Auditory forebrain neurons track temporal features of time-warped natural stimuli.

A fundamental challenge for sensory systems is to recognize natural stimuli despite stimulus variations. A compelling example occurs in speech, where the auditory system can recognize words spoken at a wide range of speeds. To date, there have been more computational models for time-warp invariance than experimental studies that investigate responses to time-warped stimuli at the neural level. Here, we address this problem in the model system of zebra finches anesthetized with urethane. In behavioral experiments, we found high discrimination accuracy well beyond the observed natural range of song variations. We artificially sped up or slowed down songs (preserving pitch) and recorded auditory responses from neurons in field L, the avian primary auditory cortex homolog. We found that field L neurons responded robustly to time-warped songs, tracking the temporal features of the stimuli over a broad range of warp factors. Time-warp invariance was not observed per se, but there was sufficient information in the neural responses to reliably classify which of two songs was presented. Furthermore, the average spike rate was close to constant over the range of time warps, contrary to recent modeling predictions. We discuss how this response pattern is surprising given current computational models of time-warp invariance and how such a response could be decoded downstream to achieve time-warp-invariant recognition of sounds.

Modifying spiking precision in conductance-based neuronal models.

The temporal precision of a neuron's spiking can be characterized by calculating its "jitter," defined as the standard deviation of the timing of individual spikes in response to repeated presentations of a stimulus. Sub-millisecond jitters have been measured for neurons in a variety of experimental systems and appear to be functionally important in some instances. We have investigated how modifying a neuron's maximal conductances affects jitter using the leaky integrate-and-fire (LIF) model and an eight-conductance Hodgkin-Huxley type (HH8) model. We observed that jitter can be largely understood in the LIF model in terms of the neuron's filtering properties. In the HH8 model we found the role of individual conductances in determining jitter to be complicated and dependent on the model's spiking properties. Distinct behaviors were observed for populations with slow (<11.5 Hz) and fast (>11.5 Hz) spike rates and appear to be related to differences in a particular channel's activity at times just before spiking occurs.

Competing sound sources reveal spatial effects in cortical processing.

Why is spatial tuning in auditory cortex weak, even though location is important to object recognition in natural settings? This question continues to vex neuroscientists focused on linking physiological results to auditory perception. Here we show that the spatial locations of simultaneous, competing sound sources dramatically influence how well neural spike trains recorded from the zebra finch field L (an analog of mammalian primary auditory cortex) encode source identity. We find that the location of a birdsong played in quiet has little effect on the fidelity of the neural encoding of the song. However, when the song is presented along with a masker, spatial effects are pronounced. For each spatial configuration, a subset of neurons encodes song identity more robustly than others. As a result, competing sources from different locations dominate responses of different neural subpopulations, helping to separate neural responses into independent representations. These results help elucidate how cortical processing exploits spatial information to provide a substrate for selective spatial auditory attention.

Neuron-specific stimulus masking reveals interference in spike timing at the cortical level.

The auditory system is capable of robust recognition of sounds in the presence of competing maskers (e.g., other voices or background music). This capability arises despite the fact that masking stimuli can disrupt neural responses at the cortical level. Since the origins of such interference effects remain unknown, in this study, we work to identify and quantify neural interference effects that originate due to masking occurring within and outside receptive fields of neurons. We record from single and multi-unit auditory sites from field L, the auditory cortex homologue in zebra finches. We use a novel method called spike timing-based stimulus filtering that uses the measured response of each neuron to create an individualized stimulus set. In contrast to previous adaptive experimental approaches, which have typically focused on the average firing rate, this method uses the complete pattern of neural responses, including spike timing information, in the calculation of the receptive field. When we generate and present novel stimuli for each neuron that mask the regions within the receptive field, we find that the time-varying information in the neural responses is disrupted, degrading neural discrimination performance and decreasing spike timing reliability and sparseness. We also find that, while removing stimulus energy from frequency regions outside the receptive field does not significantly affect neural responses for many sites, adding a masker in these frequency regions can nonetheless have a significant impact on neural responses and discriminability without a significant change in the average firing rate. These findings suggest that maskers can interfere with neural responses by disrupting stimulus timing information with power either within or outside the receptive fields of neurons.

