An efficient coding hypothesis links sparsity and selectivity of neural responses

To what extent are sensory responses in the brain compatible with first-order principles? The efficient coding hypothesis projects that neurons use as few spikes as possible to faithfully represent natural stimuli. However, many sparsely firing neurons in higher brain areas seem to violate this hypothesis in that they respond more to familiar stimuli than to nonfamiliar stimuli. We reconcile this discrepancy by showing that efficient sensory responses give rise to stimulus selectivity that depends on the stimulus-independent firing threshold and the balance between excitatory and inhibitory inputs. We construct a cost function that enforces minimal firing rates in model neurons by linearly punishing suprathreshold synaptic currents. By contrast, subthreshold currents are punished quadratically, which allows us to optimally reconstruct sensory inputs from elicited responses. We train synaptic currents on many renditions of a particular bird''s own song (BOS) and few renditions of conspecific birds'' songs (CONs). During training, model neurons develop a response selectivity with complex dependence on the firing threshold. At low thresholds, they fire densely and prefer CON and the reverse BOS (REV) over BOS. However, at high thresholds or when hyperpolarized, they fire sparsely and prefer BOS over REV and over CON. Based on this selectivity reversal, our model suggests that preference for a highly familiar stimulus corresponds to a high-threshold or strong-inhibition regime of an efficient coding strategy. Our findings apply to songbird mirror neurons, and in general, they suggest that the brain may be endowed with simple mechanisms to rapidly change selectivity of neural responses to focus sensory processing on either familiar or nonfamiliar stimuli. In summary, we find support for the efficient coding hypothesis and provide new insights into the interplay between the sparsity and selectivity of neural responses.     

Spatiotemporal burst coding for extracting features of spatiotemporally varying stimuli

Encoding features of spatiotemporally varying stimuli is quite important for understanding the neural mechanisms of various sensory coding. Temporal coding can encode features of time-varying stimulus, and population coding with temporal coding is adequate for encoding spatiotemporal correlation of stimulus features into spatiotemporal activity of neurons. However, little is known about how spatiotemporal features of stimulus are encoded by spatiotemporal property of neural activity. To address this issue, we propose here a population coding with burst spikes, called here spatiotemporal burst (STB) coding. In STB coding, the temporal variation of stimuli is encoded by the precise onset timing of burst spike, and the spatiotemporal correlation of stimuli is emphasized by one specific aspect of burst firing, or spike packet followed by silent interval. To show concretely the role of STB coding, we study the electrosensory system of a weakly electric fish. Weakly electric fish must perceive the information about an object nearby by analyzing spatiotemporal modulations of electric field around it. On the basis of well-characterized circuitry, we constructed a neural network model of the electrosensory system. Here we show that STB coding encodes well the information of object distance and size by extracting the spatiotemporal correlation of the distorted electric field. The burst activity of electrosensory neurons is also affected by feedback signals through synaptic plasticity. We show that the control of burst activity caused by the synaptic plasticity leads to extracting the stimulus features depending on the stimulus context. Our results suggest that sensory systems use burst spikes as a unit of sensory coding in order to extract spatiotemporal features of stimuli from spatially distributed stimuli.     

Context-dependent coding in single neurons

The linear-nonlinear cascade model (LN model) has proven very useful in representing a neural system’s encoding properties, but has proven less successful in reproducing the firing patterns of individual neurons whose behavior is strongly dependent on prior firing history. While the cell’s behavior can still usefully be considered as feature detection acting on a fluctuating input, some of the coding capacity of the cell is taken up by the increased firing rate due to a constant “driving” direct current (DC) stimulus. Furthermore, both the DC input and the post-spike refractory period generate regular firing, reducing the spike-timing entropy available for encoding time-varying fluctuations. In this paper, we address these issues, focusing on the example of motoneurons in which an afterhyperpolarization (AHP) current plays a dominant role regularizing firing behavior. We explore the accuracy and generalizability of several alternative models for single neurons under changes in DC and variance of the stimulus input. We use a motoneuron simulation to compare coding models in neurons with and without the AHP current. Finally, we quantify the tradeoff between instantaneously encoding information about fluctuations and about the DC.     

