Cortical auditory neuron interactions during presentation of 3-tone sequences: effective connectivity

Training and Spontaneous Reinforcement of Neuronal Assemblies by Spike Timing Plasticity

The synaptic connectivity of cortex is plastic, with experience shaping the ongoing interactions between neurons. Theoretical studies of spike timing-dependent plasticity (STDP) have focused on either just pairs of neurons or large-scale simulations. A simple analytic account for how fast spike time correlations affect both microscopic and macroscopic network structure is lacking. We develop a low-dimensional mean field theory for STDP in recurrent networks and show the emergence of assemblies of strongly coupled neurons with shared stimulus preferences. After training, this connectivity is actively reinforced by spike train correlations during the spontaneous dynamics. Furthermore, the stimulus coding by cell assemblies is actively maintained by these internally generated spiking correlations, suggesting a new role for noise correlations in neural coding. Assembly formation has often been associated with firing rate-based plasticity schemes; our theory provides an alternative and complementary framework, where fine temporal correlations and STDP form and actively maintain learned structure in cortical networks.

Attention Configures Synchronization Within Local Neuronal Networks for Processing of the Behaviorally Relevant Stimulus

The need for fast and dynamic processing of relevant information imposes high demands onto the flexibility and efficiency of the nervous system. A good example for such flexibility is the attention-dependent selection of relevant sensory information. Studies investigating attentional modulations of neuronal responses to simultaneously arriving input showed that neurons respond, as if only the attended stimulus would be present within their receptive fields (RF). However, attention also improves neuronal representation and behavioral performance, when only one stimulus is present. Thus, attention serves for selecting relevant input and changes the neuronal processing of signals representing selected stimuli, ultimately leading to a more efficient behavioral performance. Here, we tested the hypothesis that attention configures the strength of functional coupling between a local neuronal network's neurons specifically for effective processing of signals representing attended stimuli. This coupling is measured as the strength of γ-synchronization between these neurons. The hypothesis predicts that the pattern of synchronization in local networks should depend on which stimulus is attended. Furthermore, we expect this pattern to be similar for the attended stimulus presented alone or together with irrelevant stimuli in the RF. To test these predictions, we recorded spiking-activity and local field potentials (LFP) with closely spaced electrodes in area V4 of monkeys performing a demanding attention task. Our results show that the γ-band phase coherence (γ-PhC) between spiking-activity and the LFP, as well as the spiking-activity of two groups of neurons, strongly depended on which of the two stimuli in the RF was attended. The γ-PhC was almost identical for the attended stimulus presented either alone or together with a distractor. The functional relevance of dynamic γ-band synchronization is further supported by the observation of strongly degraded γ-PhC before behavioral errors, while firing rates were barely affected. These qualitatively different results point toward a failure of attention-dependent top-down mechanisms to correctly synchronize the local neuronal network in V4, even though this network receives the correctly selected input. These findings support the idea of a flexible, demand-dependent dynamic configuration of local neuronal networks, for performing different functions, even on the same sensory input.

Encoding by synchronization in the primate striatum

Information is encoded in the nervous system through the discharge and synchronization of single neurons. The striatum, the input stage of the basal ganglia, is divided into three territories: the putamen, the caudate, and the ventral striatum, all of which converge onto the same motor pathway. This parallel organization suggests that there are multiple and competing systems in the basal ganglia network controlling behavior. To explore which mechanism(s) enables the different striatal domains to encode behavioral events and to control behavior, we compared the neural activity of phasically active neurons [medium spiny neurons (MSNs), presumed projection neurons] and tonically active neurons (presumed cholinergic interneurons) across striatal territories from monkeys during the performance of a well practiced task. Although neurons in all striatal territories displayed similar spontaneous discharge properties and similar temporal modulations of their discharge rates to the behavioral events, their correlation structure was profoundly different. The distributions of signal and noise correlation of pairs of putamen MSNs were strongly shifted toward positive correlations and these two measures were correlated. In contrast, MSN pairs in the caudate and ventral striatum displayed symmetrical, near-zero signal and noise correlation distributions. Furthermore, only putamen MSN pairs displayed different noise correlation dynamics to rewarding versus neutral/aversive cues. Similarly, the noise correlation between tonically active neuron pairs was stronger in the putamen than in the caudate. We suggest that the level of synchronization of the neuronal activity and its temporal dynamics differentiate the striatal territories and may thus account for the different roles that striatal domains play in behavioral control.