A robust and biologically plausible spike pattern recognition network.

The neural mechanisms that enable recognition of spiking patterns in the brain are currently unknown. This is especially relevant in sensory systems, in which the brain has to detect such patterns and recognize relevant stimuli by processing peripheral inputs; in particular, it is unclear how sensory systems can recognize time-varying stimuli by processing spiking activity. Because auditory stimuli are represented by time-varying fluctuations in frequency content, it is useful to consider how such stimuli can be recognized by neural processing. Previous models for sound recognition have used preprocessed or low-level auditory signals as input, but complex natural sounds such as speech are thought to be processed in auditory cortex, and brain regions involved in object recognition in general must deal with the natural variability present in spike trains. Thus, we used neural recordings to investigate how a spike pattern recognition system could deal with the intrinsic variability and diverse response properties of cortical spike trains. We propose a biologically plausible computational spike pattern recognition model that uses an excitatory chain of neurons to spatially preserve the temporal representation of the spike pattern. Using a single neural recording as input, the model can be trained using a spike-timing-dependent plasticity-based learning rule to recognize neural responses to 20 different bird songs with >98% accuracy and can be stimulated to evoke reverse spike pattern playback. Although we test spike train recognition performance in an auditory task, this model can be applied to recognize sufficiently reliable spike patterns from any neuronal system.

Analyzing variability in neural responses to complex natural sounds in the awake songbird.

Studies of auditory processing in awake, behaving songbirds allow for the possibility of new classes of experiments, including those involving attention and plasticity. Detecting and determining the significance of plasticity, however, requires assessing the intrinsic variability in neural responses. Effects such as rapid plasticity have been investigated in the auditory system through the use of the spectrotemporal receptive field (STRF), a characterization of the properties of sounds to which a neuron best responds. Here we investigated neural response variability in awake recordings obtained from zebra finch field L, the analog of the primary auditory cortex. To quantify the level of variability in the neural recordings, we used three similarity measures: an STRF-based metric, a spike-train correlation-based metric, and a spike-train discrimination-based metric. We then extracted a number of parameters from these measures, quantifying how they fluctuated over time. Our results indicate that 1) awake responses are quite stable over time; 2) the different measures of response are complementary-specifically, the spike-train-based measures yield new information complementary to the STRF; and 3) different STRF parameters show distinct levels of variability. These results provide critical constraints for the design of robust decoding strategies and novel experiments on attention and plasticity in the awake songbird.

A biologically plausible computational model for auditory object recognition.

Object recognition is a task of fundamental importance for sensory systems. Although this problem has been intensively investigated in the visual system, relatively little is known about the recognition of complex auditory objects. Recent work has shown that spike trains from individual sensory neurons can be used to discriminate between and recognize stimuli. Multiple groups have developed spike similarity or dissimilarity metrics to quantify the differences between spike trains. Using a nearest-neighbor approach the spike similarity metrics can be used to classify the stimuli into groups used to evoke the spike trains. The nearest prototype spike train to the tested spike train can then be used to identify the stimulus. However, how biological circuits might perform such computations remains unclear. Elucidating this question would facilitate the experimental search for such circuits in biological systems, as well as the design of artificial circuits that can perform such computations. Here we present a biologically plausible model for discrimination inspired by a spike distance metric using a network of integrate-and-fire model neurons coupled to a decision network. We then apply this model to the birdsong system in the context of song discrimination and recognition. We show that the model circuit is effective at recognizing individual songs, based on experimental input data from field L, the avian primary auditory cortex analog. We also compare the performance and robustness of this model to two alternative models of song discrimination: a model based on coincidence detection and a model based on firing rate.

Invariance and sensitivity to intensity in neural discrimination of natural sounds.

Intensity variation poses a fundamental problem for sensory discrimination because changes in the response of sensory neurons as a result of stimulus identity, e.g., a change in the identity of the speaker uttering a word, can potentially be confused with changes resulting from stimulus intensity, for example, the loudness of the utterance. Here we report on the responses of neurons in field L, the primary auditory cortex homolog in songbirds, which allow for accurate discrimination of birdsongs that is invariant to intensity changes over a large range. Such neurons comprise a subset of a population that is highly diverse, in terms of both discrimination accuracy and intensity sensitivity. We find that the neurons with a high degree of invariance also display a high discrimination performance, and that the degree of invariance is significantly correlated with the reproducibility of spike timing on a short time scale and the temporal sparseness of spiking activity. Our results indicate that a temporally sparse spike timing-based code at a primary cortical stage can provide a substrate for intensity-invariant discrimination of natural sounds.

