements at high rate. This latter scenario was highly effective in modulating cortical dynamics at a subsecond timescale. To distinguish intrastriatal processing from loop dynamics, cortex-striatum-midbrain cultures, which lack feedback to cortex, were used. Cortical avalanches in vitro were accompanied by low-correlated resting activity in the striatum and nonselective reduction in striatal inhibition synchronized striatal neurons similar to in vivo. Importantly, reduction of inhibition from striatal INs maintained low correlations in the striatum while reorganizing functional connectivities among striatal neurons. Our results demonstrate the importance of two major striatal microcircuits in distinctly regulating striatal and cortical resting state dynamics. These findings suggest that specific functional connectivities of the striatum that are maintained by local inhibition are important in movement control.

Why don’t neuronal “avalanches” in resting-state cortex cause involuntary movements? This study shows that a low-correlation striatal resting state suppresses such movements and explores mechanisms that disrupt this inhibition.

Even in the absence of apparent motor output, the brain produces a rich repertoire of neuronal activity patterns known as “resting state” activity. In the outer layer of the cortex, resting state patterns emerge as neuronal avalanches, precisely scale-invariant spatiotemporal bursts that often engage large populations of neurons. Little is known about how the brain suppresses involuntary movements during such activity. Here, we show that the striatum, which is part of the cortex-basal ganglia loop, maintains a low-correlation state during resting activity. By using a combination of in vivo and in vitro approaches with pharmacological manipulations, we demonstrate that the precise configuration of this low-correlation state effectively contributes to involuntary movements. Nonselective blockade of intra-striatal inhibition abolished the low-correlation striatal resting state, barely affected cortical avalanches, and led to involuntary movements at low rate. In contrast, selectively reducing striatal interneuron inhibition strongly affected cortical avalanches and triggered involuntary movements at high rate while maintaining a relatively decorrelated striatal resting state. Our results demonstrate the importance of different inhibitory striatal circuits in the suppression of involuntary movements and suggest that the precise spatiotemporal configuration of striatal activity plays an active role in the regulation of cortical resting state activity and motor control.

A probabilistic approach to demixing odors

The olfactory system faces a hard problem: on the basis of noisy information from olfactory receptor neurons (the neurons that transduce chemicals to neural activity), it must figure out which odors are present in the world. Odors almost never occur in isolation, and different odors excite overlapping populations of olfactory receptor neurons, so the central challenge of the olfactory system is to demix its input. Because of noise and the large number of possible odors, demixing is fundamentally a probabilistic inference task. We propose that the early olfactory system uses approximate Bayesian inference to solve it. The computations involve a dynamical loop between the olfactory bulb and the piriform cortex, with cortex explaining incoming activity from the olfactory receptor neurons in terms of a mixture of odors. The model is compatible with known anatomy and physiology, including pattern decorrelation, and it performs better than other models at demixing odors.

Dense encoding of natural odorants by ensembles of sparsely activated neurons in the olfactory bulb

Sensory information undergoes substantial transformation along sensory pathways, usually encompassing sparsening of activity. In the olfactory bulb, though natural odorants evoke dense glomerular input maps, mitral and tufted (M/T) cells tuning is considered to be sparse because of highly odor-specific firing rate change. However, experiments used to draw this conclusion were either based on recordings performed in anesthetized preparations or used monomolecular odorants presented at arbitrary concentrations. In this study, we evaluated the lifetime and population sparseness evoked by natural odorants by capturing spike temporal patterning of neuronal assemblies instead of individual M/T tonic activity. Using functional imaging and tetrode recordings in awake mice, we show that natural odorants at their native concentrations are encoded by broad assemblies of M/T cells. While reducing odorant concentrations, we observed a reduced number of activated glomeruli representations and consequently a narrowing of M/T tuning curves. We conclude that natural odorants at their native concentrations recruit M/T cells with phasic rather than tonic activity. When encoding odorants in assemblies, M/T cells carry information about a vast number of odorants (lifetime sparseness). In addition, each natural odorant activates a broad M/T cell assembly (population sparseness).