Noise-tolerant stimulus discrimination by synchronization with depressing synapses

 Some synapses between cortical pyramidal neurons exhibit a rapid depression of excitatory postsynaptic potentials for successive presynaptic spikes. Since depressing synapses do not transmit information on sustained presynaptic firing rates, it has been speculated that they are favorable for temporal coding. In this paper, we study the dynamical effects of depressing synapses on stimulus-induced transient synchronization in a simple network of inhibitory interneurons and excitatory neurons, assuming that the recurrent excitation is mediated by depressing synapses. This synchronization occurs in a temporal pattern which depends on a given stimulus. Since the presence of noise is always a potential hazard in temporal coding, we investigate the extent to which noise in stimuli influences the synchronization phenomena. It is demonstrated that depressing synapses greatly contribute to suppressing the influences of noise on the stimulus-specific temporal patterns of synchronous firing. The timing-based Hebbian learning revealed by physiological experiments is shown to stabilize the temporal patterns in cooperation with synaptic depression. Thus, the times at which synchronous firing occurs provides a reliable information representation in the presence of synaptic depression. Received: 5 July 2000 / Accepted in revised form: 12 January 2001     

Neural coding properties based on spike timing and pattern correlation of retinal ganglion cells

Correlation between spike trains or neurons sometimes indicates certain neural coding rules in the visual system. In this paper, the relationship between spike timing correlation and pattern correlation is discussed, and their ability to represent stimulus features is compared to examine their coding strategies not only in individual neurons but also in population. Two kinds of stimuli, natural movies and checkerboard, are used to arouse firing activities in chicken retinal ganglion cells. The spike timing correlation and pattern correlation are calculated by cross-correlation function and Lempel–Ziv distance respectively. According to the correlation values, it is demonstrated that spike trains with similar spike patterns are not necessarily concerted in firing time. Moreover, spike pattern correlation values between individual neurons’ responses reflect the difference of natural movies and checkerboard; neurons cooperate with each other with higher pattern correlation values which represent spatiotemporal correlations during response to natural movies. Spike timing does not reflect stimulus features as obvious as spike patterns, caused by their particular coding properties or physiological foundation. As a result, separating the pattern correlation out of traditional timing correlation concept uncover additional insight in neural coding.     

Balancing homeostasis and learning in neural circuits

Neural circuits are remarkably adaptable, providing animals with the ability to modify their behavior on the basis of experience. At the same time, they are extremely robust and maintain stability despite the changes associated with adaptation. This combination of adaptability and stability is difficult to achieve, and it provides a strong constraint on any models of plasticity in neural circuits. New evidence suggests that the effect of action potential timing on synaptic plasticity may be an important element in reconciling homeostasis with adaptability. In particular, spike-timing dependent plasticity can act as both an adaptive and a homeostatic mechanism, controlling overall firing rates and distributions of synaptic efficacies while making neurons selective for certain aspects of their inputs. It can also cause networks that initially represent the present state of a stimulus to predict its future state on the basis of experience, a theoretical result supported by experimental data in behaving rats.     

Constrained brain volume in an efficient coding model explains the fraction of excitatory and inhibitory neurons in sensory cortices

The number of neurons in mammalian cortex varies by multiple orders of magnitude across different species. In contrast, the ratio of excitatory to inhibitory neurons (E:I ratio) varies in a much smaller range, from 3:1 to 9:1 and remains roughly constant for different sensory areas within a species. Despite this structure being important for understanding the function of neural circuits, the reason for this consistency is not yet understood. While recent models of vision based on the efficient coding hypothesis show that increasing the number of both excitatory and inhibitory cells improves stimulus representation, the two cannot increase simultaneously due to constraints on brain volume. In this work, we implement an efficient coding model of vision under a constraint on the volume (using number of neurons as a surrogate) while varying the E:I ratio. We show that the performance of the model is optimal at biologically observed E:I ratios under several metrics. We argue that this happens due to trade-offs between the computational accuracy and the representation capacity for natural stimuli. Further, we make experimentally testable predictions that 1) the optimal E:I ratio should be higher for species with a higher sparsity in the neural activity and 2) the character of inhibitory synaptic distributions and firing rates should change depending on E:I ratio. Our findings, which are supported by our new preliminary analyses of publicly available data, provide the first quantitative and testable hypothesis based on optimal coding models for the distribution of excitatory and inhibitory neural types in the mammalian sensory cortices.     

