Object Boundary Detection in Natural Images May Depend on "Incitatory" Cell-Cell Interactions.

Detecting object boundaries is crucial for recognition, but how the process unfolds in visual cortex remains unknown. To study the problem faced by a hypothetical boundary cell, and to predict how cortical circuitry could produce a boundary cell from a population of conventional "simple cells," we labeled 30,000 natural image patches and used Bayes' rule to help determine how a simple cell should influence a nearby boundary cell depending on its relative offset in receptive field position and orientation. We identified the following three basic types of cell-cell interactions: rising and falling interactions with a range of slopes and saturation rates, and nonmonotonic (bump-shaped) interactions with varying modes and amplitudes. Using simple models, we show that a ubiquitous cortical circuit motif consisting of direct excitation and indirect inhibition-a compound effect we call "incitation"-can produce the entire spectrum of simple cell-boundary cell interactions found in our dataset. Moreover, we show that the synaptic weights that parameterize an incitation circuit can be learned by a single-layer "delta" rule. We conclude that incitatory interconnections are a generally useful computing mechanism that the cortex may exploit to help solve difficult natural classification problems.SIGNIFICANCE STATEMENT Simple cells in primary visual cortex (V1) respond to oriented edges and have long been supposed to detect object boundaries, yet the prevailing model of a simple cell-a divisively normalized linear filter-is a surprisingly poor natural boundary detector. To understand why, we analyzed image statistics on and off object boundaries, allowing us to characterize the neural-style computations needed to perform well at this difficult natural classification task. We show that a simple circuit motif known to exist in V1 is capable of extracting high-quality boundary probability signals from local populations of simple cells. Our findings suggest a new, more general way of conceptualizing cell-cell interconnections in the cortex.

Classical-Contextual Interactions in V1 May Rely on Dendritic Computations.

A signature feature of the neocortex is the dense network of horizontal connections (HCs) through which pyramidal neurons (PNs) exchange "contextual" information. In primary visual cortex (V1), HCs are thought to facilitate boundary detection, a crucial operation for object recognition, but how HCs modulate PN responses to boundary cues within their classical receptive fields (CRF) remains unknown. We began by "asking" natural images, through a structured data collection and ground truth labeling process, what function a V1 cell should use to compute boundary probability from aligned edge cues within and outside its CRF. The "answer" was an asymmetric 2-D sigmoidal function, whose nonlinear form provides the first normative account for the "multiplicative" center-flanker interactions previously reported in V1 neurons (Kapadia et al., 1995, 2000; Polat et al., 1998). Using a detailed compartmental model, we then show that this boundary-detecting classical-contextual interaction function can be computed by NMDAR-dependent spatial synaptic interactions within PN dendrites - the site where classical and contextual inputs first converge in the cortex. In additional simulations, we show that local interneuron circuitry activated by HCs can powerfully leverage the nonlinear spatial computing capabilities of PN dendrites, providing the cortex with a highly flexible substrate for integration of classical and contextual information.

Optimizing a Neuron for Reliable Dendritic Subunit Pooling.

In certain biologically relevant computing scenarios, a neuron "pools" the outputs of multiple independent functional subunits, firing if any one of them crosses threshold. Recent studies suggest that active dendrites could provide the thresholding mechanism, so that both the thresholding and pooling operations could take place within a single neuron. A pooling neuron faces a difficult task, however. Dendrites can produce highly variable responses depending on the density and spatial patterning of their synaptic inputs, and bona fide dendritic firing may be very rare, making it difficult for a neuron to reliably detect when one of its many dendrites has "gone suprathreshold". Our goal has been to identify biological adaptations that optimize a neuron's performance at the binary subunit pooling (BSP) task. Katz et al. (2009) pointed to the importance of spine density gradients in shaping dendritic responses. In a similar vein, we used a compartmental model to study how a neuron's performance at the BSP task is affected by different spine density layouts and other biological variables. We found BSP performance was optimized when dendrites have (1) a decreasing spine density gradient (true for many types of pyramidal neurons); (2) low-to-medium resistance spine necks; (3) strong NMDA currents; (4) fast spiking Na+ channels; and (5) powerful hyperpolarizing inhibition. Our findings provide a normative account that links several neuronal properties within the context of a behaviorally relevant task, and may provide new insights into nature's subtle strategies for optimizing the computing capabilities of neural tissue.