Functional circuits and anatomical distribution of response properties in the primate amygdala

Recent electrophysiological studies on the primate amygdala have advanced our understanding of how individual neurons encode information relevant to emotional processes, but it remains unclear how these neurons are functionally and anatomically organized. To address this, we analyzed cross-correlograms of amygdala spike trains recorded during a task in which monkeys learned to associate novel images with rewarding and aversive outcomes. Using this task, we have recently described two populations of amygdala neurons: one that responds more strongly to images predicting reward (positive value-coding), and another that responds more strongly to images predicting an aversive stimulus (negative value-coding). Here, we report that these neural populations are organized into distinct, but anatomically intermingled, appetitive and aversive functional circuits, which are dynamically modulated as animals used the images to predict outcomes. Furthermore, we report that responses to sensory stimuli are prevalent in the lateral amygdala, and are also prevalent in the medial amygdala for sensory stimuli that are emotionally significant. The circuits identified here could potentially mediate valence-specific emotional behaviors thought to involve the amygdala.

Beyond traditional approaches to understanding the functional role of neuromodulators in sensory cortices

Over the last two decades, a vast literature has described the influence of neuromodulatory systems on the responses of sensory cortex neurons (review in Gu, 2002; Edeline, 2003; Weinberger, 2003; Metherate, 2004, 2011). At the single cell level, facilitation of evoked responses, increases in signal-to-noise ratio, and improved functional properties of sensory cortex neurons have been reported in the visual, auditory, and somatosensory modality. At the map level, massive cortical reorganizations have been described when repeated activation of a neuromodulatory system are associated with a particular sensory stimulus. In reviewing our knowledge concerning the way the noradrenergic and cholinergic system control sensory cortices, I will point out that the differences between the protocols used to reveal these effects most likely reflect different assumptions concerning the role of the neuromodulators. More importantly, a gap still exists between the descriptions of neuromodulatory effects and the concepts that are currently applied to decipher the neural code operating in sensory cortices. Key examples that bring this gap into focus are the concept of cell assemblies and the role played by the spike timing precision (i.e., by the temporal organization of spike trains at the millisecond time-scale) which are now recognized as essential in sensory physiology but are rarely considered in experiments describing the role of neuromodulators in sensory cortices. Thus, I will suggest that several lines of research, particularly in the field of computational neurosciences, should help us to go beyond traditional approaches and, ultimately, to understand how neuromodulators impact on the cortical mechanisms underlying our perceptual abilities.

Stimulus-induced reduction of noise correlation in rat prefrontal cortex

We have shown previously that stimulus-induced modulation of noise correlation in rat somatosensory cortex conveys additional information about the delivery of tactile stimulation. Here we investigated whether noise correlation is also modulated by an external sensory stimulus in rat prefrontal cortex and, if so, whether such modulation conveys additional information on stimulus delivery. Noise correlation was significantly reduced after the onset of a conditional stimulus (auditory tone) that signaled an electric foot shock in the prefrontal cortex. However, noise correlation contributed little to the transmission of information on stimulus delivery. These results indicate that a meaningful sensory stimulus reduces noise correlation in rat prefrontal cortex, but such modulation does not play a significant role in conveying information on stimulus delivery.

How different are the local field potentials and spiking activities? Insights from multi-electrodes arrays

Simultaneous recording of multiple neurons, or neuron groups, offers new promise for investigating fundamental questions about the neural code. We used arrays of 16 electrodes in the tonotopic, primary, auditory cortex of guinea pigs and we extracted LFP- and spike-based spectro-temporal receptive fields (STRFs). We confirm here that LFP signals provide broadly tuned activity which lacks frequency resolution compared to multiunit signals and, therefore, lead to large redundancy in neural responses even between recording sites far apart. Thanks to the use of multi-electrode arrays which allows simultaneous recordings, we also focused on functional relationships between neuronal discharges (through cross-correlations) and between LFPs (through coherence). Since the LFP is composed of distinct brain rhythms, the LFP results were split into three frequency bands from the slowest to the fastest components of LFPs. For driven as well as spontaneous activity, we show that components >70 Hz in LFPs are much less coherent between recording sites than slower components. In general, coherence between LFPs from two recordings sites is positively correlated with the degree of frequency overlap between the two corresponding STRFs, similar to cross-correlation between multiunit activities. However, coherence is only weakly correlated with cross-correlation in all frequency ranges. Altogether, these results suggest that LFPs reflect global functional connectivity in the thalamocortical auditory system whereas spiking activities reflect more independent local processing.