To Burst or Not to Burst?

It is well known that some neurons tend to fire packets of action potentials followed by periods of quiescence (bursts) while others within the same stage of sensory processing fire in a tonic manner. However, the respective computational advantages of bursting and tonic neurons for encoding time varying signals largely remain a mystery. Weakly electric fish use cutaneous electroreceptors to convey information about sensory stimuli and it has been shown that some electroreceptors exhibit bursting dynamics while others do not. In this study, we compare the neural coding capabilities of tonically firing and bursting electroreceptor model neurons using information theoretic measures. We find that both bursting and tonically firing model neurons efficiently transmit information about the stimulus. However, the decoding mechanisms that must be used for each differ greatly: a non-linear decoder would be required to extract all the available information transmitted by the bursting model neuron whereas a linear one might suffice for the tonically firing model neuron. Further investigations using stimulus reconstruction techniques reveal that, unlike the tonically firing model neuron, the bursting model neuron does not encode the detailed time course of the stimulus. A novel measure of feature detection reveals that the bursting neuron signals certain stimulus features. Finally, we show that feature extraction and stimulus estimation are mutually exclusive computations occurring in bursting and tonically firing model neurons, respectively. Our results therefore suggest that stimulus estimation and feature extraction might be parallel computations in certain sensory systems rather than being sequential as has been previously proposed.     

Bidirectional plasticity gated by hyperpolarization controls the gain of postsynaptic firing responses at central vestibular nerve synapses

Linking synaptic plasticity with behavioral learning requires understanding how synaptic efficacy influences postsynaptic firing in neurons whose role in behavior is understood. Here, we examine plasticity at a candidate site of motor learning: vestibular nerve synapses onto neurons that mediate reflexive movements. Pairing nerve activity with changes in postsynaptic voltage induced bidirectional synaptic plasticity in vestibular nucleus projection neurons: long-term potentiation relied on calcium-permeable AMPA receptors and postsynaptic hyperpolarization, whereas long-term depression relied on NMDA receptors and postsynaptic depolarization. Remarkably, both forms of plasticity uniformly scaled synaptic currents evoked by pulse trains, and these changes in synaptic efficacy were translated into linear increases or decreases in postsynaptic firing responses. Synapses onto local inhibitory neurons were also plastic but expressed only long-term depression. Bidirectional, linear gain control of vestibular nerve synapses onto projection neurons provides a plausible mechanism for motor learning underlying adaptation of vestibular reflexes.     

Transformation of Adaptation and Gain Rescaling along the Whisker Sensory Pathway

Neurons in all sensory systems have a remarkable ability to adapt their sensitivity to the statistical structure of the sensory signals to which they are tuned. In the barrel cortex, firing rate adapts to the variance of a whisker stimulus and neuronal sensitivity (gain) adjusts in inverse proportion to the stimulus standard deviation. To determine how adaptation might be transformed across the ascending lemniscal pathway, we measured the responses of single units in the first and last subcortical stages, the trigeminal ganglion (TRG) and ventral posterior medial thalamic nucleus (VPM), to controlled whisker stimulation in urethane-anesthetized rats. We probed adaptation using a filtered white noise stimulus that switched between low- and high-variance epochs. We found that the firing rate of both TRG and VPM neurons adapted to stimulus variance. By fitting the responses of each unit to a Linear-Nonlinear-Poisson model, we tested whether adaptation changed feature selectivity and/or sensitivity. We found that, whereas feature selectivity was unaffected by stimulus variance, units often exhibited a marked change in sensitivity. The extent of these sensitivity changes increased systematically along the pathway from TRG to barrel cortex. However, there was marked variability across units, especially in VPM. In sum, in the whisker system, the adaptation properties of subcortical neurons are surprisingly diverse. The significance of this diversity may be that it contributes to a rich population representation of whisker dynamics.     

