Tetrodes markedly improve the reliability and yield of multiple single-unit isolation from multi-unit recordings in cat striate cortex

Tracking neurons recorded from tetrodes across time

Tetrodes allow isolation of multiple neurons at a single recording site by clustering spikes. Due to electrode drift and perhaps due to time-varying neuronal properties, positions and shapes of clusters change in time. As data is typically collected in sequential files, to track neurons across files one has to decide which clusters from different files belong to the same neuron. We report on a semi-automated neuron tracking procedure that uses computed similarities between the mean spike waveforms of the clusters. The clusters with the most similar waveforms are assigned to the same neuron, provided their similarity exceeds a threshold. To set this threshold, we calculate two distributions: of within-file similarities, and of best matches in the across adjacent file similarities. The threshold is set to the value that optimally separates the two distributions. We compare different measures of similarity (metrics) by their ability to separate these distributions. We find that these metrics do not differ drastically in their performance, but that taking into account the cross-channel noise correlation significantly improves performance of all metrics. We also demonstrate the method on an independent dataset and show that neurons, as assigned by the procedure, have consistent physiological properties across files.

Nanostructured Surface Modification of Ceramic-Based Microelectrodes to Enhance Biocompatibility for a Direct Brain-Machine Interface

Many different types of microelectrodes have been developed for use as a direct Brain-Machine Interface (BMI) to chronically recording single neuron action potentials from ensembles of neurons. Unfortunately, the recordings from these microelectrode devices are not consistent and often last for only a few weeks. For most microelectrode types, the loss of these recordings is not due to failure of the electrodes but most likely due to damage to surrounding tissue that results in the formation of nonconductive glial-scar. Since the extracellular matrix consists of nanostructured microtubules, we have postulated that neurons may prefer a more complex surface structure than the smooth surface typical of thin-film microelectrodes. We, therefore, investigated the suitability of a nano-porous silicon surface layer to increase the biocompatibility of our thin film ceramic-insulated multisite electrodes. In-vitro testing demonstrated, for the first time, decreased adhesion of astrocytes and increased extension of neurites from pheochromocytoma cells on porous silicon surfaces compared to smooth silicon sufaces. Moreover, nano-porous surfaces were more biocompatible than macroporous surfaces. Collectively, these results support our hypothesis that nano-porous silicon may be an ideal material to improve biocompatibility of chronically implanted microelectrodes. We next developed a method to apply nano-porous surfaces to ceramic insulated, thin-film, microelectrodes and tested them in vivo. Chronic testing demonstrated that the nano-porous surface modification did not alter the electrical properties of the recording sites and did not interfere with proper functioning of the microelectrodes in vivo.

Ceramic-based multisite electrode arrays for chronic single-neuron recording

A method is described for the manufacture of a microelectrode array for chronic, multichannel, single neuron recording. The ceramic-based, multisite electrode array has four recording sites patterned onto a ceramic shaft the size of a single typical microwire electrode. The sites and connecting wires are applied to the ceramic substrate using a reverse photolithographic procedure. Recording sites (22 x 80 microm) are separated by 200 microm along the shaft. A layer of alumina insulation is applied over the whole array (exclusive of recording sites) by ion-beam assisted deposition. These arrays were capable of recording single neuron activity from each of their recording sites for at least three weeks during chronic implantation in the somatosensory cortex of rats, and several sites had recordings that lasted for more than 8 weeks. The vertical arrangement of the recording sites on these electrodes is ideal for simultaneously recording across the different layers of brain areas such as the cerebral cortex and hippocampus in chronic preparations.

Telemetric recordings of single neuron activity and visual scenes in monkeys walking in an open field

This paper describes a portable recording system and methods for obtaining chronic recordings of single units and tracking rhesus monkey behavior in an open field. The integrated system consists of four major components: (1) microelectrode assembly; (2) head-stage; (3) recording station; and (4) data storage station, the first three of which are carried by the monkey and weigh 800 g. Our system provides synchronized video and electrophysiological signals, which are transmitted by a wireless system to a distance of 50 m. Its major advantages are that neuronal recordings are made in freely moving monkeys, and well-separated action potentials with amplitude five times higher than the background noise are usually recorded and readily kept for many hours. Using this system, we were able to study "place cells" in non-human primate brains. The described methods provide a new way to examine correlations between single neuron activity and primate behaviors, and can also be used to study the cellular basis of social behaviors in non-human primates.

Contextual modulation of synchronization to random dots in the cat visual cortex

Synchronization of neuronal activity has been proposed as a binding mechanism for integration of image properties into one coherent percept. In the present study, we investigated the contextual modulation of synchronization to random dot patterns. Coherent motion of random dots evoked well synchronized responses in area 17 of anaesthetized cats when the stimulus was presented in the compound receptive field of recorded sites. Gradually changing the directional coherence of random dots in the surround while maintaining fully coherent motion of the stimulus in the receptive field significantly suppressed synchronization of neuronal activity for some stimulus conditions. However, usually one or two peaks of increased synchronization were found in the surround coherence tuning curves with low (8-12%) and/or moderate (25-50%) coherence in the surround. At the population level, synchronization was significantly depressed with incoherent motion in the receptive field and when both the surround and the receptive field were jointly stimulated with 0% coherence. The intriguing finding was the discovery of two distinct groups of cells with opposite synchronization changes dependent on the presence or absence of significant synchronization in their spontaneous activity. The latter group of neurons showed peaks of increased synchronization with lower surround coherence, thus probably being more sensitive to the direction of the surround motion. Overall, our findings support the notion that binding of stimulus properties can be achieved by synchronized activity of cortical cells. However, our findings go further than the original hypothesis of feature binding by synchrony to show that synchronization of cortical activity may be directly related to the decision making processes, which in turn are related to the threshold of perception of coherent motion.