How Dendrites Affect Online Recognition Memory.

In order to record the stream of autobiographical information that defines our unique personal history, our brains must form durable memories from single brief exposures to the patterned stimuli that impinge on them continuously throughout life. However, little is known about the computational strategies or neural mechanisms that underlie the brain's ability to perform this type of "online" learning. Based on increasing evidence that dendrites act as both signaling and learning units in the brain, we developed an analytical model that relates online recognition memory capacity to roughly a dozen dendritic, network, pattern, and task-related parameters. We used the model to determine what dendrite size maximizes storage capacity under varying assumptions about pattern density and noise level. We show that over a several-fold range of both of these parameters, and over multiple orders-of-magnitude of memory size, capacity is maximized when dendrites contain a few hundred synapses-roughly the natural number found in memory-related areas of the brain. Thus, in comparison to entire neurons, dendrites increase storage capacity by providing a larger number of better-sized learning units. Our model provides the first normative theory that explains how dendrites increase the brain's capacity for online learning; predicts which combinations of parameter settings we should expect to find in the brain under normal operating conditions; leads to novel interpretations of an array of existing experimental results; and provides a tool for understanding which changes associated with neurological disorders, aging, or stress are most likely to produce memory deficits-knowledge that could eventually help in the design of improved clinical treatments for memory loss.

NMDA spikes mediate amplification of inputs in the rat piriform cortex.

The piriform cortex (PCx) receives direct input from the olfactory bulb (OB) and is the brain's main station for odor recognition and memory. The transformation of the odor code from OB to PCx is profound: mitral and tufted cells in olfactory glomeruli respond to individual odorant molecules, whereas pyramidal neurons (PNs) in the PCx responds to multiple, apparently random combinations of activated glomeruli. How these 'discontinuous' receptive fields are formed from OB inputs remains unknown. Counter to the prevailing view that olfactory PNs sum their inputs passively, we show for the first time that NMDA spikes within individual dendrites can both amplify OB inputs and impose combination selectivity upon them, while their ability to compartmentalize voltage signals allows different dendrites to represent different odorant combinations. Thus, the 2-layer integrative behavior of olfactory PN dendrites provides a parsimonious account for the nonlinear remapping of the odor code from bulb to cortex.

Synaptic plasticity in dendrites: complications and coping strategies.

The elaborate morphology, nonlinear membrane mechanisms and spatiotemporally varying synaptic activation patterns of dendrites complicate the expression, compartmentalization and modulation of synaptic plasticity. To grapple with this complexity, we start with the observation that neurons in different brain areas face markedly different learning problems, and dendrites of different neuron types contribute to the cell's input-output function in markedly different ways. By committing to specific assumptions regarding a neuron's learning problem and its input-output function, specific inferences can be drawn regarding the synaptic plasticity mechanisms and outcomes that we 'ought' to expect for that neuron. Exploiting this assumption-driven approach can help both in interpreting existing experimental data and designing future experiments aimed at understanding the brain's myriad learning processes.

Multimap formation in visual cortex.

An extrastriate visual area such as V2 or V4 contains neurons selective for a multitude of complex shapes, all sharing a common topographic organization. Simultaneously developing multiple interdigitated maps--hereafter a "multimap"--is challenging in that neurons must compete to generate a diversity of response types locally, while cooperating with their dispersed same-type neighbors to achieve uniform visual field coverage for their response type at all orientations, scales, etc. Previously proposed map development schemes have relied on smooth spatial interaction functions to establish both topography and columnar organization, but by locally homogenizing cells' response properties, local smoothing mechanisms effectively rule out multimap formation. We found in computer simulations that the key requirements for multimap development are that neurons are enabled for plasticity only within highly active regions of cortex designated "learning eligibility regions" (LERs), but within an LER, each cell's learning rate is determined only by its activity level with no dependence on location. We show that a hybrid developmental rule that combines spatial and activity-dependent learning criteria in this way successfully produces multimaps when the input stream contains multiple distinct feature types, or in the degenerate case of a single feature type, produces a V1-like map with "salt-and-pepper" structure. Our results support the hypothesis that cortical maps containing a fine mixture of different response types, whether in monkey extrastriate cortex, mouse V1 or elsewhere in the cortex, rather than signaling a breakdown of map formation mechanisms at the fine scale, are a product of a generic cortical developmental scheme designed to map cells with a diversity of response properties across a shared topographic space.