Using neuronal populations to study the mechanisms underlying spatial and feature attention

Visual attention affects both perception and neuronal responses. Whether the same neuronal mechanisms mediate spatial attention, which improves perception of attended locations, and nonspatial forms of attention has been a subject of considerable debate. Spatial and feature attention have similar effects on individual neurons. Because visual cortex is retinotopically organized, however, spatial attention can comodulate local neuronal populations, whereas feature attention generally requires more selective modulation. We compared the effects of feature and spatial attention on local and spatially separated populations by recording simultaneously from dozens of neurons in both hemispheres of V4. Feature and spatial attention affect the activity of local populations similarly, modulating both firing rates and correlations between pairs of nearby neurons. However, whereas spatial attention appears to act on local populations, feature attention is coordinated across hemispheres. Our results are consistent with a unified attentional mechanism that can modulate the responses of arbitrary subgroups of neurons.

Measuring and interpreting neuronal correlations

Mounting evidence suggests that understanding how the brain encodes information and performs computations will require studying the correlations between neurons. The recent advent of recording techniques such as multielectrode arrays and two-photon imaging has made it easier to measure correlations, opening the door for detailed exploration of their properties and contributions to cortical processing. However, studies have reported discrepant findings, providing a confusing picture. Here we briefly review these studies and conduct simulations to explore the influence of several experimental and physiological factors on correlation measurements. Differences in response strength, the time window over which spikes are counted, spike sorting conventions and internal states can all markedly affect measured correlations and systematically bias estimates. Given these complicating factors, we offer guidelines for interpreting correlation data and a discussion of how best to evaluate the effect of correlations on cortical processing.

Neural codes in the thalamocortical auditory system: from artificial stimuli to communication sounds

Over the last 15 years, an increasing number of studies have described the responsiveness of thalamic and cortical neurons to communication sounds. Whereas initial studies have simply looked for neurons exhibiting higher firing rate to conspecific vocalizations over their modified, artificially synthesized versions, more recent studies determine the relative contribution of "rate coding" and "temporal coding" to the information transmitted by spike trains. In this article, we aim at reviewing the different strategies employed by thalamic and cortical neurons to encode information about acoustic stimuli, from artificial to natural sounds. Considering data obtained with simple stimuli, we first illustrate that different facets of temporal code, ranging from a strict correspondence between spike-timing and stimulus temporal features to more complex coding strategies, do already exist with artificial stimuli. We then review lines of evidence indicating that spike-timing provides an efficient code for discriminating communication sounds from thalamus, primary and non-primary auditory cortex up to frontal areas. As the neural code probably developed, and became specialized, over evolution to allow precise and reliable processing of sounds that are of survival value, we argue that spike-timing based coding strategies might set the foundations of our perceptive abilities.

Maximum decoding abilities of temporal patterns and synchronized firings: application to auditory neurons responding to click trains and amplitude modulated white noise

Simultaneous recordings of an increasing number of neurons have recently become available, but few methods have been proposed to handle this activity. Here, we extract and investigate all the possible temporal neural activity patterns based on synchronized firings of neurons recorded on multiple electrodes, or based on bursts of single-electrode activity in cat primary auditory cortex. We apply this to responses to periodic click trains or sinusoïdal amplitude modulated noise by obtaining for each pattern its temporal modulation transfer function. An algorithm that maximizes the mutual information between all patterns and stimuli subsequently leads to the identification of patterns that optimally decode modulation frequency (MF). We show that stimulus information contained in multi-electrode synchronized firing is not redundant with single-electrode firings and leads to improved efficiency of MF decoding. We also show that the combined use of firing rate and temporal codes leads to a better discrimination of the MF.

Organization of interneuronal connections in the nucleus accumbens in "impulsive" and "self-controlled" behavior in cats

In behavioral experiments, cats placed in a situation of choosing between a high-value time-delayed and a low-value rapid food reinforcement elected to wait for the preferred reward (they demonstrated "self-control") or to obtain the worse reward quickly (they demonstrated impulsive behavior). On the basis of the selected behavioral strategy, the cats were divided into three groups - "impulsive," "ambivalent," and "self-controlled." Cross-correlation analysis was used to assess the linked activity of cells in the nucleus accumbens, which reflects the nature of interactions between close-lying neurons. In cats with self-control, interneuronal interactions appeared in a significantly larger proportion of cases than in impulsive cats. In combinations resulting in long-latency reactions, cats with self-controlled and impulsive behavior showed no significant difference in the occurrence frequency of interneuronal interactions. The numbers of interneuronal interactions were greater during erroneous responses as compared with correctly performed reactions in animals of the different groups. These data indicate a key role for the interrelated activity of nucleus accumbens neurons in organizing the pattern of long-latency responses typical of self-controlled behavior.