Information rate and spike-timing precision of proprioceptive afferents.

Proprioception in the first two joints of crustacean limbs is mediated by chordotonal organs that utilize spike-mediated information coding and transmission and by nonspiking proprioceptive afferents that use graded transmission at information rates in excess of 2,500 bits/s. Chordotonal organs operate in parallel with the graded receptors, but the information rates of the spiking chordotonal afferents have not been previously determined. Lower-bound estimates of chordotonal afferent information rates were calculated using stimulus reconstruction, which assumes linear encoding of the stimulus. The information rate was also directly estimated from the spike train entropy, which makes no a priori assumptions with respect to the coding scheme used by the system. Lower-bound information rate estimates ranged from 43 to 69 bits/s, whereas the direct estimates ranged from 24 to 278 bits/s. Comparison of both estimates derived from the same data set indicates that a linear decoder could recover an average of 59% of the information from the spike train. Afferent spike timing was found to be extremely precise, with spikes evoked with an average timing jitter of 0.55 ms. Information rate was correlated with the mean jitter and the noise entropy of the spike train could be predicted from the mean firing rate and mean jitter. Direct stimulation of single afferents by current injection into the soma revealed that the average timing jitter was <0.1 ms, indicating that intrinsic membrane properties, spike generation, and mechanotransduction mechanisms are the major sources of timing jitter in this system.

Neuromodulation of spike-timing precision in sensory neurons.

The neuropeptide allatostatin decreases the spike rate in response to time-varying stretches of two different crustacean mechanoreceptors, the gastropyloric receptor 2 in the crab Cancer borealis and the coxobasal chordotonal organ (CBCTO) in the crab Carcinus maenas. In each system, the decrease in firing rate is accompanied by an increase in the timing precision of spikes triggered by discrete temporal features in the stimulus. This was quantified by calculating the standard deviation or "jitter" in the times of individual identified spikes elicited in response to repeated presentations of the stimulus. Conversely, serotonin increases the firing rate but decreases the timing precision of the CBCTO response. Intracellular recordings from the afferents of this receptor demonstrate that allatostatin increases the conductance of the neurons, consistent with its inhibitory action on spike rate, whereas serotonin decreases the overall membrane conductance. We conclude that spike-timing precision of mechanoreceptor afferents in response to dynamic stimulation can be altered by neuromodulators acting directly on the afferent neurons.

Profiling of neuropeptides released at the stomatogastric ganglion of the crab, Cancer borealis with mass spectrometry.

Studies of release under physiological conditions provide more direct data about the identity of neuromodulatory signaling molecules than studies of tissue localization that cannot distinguish between processing precursors and biologically active neuropeptides. We have identified neuropeptides released by electrical stimulation of nerves that contain the axons of the modulatory projection neurons to the stomatogastric ganglion of the crab, Cancer borealis. Preparations were bathed in saline containing a cocktail of peptidase inhibitors to minimize peptide degradation. Both electrical stimulation of projection nerves and depolarization with high K+ saline were used to evoke release. Releasates were desalted and then identified by mass using MALDI-TOF (matrix-assisted laser desorption/ionization-time-of-flight) mass spectrometry. Both previously known and novel peptides were detected. Subsequent to electrical stimulation proctolin, Cancer borealis tachykinin-related peptide (CabTRP), FVNSRYa, carcinustatin-8, allatostatin-3 (AST-3), red pigment concentrating hormone, NRNFLRFa, AST-5, SGFYANRYa, TNRNFLRFa, AST-9, orcomyotropin-related peptide, corazonin, Ala13-orcokinin, and Ser9-Val13-orcokinin were detected. Some of these were also detected after high K+ depolarization. Release was calcium dependent. In summary, we have shown release of the neuropeptides thought to play an important neuromodulatory role in the stomatogastric ganglion, as well as numerous other candidate neuromodulators that remain to be identified.