Population imaging at subcellular resolution supports specific and local inhibition by granule cells in the olfactory bulb

Information processing in early sensory regions is modulated by a diverse range of inhibitory interneurons. We sought to elucidate the role of olfactory bulb interneurons called granule cells (GCs) in odor processing by imaging the activity of hundreds of these cells simultaneously in mice. Odor responses in GCs were temporally diverse and spatially disperse, with some degree of non-random, modular organization. The overall sparseness of activation of GCs was highly correlated with the extent of glomerular activation by odor stimuli. Increasing concentrations of single odorants led to proportionately larger population activity, but some individual GCs had non-monotonic relations to concentration due to local inhibitory interactions. Individual dendritic segments could sometimes respond independently to odors, revealing their capacity for compartmentalized signaling in vivo. Collectively, the response properties of GCs point to their role in specific and local processing, rather than global operations such as response normalization proposed for other interneurons.

Persistent Structural Plasticity Optimizes Sensory Information Processing in the Olfactory Bulb

In the mammalian brain, the anatomical structure of neural circuits changes little during adulthood. As a result, adult learning and memory are thought to result from specific changes in synaptic strength. A possible exception is the olfactory bulb (OB), where activity guides interneuron turnover throughout adulthood. These adult-born granule cell (GC) interneurons form new GABAergic synapses that have little synaptic strength plasticity. In the face of persistent neuronal and synaptic turnover, how does the OB balance flexibility, as is required for adapting to changing sensory environments, with perceptual stability? Here we show that high dendritic spine turnover is a universal feature of GCs, regardless of their developmental origin and age. We find matching dynamics among postsynaptic sites on the principal neurons receiving the new synaptic inputs. We further demonstrate in silico that this coordinated structural plasticity is consistent with stable, yet flexible, decorrelated sensory representations. Together, our study reveals that persistent, coordinated synaptic structural plasticity between interneurons and principal neurons is a major mode of functional plasticity in the OB.

A neural basis for the spatial suppression of visual motion perception

In theory, sensory perception should be more accurate when more neurons contribute to the representation of a stimulus. However, psychophysical experiments that use larger stimuli to activate larger pools of neurons sometimes report impoverished perceptual performance. To determine the neural mechanisms underlying these paradoxical findings, we trained monkeys to discriminate the direction of motion of visual stimuli that varied in size across trials, while simultaneously recording from populations of motion-sensitive neurons in cortical area MT. We used the resulting data to constrain a computational model that explained the behavioral data as an interaction of three main mechanisms: noise correlations, which prevented stimulus information from growing with stimulus size; neural surround suppression, which decreased sensitivity for large stimuli; and a read-out strategy that emphasized neurons with receptive fields near the stimulus center. These results suggest that paradoxical percepts reflect tradeoffs between sensitivity and noise in neuronal populations.

DOI:http://dx.doi.org/10.7554/eLife.16167.001

People usually find it easier to see things when they are big and bright, but there are occasionally exceptions. One example concerns moving objects: when they are small, we can identify their direction of motion easily, but this becomes much more difficult for larger objects. This decreased perceptual sensitivity appears to be linked to other mental processes. For example, studies have suggested that people with high IQs have more difficulty perceiving the motion of large objects, whereas people with various psychiatric disorders, such as schizophrenia, are better able to see such movement. Although several theories have been proposed, there is currently no good explanation for these findings.

Liu et al. set out to determine why the part of the brain that is responsible for vision (the visual cortex) fails to register the direction of large moving objects and how this failure might relate to mental function in general. To do this, Liu et al. trained monkeys to report which direction different sized stimuli were moving on a screen. The electrical activity of nerve cells in the part of the visual cortex that deals with movement was recorded while the monkeys performed this task. The results of the experiments revealed that, on average, these cells responded strongly to large moving stimuli, even though the monkeys had trouble seeing their motion. However, nerve cells are “noisy” – they respond a bit differently every time they are presented with the same stimulus – and this noise was stronger for larger stimuli.

By studying the mathematical relationship between the noise and what the animals perceived, Liu et al. found that the visual cortex attempts to suppress the noise and in the process often shuts off the responses to large stimuli entirely. This suppression is likely to cause the movement of large stimuli to be poorly perceived.

If suppressing this kind of noise is really responsible for failures in perceiving motion, then this mechanism could also explain the connection between motion perception and other mental processes. Liu et al. are currently testing this idea.