Large-scale recording of neuronal ensembles

How does the brain orchestrate perceptions, thoughts and actions from the spiking activity of its neurons? Early single-neuron recording research treated spike pattern variability as noise that needed to be averaged out to reveal the brain's representation of invariant input. Another view is that variability of spikes is centrally coordinated and that this brain-generated ensemble pattern in cortical structures is itself a potential source of cognition. Large-scale recordings from neuronal ensembles now offer the opportunity to test these competing theoretical frameworks. Currently, wire and micro-machined silicon electrode arrays can record from large numbers of neurons and monitor local neural circuits at work. Achieving the full potential of massively parallel neuronal recordings, however, will require further development of the neuron-electrode interface, automated and efficient spike-sorting algorithms for effective isolation and identification of single neurons, and new mathematical insights for the analysis of network properties.

Functional stability of dorsolateral prefrontal neurons

Stable multiday recordings from the dorsolateral prefrontal cortex of 2 monkeys performing 2 Go/NoGo visual-discrimination tasks (one requiring well-learned responses, the other requiring learning) demonstrate that the majority of prefrontal neurons were "functionally stable". Recordings were made using a series of removable microdrives, each implanted for 3-6 mo, housing independently mobile electrodes. Action potential waveforms of 94 neurons were stable over 2-9 days; 66/94 (70%) of these cells responded each day, 22/94 (23%) never responded significantly, and 6/94 (6%) responded one day but not the next. Of 66 responsive neurons, 55 were selective for either Go or NoGo trials, individual stimuli, or eye movements. This selectivity was functionally stable (i.e., maintained) for 46/55 neurons across all recording days. Functional stability was also noted in terms of response strength (baseline firing rates compared with poststimulation firing rates) and event-related response timing. Two neurons with consistent responses in familiar testing conditions responded flexibly when the monkeys learned to make correct responses to novel stimuli. We conclude that the majority of prefrontal neurons were functionally stable during the performance of well-learned tasks. Such stability may be a general property of prefrontal neurons, given that neurons with 4 different types of task selectivity were found to be functionally stable. Conceptually similar studies based on long-term recordings in other cortical regions reached similar conclusions, suggesting that neurons throughout the brain are functionally stable.

Editing trains of action potentials from multi-electrode arrays

When recording from multi-electrode arrays, only a short period around the time of a threshold crossing is generally saved for later analysis. Then, waveforms are often sorted automatically to identify templates of spikes from individual neurons near an electrode. As spikes sum from different neurons and noise is present, some spikes may be missed and others erroneously accepted. This paper describes methods for identifying and correcting errors in recorded spike trains to recover the pattern of spikes from each neuron as faithfully as possible. These methods are complementary to, but distinct from methods to reconstruct waveforms that arise from summation of individual templates that overlap one another. Our methods are based on the local statistics of the firing rates or inter-spike intervals and the methods work best for neurons that fire regularly (small standard deviation relative to the mean interval). First, we test whether accepting more spikes, whose waveforms are close to the templates that have been identified, will increase the regularity or smoothness of the firing rates. Then, after accepting spikes that increase regularity, we test whether individual intervals are sufficiently longer (or shorter) than their neighbors to identify spikes that have been omitted (or accepted) erroneously. The methods are tested on simulated spike trains, where spikes have been inserted or deleted at random, and on spike trains recorded from multi-electrode arrays in dorsal root ganglia of cats walking on a treadmill.

Interpreting the BOLD Signal

The development of functional magnetic resonance imaging (fMRI) has brought together a broad community of scientists interested in measuring the neural basis of the human mind. Because fMRI signals are an indirect measure of neural activity, interpreting these signals to make deductions about the nervous system requires some understanding of the signaling mechanisms. We describe our current understanding of the causal relationships between neural activity and the blood-oxygen-level-dependent (BOLD) signal, and we review how these analyses have challenged some basic assumptions that have guided neuroscience. We conclude with a discussion of how to use the BOLD signal to make inferences about the neural signal.

Learning-related changes in response patterns of prefrontal neurons during instrumental conditioning

A crucial aspect of organizing goal-directed behavior is the ability to form neural representations of relationships between environmental stimuli, actions and reinforcement. Very little is known yet about the neural encoding of response-reward relationships, a process which is deemed essential for purposeful behavior. To investigate this, tetrode recordings were made in the medial prefrontal cortex (PFC) of rats performing a Go-NoGo task. After task acquisition, a subset of neurons showed a sustained change in firing during the rewarded action sequence that was triggered by a specific visual cue. When these changes were monitored in the course of learning, they were seen to develop in parallel with the behavioral learning curve and were highly sensitive to a switch in reward contingencies. These sustained changes correlated with the reward-associated action sequence, not with sensory or reward-predicting properties of the cue or individual motor acts per se. This novel type of neural plasticity may contribute to the formation of response-reinforcer associations and of behavioral strategies for guiding goal-directed action.

Improved spike-sorting by modeling firing statistics and burst-dependent spike amplitude attenuation: a Markov chain Monte Carlo approach

Spike-sorting techniques attempt to classify a series of noisy electrical waveforms according to the identity of the neurons that generated them. Existing techniques perform this classification ignoring several properties of actual neurons that can ultimately improve classification performance. In this study, we propose a more realistic spike train generation model. It incorporates both a description of "nontrivial" (i.e., non-Poisson) neuronal discharge statistics and a description of spike waveform dynamics (e.g., the events amplitude decays for short interspike intervals). We show that this spike train generation model is analogous to a one-dimensional Potts spin-glass model. We can therefore tailor to our particular case the computational methods that have been developed in fields where Potts models are extensively used, including statistical physics and image restoration. These methods are based on the construction of a Markov chain in the space of model parameters and spike train configurations, where a configuration is defined by specifying a neuron of origin for each spike. This Markov chain is built such that its unique stationary density is the posterior density of model parameters and configurations given the observed data. A Monte Carlo simulation of the Markov chain is then used to estimate the posterior density. We illustrate the way to build the transition matrix of the Markov chain with a simple, but realistic, model for data generation. We use simulated data to illustrate the performance of the method and to show that this approach can easily cope with neurons firing doublets of spikes and/or generating spikes with highly dynamic waveforms. The method cannot automatically find the "correct" number of neurons in the data. User input is required for this important problem and we illustrate how this can be done. We finally discuss further developments of the method.

