The estimation of visual motion has long been studied as a paradigmatic neural computation, and multiple models have been advanced to explain behavioral and neural responses to motion signals. A broad class of models, originating with the Reichardt correlator model, proposes that animals estimate motion by computing a temporal cross-correlation of light intensities from two neighboring points in visual space. These models provide a good description of experimental data in specific contexts but cannot explain motion percepts in stimuli lacking pairwise correlations. Here, we develop a theoretical formalism that can accommodate diverse stimuli and behavioral goals. To achieve this, we treat motion estimation as a problem of Bayesian inference. Pairwise models emerge as one component of the generalized strategy for motion estimation. However, correlation functions beyond second order enable more accurate motion estimation. Prior expectations that are asymmetric with respect to bright and dark contrast use correlations of both even and odd orders, and we show that psychophysical experiments using visual stimuli with symmetric probability distributions for contrast cannot reveal whether the subject uses odd-order correlators for motion estimation. This result highlights a gap in previous experiments, which have largely relied on symmetric contrast distributions. Our theoretical treatment provides a natural interpretation of many visual motion percepts, indicates that motion estimation should be revisited using a broader class of stimuli, demonstrates how correlation-based motion estimation is related to stimulus statistics, and provides multiple experimentally testable predictions.

Competing models have been proposed to explain how neurons integrate the thousands of inputs distributed throughout their dendritic trees. In a simple global integration model, inputs from all locations sum in the axon. In a two-stage integration model, inputs contribute directly to dendritic spikes, and outputs from multiple branches sum in the axon. These two models yield opposite predictions of how synapses at different dendritic locations should be scaled if they are to contribute equally to neuronal output. We used serial-section electron microscopy to reconstruct individual apical oblique dendritic branches of CA1 pyramidal neurons and observe a synapse distribution consistent with the two-stage integration model. Computational modeling suggests that the observed synapse distribution enhances the contribution of each dendritic branch to neuronal output.

The neuronal immediate early gene Arc/Arg-3.1 is widely used as one of the most reliable molecular markers for intense synaptic activity in vivo. However, the cis-acting elements responsible for such stringent activity dependence have not been firmly identified. Here we combined luciferase reporter assays in cultured cortical neurons and comparative genome mapping to identify the critical synaptic activity-responsive elements (SARE) of the Arc/Arg-3.1 gene. A major SARE was found as a unique approximately 100-bp element located at >5 kb upstream of the Arc/Arg-3.1 transcription initiation site in the mouse genome. This single element, when positioned immediately upstream of a minimal promoter, was necessary and sufficient to replicate crucial properties of endogenous Arc/Arg-3.1’s transcriptional regulation, including rapid onset of transcription triggered by synaptic activity and low basal expression during synaptic inactivity. We identified the major determinants of SARE as a unique cluster of neuronal activity-dependent cis-regulatory elements consisting of closely localized binding sites for CREB, MEF2, and SRF. Consistently, a SARE reporter could readily trace and mark an ensemble of cells that have experienced intense activity in the recent past in vivo. Taken together, our work uncovers a novel transcriptional mechanism by which a critical 100-bp element, SARE, mediates a predominant component of the synapse-to-nucleus signaling in ensembles of Arc/Arg-3.1-positive activated neurons.

Dendritic spines are the nearly ubiquitous site of excitatory synaptic input onto neurons and as such are critically positioned to influence diverse aspects of neuronal signalling. Decades of theoretical studies have proposed that spines may function as highly effective and modifiable chemical and electrical compartments that regulate synaptic efficacy, integration and plasticity. Experimental studies have confirmed activity-dependent structural dynamics and biochemical compartmentalization by spines. However, there is a longstanding debate over the influence of spines on the electrical aspects of synaptic transmission and dendritic operation. Here we measure the amplitude ratio of spine head to parent dendrite voltage across a range of dendritic compartments and calculate the associated spine neck resistance (R(neck)) for spines at apical trunk dendrites in rat hippocampal CA1 pyramidal neurons. We find that R(neck) is large enough ( 500 MΩ) to amplify substantially the spine head depolarization associated with a unitary synaptic input by  1.5- to  45-fold, depending on parent dendritic impedance. A morphologically realistic compartmental model capable of reproducing the observed spatial profile of the amplitude ratio indicates that spines provide a consistently high-impedance input structure throughout the dendritic arborization. Finally, we demonstrate that the amplification produced by spines encourages electrical interaction among coactive inputs through an R(neck)-dependent increase in spine head voltage-gated conductance activation. We conclude that the electrical properties of spines promote nonlinear dendritic processing and associated forms of plasticity and storage, thus fundamentally enhancing the computational capabilities of neurons.