Inhibitory Contribution to Suprathreshold Corticostriatal Responses: An Experimental and Modeling Study

Neostriatal neurons may undergo events of spontaneous synchronization as those observed in recurrent networks of excitatory neurons, even when cortical afferents are transected. It is necessary to explain these events because the neostriatum is a recurrent network of inhibitory neurons. Synchronization of neuronal activity may be caused by plateau-like depolarizations. Plateau-like orthodromic depolarizations that resemble up-states in medium spiny neostriatal neurons (MSNs) may be induced by a single corticostriatal suprathreshold stimulus. Slow synaptic depolarizations may last hundreds of milliseconds, decay slower than the monosynaptic glutamatergic synaptic potentials that induce them, and sustain repetitive firing. Because inhibitory inputs impinging onto MSNs have a reversal potential above the resting membrane potential but below the threshold for firing, they conform a type of “shunting inhibition”. This work asks if shunting GABAergic inputs onto MSNs arrive asynchronously enough as to help in sustaining the plateau-like corticostriatal response after a single cortical stimulus. This may help to begin explaining autonomous processing in the striatal micro-circuitry in the presence of a tonic excitatory drive and independently of spatio-temporally organized inputs. It is shown here that besides synaptic currents from AMPA/KA- and NMDA-receptors, as well as L-type intrinsic Ca²⁺- currents, inhibitory synapses help in maintaining the slow depolarization, although they accomplish the role of depressing firing at the beginning of the response. We then used a NEURON model of spiny cells to show that inhibitory synapses arriving asynchronously on the dendrites can help to simulate a plateau potential similar to that observed experimentally. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Rapid adaptation and efficient coding

Rapid adaptation is a prominent feature of biological neuronal systems. From a functional perspective the adaptation of neuronal properties, namely the input-output relation of sensory neurons, is usually interpreted as an adaptation of the sensory system to changing environments as characterized by their stimulus statistics. Here we argue that this interpretation is only applicable as long as the adaptation processes are slower than the time-scale at which the stimulus statistics change. We present a definition of optimality of a neuronal code which still captures the idea of efficient coding, but which can also explain rapid adaptation without referring to an adaptation to different sensory environments. Finally, we apply our new idea to a simple model of an orientation hypercolumn in the primary visual cortex and predict that the interactions between orientation columns should adapt at the time-scale of a single stimulus presentation.     

Gradients and modulation of K(+) channels optimize temporal accuracy in networks of auditory neurons

Accurate timing of action potentials is required for neurons in auditory brainstem nuclei to encode the frequency and phase of incoming sound stimuli. Many such neurons express "high threshold" Kv3-family channels that are required for firing at high rates (> -200 Hz). Kv3 channels are expressed in gradients along the medial-lateral tonotopic axis of the nuclei. Numerical simulations of auditory brainstem neurons were used to calculate the input-output relations of ensembles of 1-50 neurons, stimulated at rates between 100-1500 Hz. Individual neurons with different levels of potassium currents differ in their ability to follow specific rates of stimulation but all perform poorly when the stimulus rate is greater than the maximal firing rate of the neurons. The temporal accuracy of the combined synaptic output of an ensemble is, however, enhanced by the presence of gradients in Kv3 channel levels over that measured when neurons express uniform levels of channels. Surprisingly, at high rates of stimulation, temporal accuracy is also enhanced by the occurrence of random spontaneous activity, such as is normally observed in the absence of sound stimulation. For any pattern of stimulation, however, greatest accuracy is observed when, in the presence of spontaneous activity, the levels of potassium conductance in all of the neurons is adjusted to that found in the subset of neurons that respond better than their neighbors. This optimization of response by adjusting the K(+) conductance occurs for stimulus patterns containing either single and or multiple frequencies in the phase-locking range. The findings suggest that gradients of channel expression are required for normal auditory processing and that changes in levels of potassium currents across the nuclei, by mechanisms such as protein phosphorylation and rapid changes in channel synthesis, adapt the nuclei to the ongoing auditory environment.     