An Augmented Two-Layer Model Captures Nonlinear Analog Spatial Integration Effects in Pyramidal Neuron Dendrites.

In pursuit of the goal to understand and eventually reproduce the diverse functions of the brain, a key challenge lies in reverse engineering the peculiar biology-based "technology" that underlies the brain's remarkable ability to process and store information. The basic building block of the nervous system is the nerve cell, or "neuron," yet after more than 100 years of neurophysiological study and 60 years of modeling, the information processing functions of individual neurons, and the parameters that allow them to engage in so many different types of computation (sensory, motor, mnemonic, executive, etc.) remain poorly understood. In this paper, we review both historical and recent findings that have led to our current understanding of the analog spatial processing capabilities of dendrites, the major input structures of neurons, with a focus on the principal cell type of the neocortex and hippocampus, the pyramidal neuron (PN). We encapsulate our current understanding of PN dendritic integration in an abstract layered model whose spatially sensitive branch-subunits compute multidimensional sigmoidal functions. Unlike the 1-D sigmoids found in conventional neural network models, multidimensional sigmoids allow the cell to implement a rich spectrum of nonlinear modulation effects directly within their dendritic trees.

Mechanisms underlying subunit independence in pyramidal neuron dendrites.

Pyramidal neuron (PN) dendrites compartmentalize voltage signals and can generate local spikes, which has led to the proposal that their dendrites act as independent computational subunits within a multilayered processing scheme. However, when a PN is strongly activated, back-propagating action potentials (bAPs) sweeping outward from the soma synchronize dendritic membrane potentials many times per second. How PN dendrites maintain the independence of their voltage-dependent computations, despite these repeated voltage resets, remains unknown. Using a detailed compartmental model of a layer 5 PN, and an improved method for quantifying subunit independence that incorporates a more accurate model of dendritic integration, we first established that the output of each dendrite can be almost perfectly predicted by the intensity and spatial configuration of its own synaptic inputs, and is nearly invariant to the rate of bAP-mediated "cross-talk" from other dendrites over a 100-fold range. Then, through an analysis of conductance, voltage, and current waveforms within the model cell, we identify three biophysical mechanisms that together help make independent dendritic computation possible in a firing neuron, suggesting that a major subtype of neocortical neuron has been optimized for layered, compartmentalized processing under in-vivo-like spiking conditions.

Computing local edge probability in natural scenes from a population of oriented simple cells.

A key computation in visual cortex is the extraction of object contours, where the first stage of processing is commonly attributed to V1 simple cells. The standard model of a simple cell-an oriented linear filter followed by a divisive normalization-fits a wide variety of physiological data, but is a poor performing local edge detector when applied to natural images. The brain's ability to finely discriminate edges from nonedges therefore likely depends on information encoded by local simple cell populations. To gain insight into the corresponding decoding problem, we used Bayes's rule to calculate edge probability at a given location/orientation in an image based on a surrounding filter population. Beginning with a set of ∼ 100 filters, we culled out a subset that were maximally informative about edges, and minimally correlated to allow factorization of the joint on- and off-edge likelihood functions. Key features of our approach include a new, efficient method for ground-truth edge labeling, an emphasis on achieving filter independence, including a focus on filters in the region orthogonal rather than tangential to an edge, and the use of a customized parametric model to represent the individual filter likelihood functions. The resulting population-based edge detector has zero parameters, calculates edge probability based on a sum of surrounding filter influences, is much more sharply tuned than the underlying linear filters, and effectively captures fine-scale edge structure in natural scenes. Our findings predict nonmonotonic interactions between cells in visual cortex, wherein a cell may for certain stimuli excite and for other stimuli inhibit the same neighboring cell, depending on the two cells' relative offsets in position and orientation, and their relative activation levels.

Contour enhancement benefits older adults with simulated central field loss.