We reconstructed the synaptic circuits of seven columns in the second neuropil or medulla behind the fly's compound eye. These neurons embody some of the most stereotyped circuits in one of the most miniaturized of animal brains. The reconstructions allow us, for the first time to our knowledge, to study variations between circuits in the medulla's neighboring columns. This variation in the number of synapses and the types of their synaptic partners has previously been little addressed because methods that visualize multiple circuits have not resolved detailed connections, and existing connectomic studies, which can see such connections, have not so far examined multiple reconstructions of the same circuit. Here, we address the omission by comparing the circuits common to all seven columns to assess variation in their connection strengths and the resultant rates of several different and distinct types of connection error. Error rates reveal that, overall, <1% of contacts are not part of a consensus circuit, and we classify those contacts that supplement (E+) or are missing from it (E-). Autapses, in which the same cell is both presynaptic and postsynaptic at the same synapse, are occasionally seen; two cells in particular, Dm9 and Mi1, form ≥20-fold more autapses than do other neurons. These results delimit the accuracy of developmental events that establish and normally maintain synaptic circuits with such precision, and thereby address the operation of such circuits. They also establish a precedent for error rates that will be required in the new science of connectomics.

Neural circuit reconstruction at single synapse resolution is increasingly recognized as crucially important to decipher the function of biological nervous systems. Volume electron microscopy in serial transmission or scanning mode has been demonstrated to provide the necessary resolution to segment or trace all neurites and to annotate all synaptic connections. 
Automatic annotation of synaptic connections has been done successfully in near isotropic electron microscopy of vertebrate model organisms. Results on non-isotropic data in insect models, however, are not yet on par with human annotation. 
We designed a new 3D-U-Net architecture to optimally represent isotropic fields of view in non-isotropic data. We used regression on a signed distance transform of manually annotated synaptic clefts of the CREMI challenge dataset to train this model and observed significant improvement over the state of the art. 
We developed open source software for optimized parallel prediction on very large volumetric datasets and applied our model to predict synaptic clefts in a 50 tera-voxels dataset of the complete Drosophila brain. Our model generalizes well to areas far away from where training data was available.

Neurons often possess elaborate axonal and dendritic arbors. Why do these arbors exist and what determines their form and dimensions? To answer these questions, I consider the wiring up of a large highly interconnected neuronal network, such as the cortical column. Implementation of such a network in the allotted volume requires all the salient features of neuronal morphology: the existence of branching dendrites and axons and the presence of dendritic spines. Therefore, the requirement of high interconnectivity is, in itself, sufficient to account for the existence of these features. Moreover, the actual lengths of axons and dendrites are close to the smallest possible length for a given interconnectivity, arguing that high interconnectivity is essential for cortical function.

Long-term potentiation (LTP) requires postsynaptic depolarization that can result from EPSPs paired with action potentials or larger EPSPs that trigger dendritic spikes. We explored the relative contribution of these sources of depolarization to LTP induction during synaptically driven action potential firing in hippocampal CA1 pyramidal neurons. Pairing of a weak test input with a strong input resulted in large LTP (approximately 75% increase) when the weak and strong inputs were both located in the apical dendrites. This form of LTP did not require somatic action potentials. When the strong input was located in the basal dendrites, the resulting LTP was smaller (< or =25% increase). Pairing the test input with somatically evoked action potentials mimicked this form of LTP. Thus, back-propagating action potentials may contribute to modest LTP, but local synaptic depolarization and/or dendritic spikes mediate a stronger form of LTP that requires spatial proximity of the associated synaptic inputs.

To survive, animals must convert sensory information into appropriate behaviours. Vision is a common sense for locating ethologically relevant stimuli and guiding motor responses. How circuitry converts object location in retinal coordinates to movement direction in body coordinates remains largely unknown. Here we show through behaviour, physiology, anatomy and connectomics in Drosophila that visuomotor transformation occurs by conversion of topographic maps formed by the dendrites of feature-detecting visual projection neurons (VPNs) into synaptic weight gradients of VPN outputs onto central brain neurons. We demonstrate how this gradient motif transforms the anteroposterior location of a visual looming stimulus into the fly's directional escape. Specifically, we discover that two neurons postsynaptic to a looming-responsive VPN type promote opposite takeoff directions. Opposite synaptic weight gradients onto these neurons from looming VPNs in different visual field regions convert localized looming threats into correctly oriented escapes. For a second looming-responsive VPN type, we demonstrate graded responses along the dorsoventral axis. We show that this synaptic gradient motif generalizes across all 20 primary VPN cell types and most often arises without VPN axon topography. Synaptic gradients may thus be a general mechanism for conveying spatial features of sensory information into directed motor outputs.

To adapt to an ever-changing environment, animals consolidate some, but not all, learning experiences to long-term memory. In mammals, long-term memory consolidation often involves neural pathway reactivation hours after memory acquisition. It is not known whether this delayed-reactivation schema is common across the animal kingdom or how information is stored during the delay period. Here, we show that, during courtship suppression learning, Drosophila exhibits delayed long-term memory consolidation. We also show that the same class of dopaminergic neurons engaged earlier in memory acquisition is also both necessary and sufficient for delayed long-term memory consolidation. Furthermore, we present evidence that, during learning, the translational regulator Orb2A tags specific synapses of mushroom body neurons for later consolidation. Consolidation involves the subsequent recruitment of Orb2B and the activity-dependent synthesis of CaMKII. Thus, our results provide evidence for the role of a neuromodulated, synapse-restricted molecule bridging memory acquisition and long-term memory consolidation in a learning animal.