Phasic Firing in Vasopressin Cells: Understanding Its Functional Significance through Computational Models

Vasopressin neurons, responding to input generated by osmotic pressure, use an intrinsic mechanism to shift from slow irregular firing to a distinct phasic pattern, consisting of long bursts and silences lasting tens of seconds. With increased input, bursts lengthen, eventually shifting to continuous firing. The phasic activity remains asynchronous across the cells and is not reflected in the population output signal. Here we have used a computational vasopressin neuron model to investigate the functional significance of the phasic firing pattern. We generated a concise model of the synaptic input driven spike firing mechanism that gives a close quantitative match to vasopressin neuron spike activity recorded in vivo, tested against endogenous activity and experimental interventions. The integrate-and-fire based model provides a simple physiological explanation of the phasic firing mechanism involving an activity-dependent slow depolarising afterpotential (DAP) generated by a calcium-inactivated potassium leak current. This is modulated by the slower, opposing, action of activity-dependent dendritic dynorphin release, which inactivates the DAP, the opposing effects generating successive periods of bursting and silence. Model cells are not spontaneously active, but fire when perturbed by random perturbations mimicking synaptic input. We constructed one population of such phasic neurons, and another population of similar cells but which lacked the ability to fire phasically. We then studied how these two populations differed in the way that they encoded changes in afferent inputs. By comparison with the non-phasic population, the phasic population responds linearly to increases in tonic synaptic input. Non-phasic cells respond to transient elevations in synaptic input in a way that strongly depends on background activity levels, phasic cells in a way that is independent of background levels, and show a similar strong linearization of the response. These findings show large differences in information coding between the populations, and apparent functional advantages of asynchronous phasic firing.     

Modelling Feedback Excitation,Pacemaker Properties and Sensory Switching of Electrically Coupled Brainstem Neurons Controlling Rhythmic Activity

What cellular and network properties allow reliable neuronal rhythm generation or firing that can be started and stopped by brief synaptic inputs? We investigate rhythmic activity in an electrically-coupled population of brainstem neurons driving swimming locomotion in young frog tadpoles, and how activity is switched on and off by brief sensory stimulation. We build a computational model of 30 electrically-coupled conditional pacemaker neurons on one side of the tadpole hindbrain and spinal cord. Based on experimental estimates for neuron properties, population sizes, synapse strengths and connections, we show that: long-lasting, mutual, glutamatergic excitation between the neurons allows the network to sustain rhythmic pacemaker firing at swimming frequencies following brief synaptic excitation; activity persists but rhythm breaks down without electrical coupling; NMDA voltage-dependency doubles the range of synaptic feedback strengths generating sustained rhythm. The network can be switched on and off at short latency by brief synaptic excitation and inhibition. We demonstrate that a population of generic Hodgkin-Huxley type neurons coupled by glutamatergic excitatory feedback can generate sustained asynchronous firing switched on and off synaptically. We conclude that networks of neurons with NMDAR mediated feedback excitation can generate self-sustained activity following brief synaptic excitation. The frequency of activity is limited by the kinetics of the neuron membrane channels and can be stopped by brief inhibitory input. Network activity can be rhythmic at lower frequencies if the neurons are electrically coupled. Our key finding is that excitatory synaptic feedback within a population of neurons can produce switchable, stable, sustained firing without synaptic inhibition.     