PURPOSE: Age-related macular degeneration is the leading cause of vision loss among Americans aged >65 years. Currently, no effective treatment can reverse the central vision loss associated with most age-related macular degeneration. Digital image-processing techniques have been developed to improve image visibility for peripheral vision; however, both the selection and efficacy of such methods are limited. Progress has been difficult for two reasons: the exact nature of image enhancement that might benefit peripheral vision is not well understood, and efficient methods for testing such techniques have been elusive. The current study aims to develop both an effective image enhancement technique for peripheral vision and an efficient means for validating the technique. METHODS: We used a novel contour-detection algorithm to locate shape-defining edges in images based on natural-image statistics. We then enhanced the scene by locally boosting the luminance contrast along such contours. Using a gaze-contingent display, we simulated central visual field loss in normally sighted young (aged 18-30 years) and older adults (aged 58-88 years). Visual search performance was measured as a function of contour enhancement strength ["original" (unenhanced), "medium," and "high"]. For preference task, a separate group of subjects judged which image in a pair "would lead to better search performance." RESULTS: We found that although contour enhancement had no significant effect on search time and accuracy in young adults, Medium enhancement resulted in significantly shorter search time in older adults (about 13% reduction relative to original). Both age-groups preferred images with Medium enhancement over original (2-7 times). Furthermore, across age-groups, image content types, and enhancement strengths, there was a robust correlation between preference and performance. CONCLUSIONS: Our findings demonstrate a beneficial role of contour enhancement in peripheral vision for older adults. Our findings further suggest that task-specific preference judgments can be an efficient surrogate for performance testing.

Effects of contour enhancement on low-vision preference and visual search.

PURPOSE: To determine whether image enhancement improves visual search performance and whether enhanced images were also preferred by subjects with vision impairment. METHODS: Subjects (n = 24) with vision impairment (vision: 20/52 to 20/240) completed visual search and preference tasks for 150 static images that were enhanced to increase object contours' visual saliency. Subjects were divided into two groups and were shown three enhancement levels. Original and medium enhancements were shown to both groups. High enhancement was shown to group 1, and low enhancement was shown to group 2. For search, subjects pointed to an object that matched a search target displayed at the top left of the screen. An "integrated search performance" measure (area under the curve of cumulative correct response rate over search time) quantified performance. For preference, subjects indicated the preferred side when viewing the same image with different enhancement levels on side-by-side high-definition televisions. RESULTS: Contour enhancement did not improve performance in the visual search task. Group 1 subjects significantly (p < 0.001) rejected the High enhancement, and showed no preference for medium enhancement over the original images. Group 2 subjects significantly preferred (p < 0.001) both the medium and the low enhancement levels over original. Contrast sensitivity was correlated with both preference and performance; subjects with worse contrast sensitivity performed worse in the search task (ρ = 0.77, p < 0.001) and preferred more enhancement (ρ = -0.47, p = 0.02). No correlation between visual search performance and enhancement preference was found. However, a small group of subjects (n = 6) in a narrow range of mid-contrast sensitivity performed better with the enhancement, and most (n = 5) also preferred the enhancement. CONCLUSIONS: Preferences for image enhancement can be dissociated from search performance in people with vision impairment. Further investigations are needed to study the relationships between preference and performance for a narrow range of mid-contrast sensitivity where a beneficial effect of enhancement may exist.

Location-dependent excitatory synaptic interactions in pyramidal neuron dendrites.

Neocortical pyramidal neurons (PNs) receive thousands of excitatory synaptic contacts on their basal dendrites. Some act as classical driver inputs while others are thought to modulate PN responses based on sensory or behavioral context, but the biophysical mechanisms that mediate classical-contextual interactions in these dendrites remain poorly understood. We hypothesized that if two excitatory pathways bias their synaptic projections towards proximal vs. distal ends of the basal branches, the very different local spike thresholds and attenuation factors for inputs near and far from the soma might provide the basis for a classical-contextual functional asymmetry. Supporting this possibility, we found both in compartmental models and electrophysiological recordings in brain slices that the responses of basal dendrites to spatially separated inputs are indeed strongly asymmetric. Distal excitation lowers the local spike threshold for more proximal inputs, while having little effect on peak responses at the soma. In contrast, proximal excitation lowers the threshold, but also substantially increases the gain of distally-driven responses. Our findings support the view that PN basal dendrites possess significant analog computing capabilities, and suggest that the diverse forms of nonlinear response modulation seen in the neocortex, including uni-modal, cross-modal, and attentional effects, could depend in part on pathway-specific biases in the spatial distribution of excitatory synaptic contacts onto PN basal dendritic arbors.