Mass spectrometric investigation of the neuropeptide complement and release in the pericardial organs of the crab, Cancer borealis.

The crustacean stomatogastric ganglion (STG) is modulated by both locally released neuroactive compounds and circulating hormones. This study presents mass spectrometric characterization of the complement of peptide hormones present in one of the major neurosecretory structures, the pericardial organs (POs), and the detection of neurohormones released from the POs. Direct peptide profiling of Cancer borealis PO tissues using matrix-assisted laser desorption/ionization (MALDI) time-of-flight (TOF) mass spectrometry (MS) revealed many previously identified peptides, including proctolin, red pigment concentrating hormone (RPCH), crustacean cardioactive peptide (CCAP), several orcokinins, and SDRNFLRFamide. This technique also detected corazonin, a well-known insect hormone, in the POs for the first time. However, most mass spectral peaks did not correspond to previously known peptides. To characterize and identify these novel peptides, we performed MALDI postsource decay (PSD) and electrospray ionization (ESI) MS/MS de novo sequencing of peptides fractionated from PO extracts. We characterized a truncated form of previously identified TNRNFLRFamide, NRNFLRFamide. In addition, we sequenced five other novel peptides sharing a common C-terminus of RYamide from the PO tissue extracts. High K+ depolarization of isolated POs released many peptides present in this tissue, including several of the novel peptides sequenced in the current study.

Alternative to hand-tuning conductance-based models: construction and analysis of databases of model neurons.

Conventionally, the parameters of neuronal models are hand-tuned using trial-and-error searches to produce a desired behavior. Here, we present an alternative approach. We have generated a database of about 1.7 million single-compartment model neurons by independently varying 8 maximal membrane conductances based on measurements from lobster stomatogastric neurons. We classified the spontaneous electrical activity of each model neuron and its responsiveness to inputs during runtime with an adaptive algorithm and saved a reduced version of each neuron's activity pattern. Our analysis of the distribution of different activity types (silent, spiking, bursting, irregular) in the 8-dimensional conductance space indicates that the coarse grid of conductance values we chose is sufficient to capture the salient features of the distribution. The database can be searched for different combinations of neuron properties such as activity type, spike or burst frequency, resting potential, frequency-current relation, and phase-response curve. We demonstrate how the database can be screened for models that reproduce the behavior of a specific biological neuron and show that the contents of the database can give insight into the way a neuron's membrane conductances determine its activity pattern and response properties. Similar databases can be constructed to explore parameter spaces in multicompartmental models or small networks, or to examine the effects of changes in the voltage dependence of currents. In all cases, database searches can provide insight into how neuronal and network properties depend on the values of the parameters in the models.

Differential and history-dependent modulation of a stretch receptor in the stomatogastric system of the crab, Cancer borealis.

Neuromodulators can modify the magnitude and kinetics of the response of a sensory neuron to a stimulus. Six neuroactive substances modified the activity of the gastropyloric receptor 2 (GPR2) neuron of the stomatogastric nervous system (STNS) of the crab Cancer borealis during muscle stretch. Stretches were applied to the gastric mill 9 (gm9) and the cardio-pyloric valve 3a (cpv3a) muscles. SDRNFLRFamide and dopamine had excitatory effects on GPR2. Serotonin, GABA, and the peptide allatostatin-3 (AST) decreased GPR2 firing during stretch. Moreover, SDRNFLRFamide and TNRNFLRFamide increased the unstimulated spontaneous firing rate, whereas AST and GABA decreased it. The actions of AST and GABA were amplitude- and history-dependent. In fully recovered preparations, AST and GABA decreased the response to small-amplitude stretches proportionally more than to those evoked by large-amplitude stretches. For large-amplitude stretches, the effects of AST and GABA were more pronounced as the number of recent stretches increased. The modulators that affected the stretch-induced GPR2 firing rate were also tested when the neuron was operating in a bursting mode of activity. Application of SDRNFLRFamide increased the bursting frequency transiently, whereas high concentrations of serotonin, AST, and GABA abolished bursting altogether. Together these data demonstrate that the effects of neuromodulators depend on the previous activity and current state of the sensory neuron.