Bayesian inference of functional connectivity and network structure from spikes

Current multielectrode techniques enable the simultaneous recording of spikes from hundreds of neurons. To study neural plasticity and network structure it is desirable to infer the underlying functional connectivity between the recorded neurons. Functional connectivity is defined by a large number of parameters, which characterize how each neuron influences the other neurons. A Bayesian approach that combines information from the recorded spikes (likelihood) with prior beliefs about functional connectivity (prior) can improve inference of these parameters and reduce overfitting. Recent studies have used likelihood functions based on the statistics of point-processes and a prior that captures the sparseness of neural connections. Here we include a prior that captures the empirical finding that interactions tend to vary smoothly in time. We show that this method can successfully infer connectivity patterns in simulated data and apply the algorithm to spike data recorded from primary motor (M1) and premotor (PMd) cortices of a monkey. Finally, we present a new approach to studying structure in inferred connections based on a Bayesian clustering algorithm. Groups of neurons in M1 and PMd show common patterns of input and output that may correspond to functional assemblies.

Inferring functional connections between neurons

A central question in neuroscience is how interactions between neurons give rise to behavior. In many electrophysiological experiments, the activity of a set of neurons is recorded while sensory stimuli or movement tasks are varied. Tools that aim to reveal underlying interactions between neurons from such data can be extremely useful. Traditionally, neuroscientists have studied these interactions using purely descriptive statistics (cross-correlograms or joint peri-stimulus time histograms). However, the interpretation of such data is often difficult, particularly as the number of recorded neurons grows. Recent research suggests that model-based, maximum likelihood methods can improve these analyses. In addition to estimating neural interactions, application of these techniques has improved decoding of external variables, created novel interpretations of existing electrophysiological data, and may provide new insight into how the brain represents information.

Context-dependent changes in functional circuitry in visual area MT

Animals can flexibly change their behavior in response to a particular sensory stimulus; the mapping between sensory and motor representations in the brain must therefore be flexible as well. Changes in the correlated firing of pairs of neurons may provide a metric of changes in functional circuitry during behavior. We studied dynamic changes in functional circuitry by analyzing the noise correlations of simultaneously recorded MT neurons in two behavioral contexts: one that promotes cooperative interactions between the two neurons and another that promotes competitive interactions. We found that identical visual stimuli give rise to differences in noise correlation in the two contexts, suggesting that MT neurons receive inputs of central origin whose strength changes with the task structure. The data are consistent with a mixed feature-based attentional strategy model in which the animal sometimes alternates attention between opposite directions of motion and sometimes attends to the two directions simultaneously.

Early stages of melody processing: stimulus-sequence and task-dependent neuronal activity in monkey auditory cortical fields A1 and R

To explore the effects of acoustic and behavioral context on neuronal responses in the core of auditory cortex (fields A1 and R), two monkeys were trained on a go/no-go discrimination task in which they learned to respond selectively to a four-note target (S+) melody and withhold response to a variety of other nontarget (S-) sounds. We analyzed evoked activity from 683 units in A1/R of the trained monkeys during task performance and from 125 units in A1/R of two naive monkeys. We characterized two broad classes of neural activity that were modulated by task performance. Class I consisted of tone-sequence-sensitive enhancement and suppression responses. Enhanced or suppressed responses to specific tonal components of the S+ melody were frequently observed in trained monkeys, but enhanced responses were rarely seen in naive monkeys. Both facilitatory and suppressive responses in the trained monkeys showed a temporal pattern different from that observed in naive monkeys. Class II consisted of nonacoustic activity, characterized by a task-related component that correlated with bar release, the behavioral response leading to reward. We observed a significantly higher percentage of both Class I and Class II neurons in field R than in A1. Class I responses may help encode a long-term representation of the behaviorally salient target melody. Class II activity may reflect a variety of nonacoustic influences, such as attention, reward expectancy, somatosensory inputs, and/or motor set and may help link auditory perception and behavioral response. Both types of neuronal activity are likely to contribute to the performance of the auditory task.