Prospective and retrospective memory coding in the hippocampus

The effect of memory on hippocampal neuronal activity was assessed as rats performed a spatial task that was impaired by fornix lesions. The influences of current location, recently entered places, and places about to be entered were compared. Three new findings emerged. (1) Current, retrospective, and prospective coding were common and recorded simultaneously in neural ensembles. (2) The origin of journeys influenced firing even when rats made detours, showing that recent memory could modulate neuronal activity more than spatial trajectory. (3) Diminished retrospective coding and, more markedly, reduced prospective coding in error trials suggested that the neuronal signal was important for task performance. The population of hippocampal neurons thus encoded information about the recent past, the present, and the imminent future, consistent with a neuronal mechanism for episodic memory.

Visual resolution with epi-retinal electrical stimulation estimated from activation profiles in cat visual cortex

Blinds with receptor degeneration can perceive localized phosphenes in response to focal electrical epi-retinal stimuli. To avoid extensive basic stimulation tests in human patients, we developed techniques for estimating visual spatial resolution in anesthetized cats. Electrical epi-retinal and visual stimulation was combined with multiple-site retinal and cortical microelectrode recordings of local field potentials (LFPs) from visual areas 17 and 18. Classical visual receptive fields were characterized for retinal and cortical recording sites using multifocal visual stimulation combined with stimulus–response cross-correlation. We estimated visual spatial resolution from the size of the cortical activation profiles in response to single focal stimuli. For comparison, we determined activation profiles in response to visual stimuli at the same retinal location. Activation profiles were single peaked or multipeaked. In multipeaked profiles, the peak locations coincided with discontinuities in cortical retinotopy. Location and width of cortical activation profiles were distinct for retinal stimulation sites. On average, the activation profiles had a size of 1.28 ± 0.03 mm cortex. Projected to visual space this corresponds to a spatial resolution of 1.49 deg ± 0.04 deg visual angle. Best resolutions were 0.5 deg at low and medium stimulation currents corresponding to a visus of 1/30. Higher stimulation currents caused lower spatial, but higher temporal resolution (up to 70 stimuli/s). In analogy to the receptive-field concept in visual space, we defined and characterized electrical receptive fields. As our estimates of visual resolutions are conservative, we assume that a visual prosthesis will induce phosphenes at least at this resolution. This would enable visuomotor coordinations and object recognition in many indoor and outdoor situations of daily life.

Differential modulation of CA1 and dentate gyrus interneurons during exploration of novel environments

Parallel recordings of hippocampal principal cells and interneurons were obtained as rats foraged in familiar and adjacent, novel environments. Firing rates of each cell type were assessed as a function of spatial location. Many CA1 interneurons exhibited large decreases in activity in the novel compared with the familiar environment. Dentate gyrus interneurons, however, were much more likely to exhibit large increases in firing in the novel environment. Neither effect was correlated with basic interneuron discharge properties such as degree of theta modulation, baseline firing rate or degree of spatially modulated discharge. Both CA1 and dentate gyrus interneuron rate changes extended into regions of the familiar environment bordering the novel environment. Principal cells in CA1 and dentate gyrus exhibited similar patterns of place specific activity each being indicative of incorporation of novel spatial information into the spatial representation of the familiar environment. The data indicate that inhibitory networks in the CA1 and dentate gyrus areas are modulated in a divergent fashion during the acquisition of novel spatial information and that interneuron activities can be used to detect those regions of an environment subject to redistribution of principal cell spatial activity patterns.

Neuronal interactions are higher in the cortex than thalamus in the somatosensory pathway

Previous studies have shown significant correlated discharges (noise correlation) and synergistic information coding among adjacent cortical neurons. In order to investigate whether such interactions are present at an earlier stage of sensory processing, we compared noise correlation and synergistic information transmission in the ventral posterolateral nucleus (VPLn) of thalamus and primary somatosensory cortex (SI) of anesthetized rats. A hind paw was stimulated electrically and responses of several neighboring neurons were recorded simultaneously with a tetrode. Analyses indicated that noise correlation in the SI was about four times higher than in the VPLn, and, interestingly, it was significantly reduced following sensory stimulation in both regions. Spike count distributions of individual VPLn units contained higher amounts of information about the delivery of external stimulation compared with those of SI units. When simultaneously recorded units were considered together, transmission of information was more interactive (synergistic or redundant) among SI than VPLn units. On average, information transmission was independent in the VPLn, but synergistic in the SI. The difference in synergistic information coding was largely attributable to different levels of noise correlation and their modulation by external sensory stimulation. These results indicate that neuronal interactions are relatively low at the thalamic level, but much enhanced at the cortical level along the somatosensory pathway. The enhanced neuronal interactions in the cortex may reflect the role of cortex in extracting higher features of sensory stimuli.