Location-dependent effects of inhibition on local spiking in pyramidal neuron dendrites.

Cortical computations are critically dependent on interactions between pyramidal neurons (PNs) and a menagerie of inhibitory interneuron types. A key feature distinguishing interneuron types is the spatial distribution of their synaptic contacts onto PNs, but the location-dependent effects of inhibition are mostly unknown, especially under conditions involving active dendritic responses. We studied the effect of somatic vs. dendritic inhibition on local spike generation in basal dendrites of layer 5 PNs both in neocortical slices and in simple and detailed compartmental models, with equivalent results: somatic inhibition divisively suppressed the amplitude of dendritic spikes recorded at the soma while minimally affecting dendritic spike thresholds. In contrast, distal dendritic inhibition raised dendritic spike thresholds while minimally affecting their amplitudes. On-the-path dendritic inhibition modulated both the gain and threshold of dendritic spikes depending on its distance from the spike initiation zone. Our findings suggest that cortical circuits could assign different mixtures of gain vs. threshold inhibition to different neural pathways, and thus tailor their local computations, by managing their relative activation of soma- vs. dendrite-targeting interneurons.

Distinguishing Linear vs. Non-Linear Integration in CA1 Radial Oblique Dendrites: It's about Time.

It was recently shown that multiple excitatory inputs to CA1 pyramidal neuron dendrites must be activated nearly simultaneously to generate local dendritic spikes and supralinear responses at the soma; even slight input desynchronization prevented local spike initiation (Gasparini and Magee, 2006; Losonczy and Magee, 2006). This led to the conjecture that CA1 pyramidal neurons may only express their non-linear integrative capabilities during the highly synchronized sharp waves and ripples that occur during slow wave sleep and resting/consummatory behavior, whereas during active exploration and REM sleep (theta rhythm), inadequate synchronization of excitation would lead CA1 pyramidal cells to function as essentially linear devices. Using a detailed single neuron model, we replicated the experimentally observed synchronization effect for brief inputs mimicking single synaptic release events. When synapses were driven instead by double pulses, more representative of the bursty inputs that occur in vivo, we found that the tolerance for input desynchronization was increased by more than an order of magnitude. The effect depended mainly on paired-pulse facilitation of NMDA receptor-mediated responses at Schaffer collateral synapses. Our results suggest that CA1 pyramidal cells could function as non-linear integrative units in all major hippocampal states.

Capacity-enhancing synaptic learning rules in a medial temporal lobe online learning model.

Medial temporal lobe structures are responsible for recording the continuous stream of autobiographical memories that define our unique personal history. Remarkably, these areas can construct durable memories from brief exposures to the constantly changing activity patterns arriving from antecedent cortical areas. Using a computer model of the hippocampal Schaffer collateral pathway that incorporates evidence for dendritic spikes in CA1 pyramidal neurons, we searched for biologically-plausible long-term potentiation (LTP) and homeostatic depression (HD) rules that maximize "online" learning capacity. We found memory utilization is most efficient when (1) very few synapses are modified to store each pattern, (2) LTP, the learning operation, is dendrite-specific and gated by distinct pre- and postsynaptic thresholds, (3) HD, the forgetting operation, co-occurs with LTP and targets least-recently potentiated synapses, and (4) both LTP and HD are all-or-none, leading de facto to binary-valued synaptic weights. In networks containing up to 40 million synapses, the learning scheme led to order-of-magnitude capacity increases compared to conventional plasticity rules.

Cue combination and color edge detection in natural scenes.