DOI:http://dx.doi.org/10.7554/eLife.16167.002

Dense EM-based reconstruction of the interglomerular projectome in the zebrafish olfactory bulb

The dense reconstruction of neuronal circuits from volumetric electron microscopy (EM) data has the potential to uncover fundamental structure-function relationships in the brain. To address bottlenecks in the workflow of this emerging methodology, we developed a procedure for conductive sample embedding and a pipeline for neuron reconstruction. We reconstructed ∼98% of all neurons (>1,000) in the olfactory bulb of a zebrafish larva with high accuracy and annotated all synapses on subsets of neurons representing different types. The organization of the larval olfactory bulb showed marked differences from that of the adult but similarities to that of the insect antennal lobe. Interneurons comprised multiple types but granule cells were rare. Interglomerular projections of interneurons were complex and bidirectional. Projections were not random but biased toward glomerular groups receiving input from common types of sensory neurons. Hence, the interneuron network in the olfactory bulb exhibits a specific topological organization that is governed by glomerular identity.

Cultured Cortical Neurons Can Perform Blind Source Separation According to the Free-Energy Principle

Blind source separation is the computation underlying the cocktail party effect--a partygoer can distinguish a particular talker's voice from the ambient noise. Early studies indicated that the brain might use blind source separation as a signal processing strategy for sensory perception and numerous mathematical models have been proposed; however, it remains unclear how the neural networks extract particular sources from a complex mixture of inputs. We discovered that neurons in cultures of dissociated rat cortical cells could learn to represent particular sources while filtering out other signals. Specifically, the distinct classes of neurons in the culture learned to respond to the distinct sources after repeating training stimulation. Moreover, the neural network structures changed to reduce free energy, as predicted by the free-energy principle, a candidate unified theory of learning and memory, and by Jaynes' principle of maximum entropy. This implicit learning can only be explained by some form of Hebbian plasticity. These results are the first in vitro (as opposed to in silico) demonstration of neural networks performing blind source separation, and the first formal demonstration of neuronal self-organization under the free energy principle.

Neuronal pattern separation in the olfactory bulb improves odor discrimination learning

Neuronal pattern separation is thought to enable the brain to disambiguate sensory stimuli with overlapping features, thereby extracting valuable information. In the olfactory system, it remains unknown whether pattern separation acts as a driving force for sensory discrimination and the learning thereof. We found that overlapping odor-evoked input patterns to the mouse olfactory bulb (OB) were dynamically reformatted in the network on the timescale of a single breath, giving rise to separated patterns of activity in an ensemble of output neurons, mitral/tufted (M/T) cells. Notably, the extent of pattern separation in M/T assemblies predicted behavioral discrimination performance during the learning phase. Furthermore, exciting or inhibiting GABAergic OB interneurons, using optogenetics or pharmacogenetics, altered pattern separation and thereby odor discrimination learning in a bidirectional way. In conclusion, we propose that the OB network can act as a pattern separator facilitating olfactory stimulus distinction, a process that is sculpted by synaptic inhibition.

Supersensitive detection and discrimination of enantiomers by dorsal olfactory receptors: evidence for hierarchical odour coding

Enantiomeric pairs of mirror-image molecular structures are difficult to resolve by instrumental analyses. The human olfactory system, however, discriminates (−)-wine lactone from its (+)-form rapidly within seconds. To gain insight into receptor coding of enantiomers, we compared behavioural detection and discrimination thresholds of wild-type mice with those of ΔD mice in which all dorsal olfactory receptors are genetically ablated. Surprisingly, wild-type mice displayed an exquisite “supersensitivity” to enantiomeric pairs of wine lactones and carvones. They were capable of supersensitive discrimination of enantiomers, consistent with their high detection sensitivity. In contrast, ΔD mice showed selective major loss of sensitivity to the (+)-enantiomers. The resulting 10⁸-fold differential sensitivity of ΔD mice to (−)- vs. (+)-wine lactone matched that observed in humans. This suggests that humans lack highly sensitive orthologous dorsal receptors for the (+)-enantiomer, similarly to ΔD mice. Moreover, ΔD mice showed >10¹⁰-fold reductions in enantiomer discrimination sensitivity compared to wild-type mice. ΔD mice detected one or both of the (−)- and (+)-enantiomers over a wide concentration range, but were unable to discriminate them. This “enantiomer odour discrimination paradox” indicates that the most sensitive dorsal receptors play a critical role in hierarchical odour coding for enantiomer identification.