Massively parallel recording of unit and local field potentials with silicon-based electrodes

Parallel recording of neuronal activity in the behaving animal is a prerequisite for our understanding of neuronal representation and storage of information. Here we describe the development of micro-machined silicon microelectrode arrays for unit and local field recordings. The two-dimensional probes with 96 or 64 recording sites provided high-density recording of unit and field activity with minimal tissue displacement or damage. The on-chip active circuit eliminated movement and other artifacts and greatly reduced the weight of the headgear. The precise geometry of the recording tips allowed for the estimation of the spatial location of the recorded neurons and for high-resolution estimation of extracellular current source density. Action potentials could be simultaneously recorded from the soma and dendrites of the same neurons. Silicon technology is a promising approach for high-density, high-resolution sampling of neuronal activity in both basic research and prosthetic devices.

A microelectrode drive for long term recording of neurons in freely moving and chaired monkeys

An electrode drive is described for recordings of neurons in freely moving and chaired monkeys during the performance of behavioural tasks. The electrode drives are implanted for periods of up to 6 months, and can advance up to 42 electrodes using 14 independent drive mechanisms. The drive samples 288 points within a 12 mmx12 mm region, with 15 mm of electrode travel. Major advantages are that recordings are made in freely moving monkeys, and these recordings can be compared with those in chaired experiments; waveforms of single neurons are stable, enabling prolonged recordings of the same neurons across periods of days; recordings can be made throughout the brain, including the dorsolateral prefrontal cortex and hippocampus; the drive accommodates both sharp microelectrodes and fine wire assemblies such as tetrodes.

A new approach to spike sorting for multi-neuronal activities recorded with a tetrode--how ICA can be practical

Multi-neuronal recording with a tetrode is a powerful technique to reveal neuronal interactions in local circuits. However, it is difficult to detect precise spike timings among closely neighboring neurons because the spike waveforms of individual neurons overlap on the electrode when more than two neurons fire simultaneously. In addition, the spike waveforms of single neurons, especially in the presence of complex spikes, are often non-stationary. These problems limit the ability of ordinary spike sorting to sort multi-neuronal activities recorded using tetrodes into their single-neuron components. Though sorting with independent component analysis (ICA) can solve these problems, it has one serious limitation that the number of separated neurons must be less than the number of electrodes. Using a combination of ICA and the efficiency of ordinary spike sorting technique (k-means clustering), we developed an automatic procedure to solve the spike-overlapping and the non-stationarity problems with no limitation on the number of separated neurons. The results for the procedure applied to real multi-neuronal data demonstrated that some outliers which may be assigned to distinct clusters if ordinary spike-sorting methods were used can be identified as overlapping spikes, and that there are functional connections between a putative pyramidal neuron and its putative dendrite. These findings suggest that the combination of ICA and k-means clustering can provide insights into the precise nature of functional circuits among neurons, i.e. cell assemblies.

Hippocampal network patterns of activity in the mouse

Genetic engineering of the mouse brain allows investigators to address novel hypotheses in vivo. Because of the paucity of information on the network patterns of the mouse hippocampus, we investigated the electrical patterns in the behaving animal using multisite silicon probes and wire tetrodes. Theta (6-9 Hz) and gamma (40-100 Hz) oscillations were present during exploration and rapid eye movement sleep. Gamma power and theta power were comodulated and gamma power varied as a function of the theta cycle. Pyramidal cells and putative interneurons were phase-locked to theta oscillations. During immobility, consummatory behaviors and slow-wave sleep, sharp waves were present in cornu ammonis region CA1 of the hippocampus stratum radiatum associated with 140-200-Hz "ripples" in the pyramidal cell layer and population burst of CA1 neurons. In the hilus, large-amplitude "dentate spikes" occurred in association with increased discharge of hilar neurons. The amplitude of field patterns was larger in the mouse than in the rat, likely reflecting the higher neuron density in a smaller brain. We suggest that the main hippocampal network patterns are mediated by similar pathways and mechanisms in mouse and rat.

Simultaneous mapping of binocular and monocular receptive fields in awake monkeys for calibrating eye alignment in a dichoptical setup

We developed a modified Wheatstone stereoscope for simultaneous dichoptical and binocular stimulation in awake monkey. We, therefore, extended the conventional two-screen Wheatstone stereoscope to a setup with an additional third screen viewed binocularly via semi-transparent mirrors. With a sparse noise stimulation we mapped classical receptive field (CRF) positions via each screen independently but simultaneously. This was done for multiple recording positions (16 electrodes) at once in primary visual cortex based on multiple unit spike activity (MUA) and local field potentials (LFP), respectively. The technique can be used to (1) quickly and simultaneously determine binocular as well as left and right eye CRFs, including ocular dominance characteristics (net recording time for the given examples: approximately 2 min), (2) precisely adjust dichoptical stimulation by evaluating offsets between monocular and binocular CRF positions (average spatial incongruency between binocular and left/right eye stimulation after calibration: approximately 0.025 degrees visual angle), and (3) investigate left and right eye interaction in forming binocular CRFs. Due to the precise adjustment of the dichoptical and the simultaneous binocular stimulation investigations on the basis of stereo vision can be done with appropriate eye vergence alignment matching normal binocular viewing conditions in awake animals.