Biological vision systems are adept at combining cues to maximize the reliability of object boundary detection, but given a set of co-localized edge detectors operating on different sensory channels, how should their responses be combined to compute overall edge probability? To approach this question, we collected joint responses of red-green and blue-yellow edge detectors both ON- and OFF-edges using a human-labeled image database as ground truth (D. Martin, C. Fowlkes, D. Tal, & J. Malik, 2001). From a Bayesian perspective, the rule for combining edge cues is linear in the individual cue strengths when the ON-edge and OFF-edge joint distributions are (1) statistically independent and (2) lie in an exponential ratio to each other. Neither condition held in the color edge data we collected, and the function P(ON cues)-dubbed the "combination rule"-was correspondingly complex and nonlinear. To characterize the statistical dependencies between edge cues, we developed a generative model ("saturated common factor," SCF) that provided good fits to the measured ON-edge and OFF-edge joint distributions. We also found that a divisive normalization scheme derived from the SCF model transformed raw edge detector responses into values with simpler distributions that satisfied both preconditions for a linear combination rule. A comparison to another normalization scheme (O. Schwartz & E. Simoncelli, 2001) suggests that apparently minor details of the normalization process can strongly influence its performance. Implications of the SCF normalization scheme for cue combination in biological sensory systems are discussed.

J4 at sweet 16: a new wrinkle?

Compartmental models provide a major source of insight into the information processing functions of single neurons. Over the past 15 years, one of the most widely used neuronal morphologies has been the cell called "j4," a layer 5 pyramidal cell from cat visual cortex originally described in Douglas, Martin, and Whitteridge (1991). The cell has since appeared in at least 28 published compartmental modeling studies, including several in this journal. In recently examining why we could not reproduce certain in vitro data involving the attenuation of signals originating in distal basal dendrites, we discovered that pronounced fluctuations in the diameter measurements of j4 lead to a bottlenecking effect that increases distal input resistances and significantly reduces voltage transfer between distal sites and the cell body. Upon smoothing these diameter fluctuations, bringing j4 more in line with other reconstructions of layer 5 pyramidal neurons, we found that the attenuation of steady-state voltage signals traveling to the cell body V(distal)/V(soma) was reduced by 60% at some locations in some branches (corresponding to a 2.5-fold increase in the voltage response at the soma for the same distal depolarization) and by 30% on average (corresponding to a 45% increase in somatic response). Changes of this magnitude could lead to different outcomes in some types of compartmental modeling studies. A smoothed version of the j4 morphology is available online at http://lnc.usc.edu/j4-smooth/ .

On the fight between excitation and inhibition: location is everything.

In the intact brain, neurons are constantly subjected to both excitatory and inhibitory inputs to their dendritic trees. Although it is accepted that the overall response of a neuron--its train of output spikes--depends on the balance of excitation and inhibition, we continue to lack specific knowledge of the rules that govern how excitatory and inhibitory inputs interact in space and time within the confines of individual neurons. In a recent paper, Liu starts by providing evidence that the relative locations and numbers of excitatory and inhibitory synapses are tightly regulated in cultured neurons from the hippocampus. This is consistent with findings in other labs that suggest neurons work hard, and in a variety of different ways, to maintain their inputs in proper balance and their outputs within appropriate ranges. On this backdrop, Liu's most important finding of a functional nature is that inhibition appears to act quite locally; that is, an inhibitory synapse effectively opposes an excitatory synapse only when it is very close by within the same dendritic branch (Fig. 1). This finding provides further support for the view--anticipated by neural theorists more than 20 years ago--that the brain's principal neurons contain a potentially large number of separate computational subunits.

Computational subunits in thin dendrites of pyramidal cells.

The thin basal and oblique dendrites of cortical pyramidal neurons receive most of the synaptic inputs from other cells, but their integrative properties remain uncertain. Previous studies have most often reported global linear or sublinear summation. An alternative view, supported by biophysical modeling studies, holds that thin dendrites provide a layer of independent computational 'subunits' that sigmoidally modulate their inputs prior to global summation. To distinguish these possibilities, we combined confocal imaging and dual-site focal synaptic stimulation of identified thin dendrites in rat neocortical pyramidal neurons. We found that nearby inputs on the same branch summed sigmoidally, whereas widely separated inputs or inputs to different branches summed linearly. This strong spatial compartmentalization effect is incompatible with a global summation rule and provides the first experimental support for a two-layer 'neural network' model of pyramidal neuron thin-branch integration. Our findings could have important implications for the computing and memory-related functions of cortical tissue.