Conventional measures of intrinsic excitability are poor estimators of neuronal activity under realistic synaptic inputs

Activity-dependent regulation of intrinsic excitability has been shown to greatly contribute to the overall plasticity of neuronal circuits. Such neuroadaptations are commonly investigated in patch clamp experiments using current step stimulation and the resulting input-output functions are analyzed to quantify alterations in intrinsic excitability. However, it is rarely addressed, how such changes translate to the function of neurons when they operate under natural synaptic inputs. Still, it is reasonable to expect that a strong correlation and near proportional relationship exist between static firing responses and those evoked by synaptic drive. We challenge this view by performing a high-yield electrophysiological analysis of cultured mouse hippocampal neurons using both standard protocols and simulated synaptic inputs via dynamic clamp. We find that under these conditions the neurons exhibit vastly different firing responses with surprisingly weak correlation between static and dynamic firing intensities. These contrasting responses are regulated by two intrinsic K-currents mediated by Kv1 and K_ir channels, respectively. Pharmacological manipulation of the K-currents produces differential regulation of the firing output of neurons. Static firing responses are greatly increased in stuttering type neurons under blocking their Kv1 channels, while the synaptic responses of the same neurons are less affected. Pharmacological blocking of K_ir-channels in delayed firing type neurons, on the other hand, exhibit the opposite effects. Our subsequent computational model simulations confirm the findings in the electrophysiological experiments and also show that adaptive changes in the kinetic properties of such currents can even produce paradoxical regulation of the firing output.     

Predictive features of persistent activity emergence in regular spiking and intrinsic bursting model neurons

Proper functioning of working memory involves the expression of stimulus-selective persistent activity in pyramidal neurons of the prefrontal cortex (PFC), which refers to neural activity that persists for seconds beyond the end of the stimulus. The mechanisms which PFC pyramidal neurons use to discriminate between preferred vs. neutral inputs at the cellular level are largely unknown. Moreover, the presence of pyramidal cell subtypes with different firing patterns, such as regular spiking and intrinsic bursting, raises the question as to what their distinct role might be in persistent firing in the PFC. Here, we use a compartmental modeling approach to search for discriminatory features in the properties of incoming stimuli to a PFC pyramidal neuron and/or its response that signal which of these stimuli will result in persistent activity emergence. Furthermore, we use our modeling approach to study cell-type specific differences in persistent activity properties, via implementing a regular spiking (RS) and an intrinsic bursting (IB) model neuron. We identify synaptic location within the basal dendrites as a feature of stimulus selectivity. Specifically, persistent activity-inducing stimuli consist of activated synapses that are located more distally from the soma compared to non-inducing stimuli, in both model cells. In addition, the action potential (AP) latency and the first few inter-spike-intervals of the neuronal response can be used to reliably detect inducing vs. non-inducing inputs, suggesting a potential mechanism by which downstream neurons can rapidly decode the upcoming emergence of persistent activity. While the two model neurons did not differ in the coding features of persistent activity emergence, the properties of persistent activity, such as the firing pattern and the duration of temporally-restricted persistent activity were distinct. Collectively, our results pinpoint to specific features of the neuronal response to a given stimulus that code for its ability to induce persistent activity and predict differential roles of RS and IB neurons in persistent activity expression.     

Different Stimuli,Different Spatial Codes: A Visual Map and an Auditory Rate Code for Oculomotor Space in the Primate Superior Colliculus

Maps are a mainstay of visual, somatosensory, and motor coding in many species. However, auditory maps of space have not been reported in the primate brain. Instead, recent studies have suggested that sound location may be encoded via broadly responsive neurons whose firing rates vary roughly proportionately with sound azimuth. Within frontal space, maps and such rate codes involve different response patterns at the level of individual neurons. Maps consist of neurons exhibiting circumscribed receptive fields, whereas rate codes involve open-ended response patterns that peak in the periphery. This coding format discrepancy therefore poses a potential problem for brain regions responsible for representing both visual and auditory information. Here, we investigated the coding of auditory space in the primate superior colliculus(SC), a structure known to contain visual and oculomotor maps for guiding saccades. We report that, for visual stimuli, neurons showed circumscribed receptive fields consistent with a map, but for auditory stimuli, they had open-ended response patterns consistent with a rate or level-of-activity code for location. The discrepant response patterns were not segregated into different neural populations but occurred in the same neurons. We show that a read-out algorithm in which the site and level of SC activity both contribute to the computation of stimulus location is successful at evaluating the discrepant visual and auditory codes, and can account for subtle but systematic differences in the accuracy of auditory compared to visual saccades. This suggests that a given population of neurons can use different codes to support appropriate multimodal behavior.