Automatic sorting for multi-neuronal activity recorded with tetrodes in the presence of overlapping spikes

Multi-neuronal recording is a powerful electrophysiological technique that has revealed much of what is known about the neuronal interactions in the brain. However, it is difficult to detect precise spike timings, especially synchronized simultaneous firings, among closely neighboring neurons recorded by one common electrode because spike waveforms overlap on the electrode when two or more neurons fire simultaneously. In addition, the non-Gaussian variability (nonstationarity) of spike waveforms, typically seen in the presence of so-called complex spikes, limits the ability to sort multi-neuronal activities into their single-neuron components. Because of these problems, the ordinary spike-sorting techniques often give inaccurate results. Our previous study has shown that independent component analysis (ICA) can solve these problems and separate single-neuron components from multi-neuronal recordings. The ICA has, however, one serious limitation that the number of separated neurons must be less than the number of electrodes. The present study combines the ICA and the efficiency of the ordinary spike-sorting technique (k-means clustering) to solve the spike-overlapping and the nonstationarity problems with no limitation on the number of single neurons to be separated. First, multi-neuronal activities are sorted into an overly large number of clusters by k-means clustering. Second, the sorted clusters are decomposed by ICA. Third, the decomposed clusters are progressively aggregated into a minimal set of putative single neurons based on similarities of basis vectors estimated by ICA. We applied the present procedure to multi-neuronal waveforms recorded with tetrodes composed of four microwires in the prefrontal cortex of awake behaving monkeys. The results demonstrate that there are functional connections among neighboring pyramidal neurons, some of which fire in a precise simultaneous manner and that precisely time-locked monosynaptic connections are working between neighboring pyramidal neurons and interneurons. Detection of these phenomena suggests that the present procedure can sort multi-neuronal activities, which include overlapping spikes and realistic non-Gaussian variability of spike waveforms, into their single-neuron components. We processed several types of synthesized data sets in this procedure and confirmed that the procedure was highly reliable and stable. The present method provides insights into the local circuit bases of excitatory and inhibitory interactions among neighboring neurons.

A 64(128)-channel multisite neuronal recording system

We used a recently described all-dry silicon etch process for SOI wafers to fabricate 64-site electrode arrays in stereotrode arrangement for acute cortical recordings. The fork-like probes are connected to preamplification units by flexible, Y-shaped interconnects. This facilitates maximal experimental flexibility for simultaneously recording from all available channels from the cortex of anaesthetised rats. Preconditioned signals are amplified by a novel modular main amp, which may be software or dial controlled. Signals are 16bit digitized, recorded, analyzed, stored and processed on a DSP-based modular data acquisition system. Digital data is processed, filtered and denoised on all up to (4*32) 128 channels based on an extremely fast wavelet transformation framework.

Selective impairment of hippocampal gamma oscillations in connexin-36 knock-out mouse in vivo

The physiological roles of neuronal gap junctions in the intact brain are not known. The recent generation of the connexin-36 knock-out (Cx36 KO) mouse has offered a unique opportunity to examine this problem. Recent in vitro recordings in Cx36 KO mice suggested that Cx36 gap junction contributes to various oscillatory patterns in the theta (approximately 5-10 Hz) and gamma (approximately 30-80 Hz) frequency ranges and affects certain aspects of high-frequency (>100 Hz) patterns. However, the relevance of these pharmacologically induced patterns to the intact brain is not known. We recorded field potentials and unit activity in the CA1 stratum pyramidale of the hippocampus in the behaving wild-type (WT) and Cx36 KO mice. Fast-field "ripple" oscillations (140-200 Hz) were present in both WT and KO mice and did not differ significantly in power, intraepisode frequency, or probability of occurrence. Thus, fast-field oscillations either may not require electrical synapses or may be mediated by a hitherto unknown class of gap junctions. Theta oscillations, recorded during either wheel running or rapid eye movement sleep, were not different either. However, the power in the gamma frequency band and the magnitude of theta-phase modulation of gamma power were significantly decreased in KO mice compared with WT controls during wheel running. This suggests that Cx36 interneuronal gap junctions selectively contribute to gamma oscillations.

An application of reversible-jump Markov chain Monte Carlo to spike classification of multi-unit extracellular recordings

Multi-electrode recordings in neural tissue contain the action potential waveforms of many closely spaced neurons. While we can observe the action potential waveforms, we cannot observe which neuron is the source for which waveform nor how many source neurons are being recorded. Current spike-sorting algorithms solve this problem by assuming a fixed number of source neurons and assigning the action potentials given this fixed number. We model the spike waveforms as an anisotropic Gaussian mixture model and present, as an alternative, a reversible-jump Markov chain Monte Carlo (MCMC) algorithm to simultaneously estimate the number of source neurons and to assign each action potential to a source. We derive this MCMC algorithm and illustrate its application using simulated three-dimensional data and real four-dimensional feature vectors extracted from tetrode recordings of rat entorhinal cortex neurons. In the analysis of the simulated data our algorithm finds the correct number of mixture components (sources) and classifies the action potential waveforms with minimal error. In the analysis of real data, our algorithm identifies clusters closely resembling those previously identified by a user-dependent graphical clustering procedure. Our findings suggest that a reversible-jump MCMC algorithm could offer a new strategy for designing automated spike-sorting algorithms.

Mechanisms of gamma oscillations in the hippocampus of the behaving rat

Gamma frequency oscillations (30-100 Hz) have been suggested to underlie various cognitive and motor functions. Here, we examine the generation of gamma oscillation currents in the hippocampus, using two-dimensional, 96-site silicon probes. Two gamma generators were identified, one in the dentate gyrus and another in the CA3-CA1 regions. The coupling strength between the two oscillators varied during both theta and nontheta states. Both pyramidal cells and interneurons were phase-locked to gamma waves. Anatomical connectivity, rather than physical distance, determined the coupling strength of the oscillating neurons. CA3 pyramidal neurons discharged CA3 and CA1 interneurons at latencies indicative of monosynaptic connections. Intrahippocampal gamma oscillation emerges in the CA3 recurrent system, which entrains the CA1 region via its interneurons.