Cortical Feedback Decorrelates Olfactory Bulb Output in Awake Mice

The olfactory bulb receives rich glutamatergic projections from the piriform cortex. However, the dynamics and importance of these feedback signals remain unknown. Here, we use multiphoton calcium imaging to monitor cortical feedback in the olfactory bulb of awake mice and further probe its impact on the bulb output. Responses of feedback boutons were sparse, odor specific, and often outlasted stimuli by several seconds. Odor presentation either enhanced or suppressed the activity of boutons. However, any given bouton responded with stereotypic polarity across multiple odors, preferring either enhancement or suppression. Feedback representations were locally diverse and differed in dynamics across bulb layers. Inactivation of piriform cortex increased odor responsiveness and pairwise similarity of mitral cells but had little impact on tufted cells. We propose that cortical feedback differentially impacts these two output channels of the bulb by specifically decorrelating mitral cell responses to enable odor separation.

Bulbar microcircuit model predicts connectivity and roles of interneurons in odor coding

Stimulus encoding by primary sensory brain areas provides a data-rich context for understanding their circuit mechanisms. The vertebrate olfactory bulb is an input area having unusual two-layer dendro-dendritic connections whose roles in odor coding are unclear. To clarify these roles, we built a detailed compartmental model of the rat olfactory bulb that synthesizes a much wider range of experimental observations on bulbar physiology and response dynamics than has hitherto been modeled. We predict that superficial-layer inhibitory interneurons (periglomerular cells) linearize the input-output transformation of the principal neurons (mitral cells), unlike previous models of contrast enhancement. The linearization is required to replicate observed linear summation of mitral odor responses. Further, in our model, action-potentials back-propagate along lateral dendrites of mitral cells and activate deep-layer inhibitory interneurons (granule cells). Using this, we propose sparse, long-range inhibition between mitral cells, mediated by granule cells, to explain how the respiratory phases of odor responses of sister mitral cells can be sometimes decorrelated as observed, despite receiving similar receptor input. We also rule out some alternative mechanisms. In our mechanism, we predict that a few distant mitral cells receiving input from different receptors, inhibit sister mitral cells differentially, by activating disjoint subsets of granule cells. This differential inhibition is strong enough to decorrelate their firing rate phases, and not merely modulate their spike timing. Thus our well-constrained model suggests novel computational roles for the two most numerous classes of interneurons in the bulb.

Mixed signal learning by spike correlation propagation in feedback inhibitory circuits

The brain can learn and detect mixed input signals masked by various types of noise, and spike-timing-dependent plasticity (STDP) is the candidate synaptic level mechanism. Because sensory inputs typically have spike correlation, and local circuits have dense feedback connections, input spikes cause the propagation of spike correlation in lateral circuits; however, it is largely unknown how this secondary correlation generated by lateral circuits influences learning processes through STDP, or whether it is beneficial to achieve efficient spike-based learning from uncertain stimuli. To explore the answers to these questions, we construct models of feedforward networks with lateral inhibitory circuits and study how propagated correlation influences STDP learning, and what kind of learning algorithm such circuits achieve. We derive analytical conditions at which neurons detect minor signals with STDP, and show that depending on the origin of the noise, different correlation timescales are useful for learning. In particular, we show that non-precise spike correlation is beneficial for learning in the presence of cross-talk noise. We also show that by considering excitatory and inhibitory STDP at lateral connections, the circuit can acquire a lateral structure optimal for signal detection. In addition, we demonstrate that the model performs blind source separation in a manner similar to the sequential sampling approximation of the Bayesian independent component analysis algorithm. Our results provide a basic understanding of STDP learning in feedback circuits by integrating analyses from both dynamical systems and information theory.

Synaptic diversity enables temporal coding of coincident multisensory inputs in single neurons

The ability of the brain to rapidly process information from multiple pathways is critical for reliable execution of complex sensory-motor behaviors, yet the cellular mechanisms underlying a neuronal representation of multimodal stimuli are poorly understood. Here we explored the possibility that the physiological diversity of mossy fiber (MF) to granule cell (GC) synapses within the mouse vestibulocerebellum may contribute to the processing of coincident multisensory information at the level of individual GCs. We found that the strength and short-term dynamics of individual MF-GC synapses can act as biophysical signatures for primary vestibular, secondary vestibular and visual input pathways. The majority of GCs receive inputs from different modalities, which when co-activated, produced enhanced GC firing rates and distinct first spike latencies. Thus, pathway-specific synaptic response properties permit temporal coding of correlated multisensory input by single GCs, thereby enriching sensory representation and facilitating pattern separation.