Real-time multi-channel stimulus artifact suppression by local curve fitting

We describe an algorithm for suppression of stimulation artifacts in extracellular micro-electrode array (MEA) recordings. A model of the artifact based on locally fitted cubic polynomials is subtracted from the recording, yielding a flat baseline amenable to spike detection by voltage thresholding. The algorithm, SALPA, reduces the period after stimulation during which action potentials cannot be detected by an order of magnitude, to less than 2 ms. Our implementation is fast enough to process 60-channel data sampled at 25 kHz in real-time on an inexpensive desktop PC. It performs well on a wide range of artifact shapes without re-tuning any parameters, because it accounts for amplifier saturation explicitly and uses a statistic to verify successful artifact suppression immediately after the amplifiers become operational. We demonstrate the algorithm's effectiveness on recordings from dense monolayer cultures of cortical neurons obtained from rat embryos. SALPA opens up a previously inaccessible window for studying transient neural oscillations and precisely timed dynamics in short-latency responses to electric stimulation.

The neural basis of the blood-oxygen-level-dependent functional magnetic resonance imaging signal

Magnetic resonance imaging (MRI) has rapidly become an important tool in clinical medicine and biological research. Its functional variant (functional magnetic resonance imaging; fMRI) is currently the most widely used method for brain mapping and studying the neural basis of human cognition. While the method is widespread, there is insufficient knowledge of the physiological basis of the fMRI signal to interpret the data confidently with respect to neural activity. This paper reviews the basic principles of MRI and fMRI, and subsequently discusses in some detail the relationship between the blood-oxygen-level-dependent (BOLD) fMRI signal and the neural activity elicited during sensory stimulation. To examine this relationship, we conducted the first simultaneous intracortical recordings of neural signals and BOLD responses. Depending on the temporal characteristics of the stimulus, a moderate to strong correlation was found between the neural activity measured with microelectrodes and the BOLD signal averaged over a small area around the microelectrode tips. However, the BOLD signal had significantly higher variability than the neural activity, indicating that human fMRI combined with traditional statistical methods underestimates the reliability of the neuronal activity. To understand the relative contribution of several types of neuronal signals to the haemodynamic response, we compared local field potentials (LFPs), single- and multi-unit activity (MUA) with high spatio-temporal fMRI responses recorded simultaneously in monkey visual cortex. At recording sites characterized by transient responses, only the LFP signal was significantly correlated with the haemodynamic response. Furthermore, the LFPs had the largest magnitude signal and linear systems analysis showed that the LFPs were better than the MUAs at predicting the fMRI responses. These findings, together with an analysis of the neural signals, indicate that the BOLD signal primarily measures the input and processing of neuronal information within a region and not the output signal transmitted to other brain regions.

A chronic multi-electrode microdrive for small animals

We describe a simple microdrive device appropriate for chronic microelectrode recording in rats. No precision machining is required; all parts are stock or cut from standard stock material with hand tools and assembled with epoxy. The device together with its electrodes can be discarded at the completion of the experiment.

Tetrode technology: advances in implantable hardware, neuroimaging, and data analysis techniques

The technical advances in hardware and software for multiunit recordings have made it easier to gather data from a large number of neurons for behavioral correlations. This paper discusses several such advances in implantable hardware, magnetic resonance imaging of electrodes in situ, and data analysis software for multiple simultaneous signals.

Electrophysiological profile of avian hippocampal unit activity: a basis for regional subdivisions

Electrophysiological activity was recorded from single neurons (units) in the hippocampal formation (HF) of freely moving homing pigeons in order to provide a taxonomy of unit types found in the avian HF; a taxonomy that could be used to define regional subdivisions and be compared with unit types found in the mammalian hippocampus. Two distinct types of unit were observed in the avian HF. One type was uniformly characterized by relatively rapid firing rates and shorter spike widths, and was found throughout the HF. The other type was more variable in activity profile but, compared with the fast-firing units, was characterized by slower firing rates and longer spike widths. However, despite the variable nature of the slow-firing units, most slow-firing units recorded within a given anatomical region displayed similar firing rates, spike widths, and interspike intervals. In general, ventral HF units displayed activity patterns similar to projection cells found in the mammalian Ammon's horn. Most dorsocaudal units displayed activity patterns similar to presumed granular cells in the mammalian dentate gyrus. By contrast, most dorsorostral units displayed activity patterns similar to a type of unit found in the mammalian subiculum. Although different in some details, the overall activity profile of units found in the avian HF, and their regional distribution, is strikingly similar to unit types found in the mammalian hippocampus, suggesting that unit activity profile is one hippocampal dimension conserved through evolution.

Comparison between local field potentials and unit cluster activity in primary auditory cortex and anterior auditory field in the cat

Multi-unit (MU) activity and local field potentials (LFP) were simultaneously recorded from 161 sites in the middle cortical layers of the primary auditory cortex (AI) and the anterior auditory field (AAF) in 51 cats. Responses were obtained for frequencies between 625 Hz and 40 kHz, at intensities from 75 dB SPL to threshold. We compared the response properties of MU activity and LFP triggers, in terms of characteristic frequency (CF), threshold at CF, minimum latency and frequency tuning-curve bandwidth 20 dB above threshold. On average, thresholds at CF were significantly lower for LFP events than those for MU spikes (4.6 dB for AI, and 3 dB for AAF). Minimum latencies were significantly shorter for LFP events than for MU spikes (1.5 ms in AI, and 1.7 ms in AAF). Frequency tuning curves were significantly broader for LFP events than those for MU spikes (1.0 octave in AI, and 1.3 octaves in AAF). In contrast, the CF was not significantly different between LFP events and MU spikes. The LFP results indicate that cortical neurons receive convergent sub-cortical inputs from a broad frequency range. The sharper tuning curves for MU activity compared to those of LFP events are likely the result of intracortical inhibitory processes.

Artifactual synchrony via capacitance coupling in multi-electrode recording from cat striate cortex

Elucidation of neural connectivity patterns in the brain are thought to give us a mechanistic understanding of how the brain works. Functional connectivity is best studied by simultaneous recording of single-unit activity from many neurons. Accordingly, various types of multiple-microelectrode systems have been developed. We have studied long-range lateral interactions in cat striate cortex. To physiologically characterize interacting cells recorded simultaneously, we used two microelectrodes whose movements were controlled by two independently-movable microdrives. The tips of the two microelectrodes were separated by approximately 2 mm or more. During preliminary plotting of two receptive fields of cell pairs, we often noted the emergence of perfectly synchronous firing between two spike trains (amplitude ratio, about 20:1) registered with two microelectrodes. Synchronously firing, smaller spikes disappeared when larger spikes of the pair were lost to either substantial advancement of or placing an electrolytic lesion at the electrode registering the latter. The synchrony also disappeared when two microdrive systems were shielded individually. We concluded that the synchrony was attained through capacitance coupling between two microdrive systems. We proposed a few practical recommendations to avoid the contamination of cross correlograms with the false-positive, narrow peak at time zero due to the presence of reflected spike trains.

Signal-to-noise ratio improvement in multiple electrode recording

Recordings of spike trains made with microwires or silicon electrodes include more noise from various sources that contaminate the observed spike shapes compared with recordings using sharp microelectrodes. This is a particularly serious problem if spike shape sorting is required to separate the several trains that might be observed on a particular electrode. However, if recordings are made with an array of such electrodes, there are several mathematical methods to improve the effective signal (spikes) to noise ratio, thus considerably reducing inaccuracy in spike detection and shape sorting. We compare the theoretical basis of three such methods and evaluate their performance with simulated and real data.

The application of elastomeric connector for multi-channel electrophysiological recordings

Interest in recording multi-channel electrophysiological data from behaving animals is rapidly growing, and many laboratories tend to record a large number of EEG and/or multi-unit channels, despite the limitation of the size of the headpiece that a small behaving animal can carry. A common drawback of these experiments, therefore, is the relatively large size of even the smallest, commercially available, high-density micro-connectors for the headset. To overcome this problem, we suggest the application of elastomeric or silicone inter-rubber connectors, that are widely used in electronics. The elastomeric or "zebra" connector consists of alternating thin strips of layered electrically conductive and non-conductive materials. The conductive strips provide electrical connections between uninsulated contact surfaces of printed circuit boards such as the connector plate of the micro-drive, that holds the brain electrode wires, and the preamplifier board of the recording system. In the present paper, we provide technical details of the design of this type of connector-sets and discuss common issues arising from their use. By comparing the applicability of two designs, we aim to demonstrate the simplicity, reliability and durability of the elastomeric inter-rubber connectors in electrophysiological experiments on freely moving laboratory animals.

Redundancy reduction and sustained firing with stochastic depressing synapses

Many synapses in the CNS transmit only a fraction of the action potentials that reach them. Although unreliable, such synapses do not transmit completely randomly, because the probability of transmission depends on the recent history of synaptic activity. We examine how a variety of spike trains, including examples recorded from area V1 of monkeys freely viewing natural scenes, are transmitted through a realistic model synapse with activity-dependent depression arising from vesicle depletion or postrelease refractoriness. The resulting sequences of transmitted spikes are significantly less correlated, and hence less redundant, than the presynaptic spike trains that generate them. The spike trains we analyze, which are typical of those recorded in a variety of brain regions, have positive autocorrelations because of the occurrence of variable length periods of sustained firing at approximately constant rates. Sustained firing may, at first, seem inconsistent with input from depressing synapses. We show, however, that such a pattern of activity can arise if the postsynaptic neuron is driven by a fixed population of direct, "feedforward" inputs accompanied by a variable number of delayed, "reverberatory" inputs. This leads to a prediction for the number and latency distribution of the inputs that typically drive a cortical neuron.

Variability and information in a neural code of the cat lateral geniculate nucleus

A central theme in neural coding concerns the role of response variability and noise in determining the information transmission of neurons. This issue was investigated in single cells of the lateral geniculate nucleus of barbiturate-anesthetized cats by quantifying the degree of precision in and the information transmission properties of individual spike train responses to full field, binary (bright or dark), flashing stimuli. We found that neuronal responses could be highly reproducible in their spike timing (approximately 1-2 ms standard deviation) and spike count (approximately 0.3 ratio of variance/mean, compared with 1.0 expected for a Poisson process). This degree of precision only became apparent when an adequate length of the stimulus sequence was specified to determine the neural response, emphasizing that the variables relevant to a cell's response must be controlled to observe the cell's intrinsic response precision. Responses could carry as much as 3.5 bits/spike of information about the stimulus, a rate that was within a factor of two of the limit the spike train could transmit. Moreover, there appeared to be little sign of redundancy in coding: on average, longer response sequences carried at least as much information about the stimulus as would be obtained by adding together the information carried by shorter response sequences considered independently. There also was no direct evidence found for synergy between response sequences. These results could largely, but not entirely, be explained by a simple model of the response in which one filters the stimulus by the cell's impulse response kernel, thresholds the result at a fairly high level, and incorporates a postspike refractory period.

The application of cross-point switch arrays as input selector switch devices for multi-channel electrophysiological experiments

Integrated circuits (ICs) containing cross-point switch arrays were applied to create analog input selector switch devices for multi-channel electrophysiological experiments. The described analog input selector switch devices make it possible to connect to the main amplifier's inputs those microelectrode and preamplifier output wires that yield unit discharges of acceptable shape and amplitude, or yield other kind of acceptable electrophysiological signals (EEG, EP). This kind of selector allows to use higher number of preamplifier channels and to ignore the input channels without adequate signals or the channels with noisy inputs. No manual switching is required, as the work is done by computer controlled switches. The switch positions can be saved and reloaded at the next experimental session through an I/O port (e.g. the parallel port) of the computer.

High-resolution spatio-temporal mapping of visual pathways using multi-electrode arrays

The parallel processing of visual information was studied with penetrating microelectrode arrays. We studied the high-resolution visuotopic organization of cat primary visual cortex, and the encoding of simple visual stimuli by ensembles of ganglion cells in the isolated turtle retina. The high-resolution visuotopic organization of visual cortex is non-conformal. Regions of visual cortex separated by 400 mu may have receptive field centers that are separated by as much as 3 degrees, or they may superimpose. Ganglion cells are 'generalists', and are poor specifiers of the color of full field visual stimuli. Groups of 'luminosity' type ganglion cells can assist in the specification of stimulus color, but even individual 'chromatic' ganglion cells are not capable of quality color specification. These basic studies have relevance to the development of visual neuroprostheses based upon electrical stimulation of the retina and cortex.

Daily variation and appetitive conditioning-induced plasticity of auditory cortex receptive fields

Long-term modification of cortical receptive field maps follows learning of sensory discriminations and conditioned associations. In the process of determining whether appetitive - as opposed to aversive - conditioning is effective in causing such plastic changes, it was discovered that multineuron receptive fields, when measured in rats under ketamine-sedation, vary substantially over the course of a week, even in the absence of classical conditioning and electrode movement. Specifically, a simple correlation analysis showed that iso-intensity frequency response curves of multiunit clusters and local field potentials recorded from auditory cortex are nonstationary over 7 days. Nevertheless, significant plastic changes in receptive fields, due to conditioned pairing of a pure tone and electrical stimulation of brain reward centres, are detectable above and beyond these spontaneous daily variations. This finding is based on a novel statistical plasticity criterion which compares receptive fields recorded for three days before and three days after conditioning. Based on a more traditional criterion (i.e. one day before and after conditioning), the prevalence of learning-induced changes caused by appetitive conditioning appears to be comparable to that described in previous studies involving aversive conditioning.

Haloperidol and clozapine increase neural activity in the rat prefrontal cortex

Haloperidol and clozapine have been widely used to alleviate schizophrenic symptoms, but their physiological effects in the prefrontal cortex (PFC) are not known. Effects of haloperidol and clozapine on single unit activity were investigated in the medial PFC of anesthetized rats. Injection (intraperitoneal) of haloperidol (1 mg/kg) or clozapine (20 mg/kg) significantly elevated discharge rates of PFC neurons. Considering that hypofrontality is one characteristic of schizophrenic symptoms, these results raise the possibility that enhancement of PFC neural activity contributes to therapeutic effects of haloperidol and clozapine.

Experience-dependent changes in extracellular spike amplitude may reflect regulation of dendritic action potential back-propagation in rat hippocampal pyramidal cells

Activity-dependent attenuations in extracellular spike amplitude have been shown to correlate with a decrease in the effectiveness with which somatic action potentials back-propagate into the dendritic arbor of hippocampal pyramidal cells. In this paper we demonstrate that activity-dependent attenuations in amplitude occur during behavior and that the amount of attenuation is reduced with an animal's experience in an environment. The observed reductions are caused by an animal's experience within a specific environmental context, are dependent on functional NMDA receptors, and are accompanied by an increase in the effective coupling of pyramidal cells and interneurons. These results provide an important step in linking together in vivo studies with in vitro data and suggest that mechanisms of plasticity engaged during behavior may be sufficient to alter the biophysical and integrative properties of hippocampal pyramidal cells.

The application of printed circuit board technology for fabrication of multi-channel micro-drives

A modular multichannel microdrive ('hyperdrive') is described. The microdrive uses printed circuit board technology and flexible fused silica capillaries. The modular design allows for the fabrication of 4-32 independently movable electrodes or 'tetrodes'. The drives are re-usable and re-loading the drive with electrodes is simple.

How the brain uses time to represent and process visual information(1)

Information theory provides a theoretical framework for addressing fundamental questions concerning the nature of neural codes. Harnessing its power is not straightforward, because of the differences between mathematical abstractions and laboratory reality. We describe an approach to the analysis of neural codes that seeks to identify the informative features of neural responses, rather than to estimate the information content of neural responses per se. Our analysis, applied to neurons in primary visual cortex (V1), demonstrates that the informative precision of spike times varies with the stimulus modality being represented. Contrast is represented by spike times on the shortest time scale, and different kinds of pattern information are represented on longer time scales. The interspike interval distribution has a structure that is unanticipated from the firing rate. The significance of this structure is not that it contains additional information, but rather, that it may provide a means for simple synaptic mechanisms to decode the information that is multiplexed within a spike train. Extensions of this analysis to the simultaneous responses of pairs of neurons indicate that neighboring neurons convey largely independent information, if the decoding process is sensitive to the neuron of origin and not just the average firing rate. In summary, stimulus-related information is encoded into the precise times of spikes fired by V1 neurons. Much of this information would be obscured if individual spikes were merely taken to be estimators of the firing rate. Additional information would be lost by averaging across the responses of neurons in a local population. We propose that synaptic mechanisms sensitive to interspike intervals and dendritic processing beyond simple summation exist at least in part to enable the brain to take advantage of this extra information.

Clustering of perirhinal neurons with similar properties following visual experience in adult monkeys

The functional organization of early visual areas seems to be largely determined during development. However, the organization of areas important for learning and memory, such as perirhinal cortex, may be modifiable in adults. To test this hypothesis, we recorded from pairs of neurons in perirhinal cortex of macaques while they viewed multiple complex stimuli. For novel stimuli, neuronal response preferences for pairs of nearby neurons and far-apart neurons were uncorrelated. However, after one day of experience with the stimuli, response preferences of nearby neurons became more similar. We conclude that specific visual experience induces development of clusters of perirhinal neurons with similar stimulus preferences.