Correlated neural activity as the driving force for functional changes in auditory cortex

The functional role of neural synchrony is reflected in cortical tonotopic map reorganization and in the emergence of pathological phenomena such as tinnitus. First of all experimenter-centered and subject-centered views of neural activity will be contrasted; this argues against the use of stimulus-correction procedures and favors the use of a correction procedure based on neural activity without reference to stimulus timing. Within a cortical column neurons fired synchronously with on average about 6% of their spikes in a 1 ms bin and occasionally showing 30% or more of such coincident spikes. For electrode separations exceeding 200 microm the average peak correlation strength only occasionally reached 3%. The experimental evidence for coincidence of neural activity, neural correlation and neural synchrony shows that horizontal fibers activity can induce strong neural correlations. Cortico-cortical connections for a large part connect cell groups with characteristic frequencies differing by more than one octave. Such neurons have generally non-overlapping receptive fields but still can have sizeable peak cross-correlations. Correlated neural activity and heterotopic neural interconnections are presented as the substrates for cortical reorganization; increased neural synchrony and tonotopic map reorganization go hand in hand. This links cortical reorganization with hypersynchrony that can be considered as an important driving force underlying tinnitus.

Neurophysiology and neuroanatomy of pitch perception: auditory cortex

We present original results and review literature from the past fifty years that address the role of primate auditory cortex in the following perceptual capacities: (1) the ability to perceive small differences between the pitches of two successive tones; (2) the ability to perceive the sign (i.e., direction) of the pitch difference [higher (+) vs. lower (-)]; and (3) the ability to abstract pitch constancy across changes in stimulus acoustics. Cortical mechanisms mediating pitch perception are discussed with respect to (1) gross and microanatomical distribution; and (2) candidate neural coding schemes. Observations by us and others suggest that (1) frequency-selective neurons in primary auditory cortex (A1) and surrounding fields play a critical role in fine-grained pitch discrimination at the perceptual level; (2) cortical mechanisms that detect pitch differences are neuroanatomically dissociable from those mediating pitch direction discrimination; (3) cortical mechanisms mediating perception of the "missing fundamental frequency (F0)" are neuroanatomically dissociable from those mediating pitch perception when F0 is present; (4) frequency-selective neurons in both right and left A1 contribute to pitch change detection and pitch direction discrimination; (5) frequency-selective neurons in right A1 are necessary for normal pitch direction discrimination; (6) simple codes for pitch that are based on single- and multiunit firing rates of frequency-selective neurons face both a "hyperacuity problem" and a "pitch constancy problem"-that is, frequency discrimination thresholds for pitch change direction and pitch direction discrimination are much smaller than neural tuning curves predict, and firing rate patterns change dramatically under conditions in which pitch percepts remain invariant; (7) cochleotopic organization of frequency-selective neurons bears little if any relevance to perceptual acuity and pitch constancy; and (8) simple temporal codes for pitch capable of accounting for pitches higher than a few hundred hertz have not been found in the auditory cortex. The cortical code for pitch is therefore not likely to be a function of simple rate profiles or synchronous temporal patterns. Studies motivated by interest in the neurophysiology and neuroanatomy of music perception have helped correct longstanding misconceptions about the functional role of auditory cortex in frequency discrimination and pitch perception. Advancing knowledge about the neural coding of pitch is of fundamental importance to the future design of neurobionic therapies for hearing loss.

Properties of Correlated Neural Activity Clusters in Cat Auditory Cortex Resemble Those of Neural Assemblies

Spiking activity was recorded from cat auditory cortex using multi-electrode arrays. Cross-correlograms were calculated for spikes recorded on separate microelectrodes. The pair-wise cross-correlation matrix was constructed for the peak values of the correlograms. Hierarchical clustering was performed on the cross-correlation matrix for six stimulus conditions. These were silence, three multi-tone stimulus ensembles with different spectral densities, low-pass amplitude-modulated noise, and Poisson-distributed click trains that each lasted 15 min. The resulting neuron clusters reflect patches in cortex of up to several mm2 in size that expand and contract in response to different stimuli. Cluster positions and size were very similar for spontaneous activity and multi-tone stimulus-evoked activity but differed between those conditions and the noise and click stimuli. Cluster size was significantly larger in posterior auditory field (PAF) compared with primary auditory cortex (AI), whereas the fraction of common spikes (within a 10-ms window) across all electrode activity participating in a cluster was significantly higher in AI compared with PAF. Clusters crossed area boundaries in <5% of the cases were simultaneous recording were made in AI and PAF. Clusters are therefore similar to but not synonymous with the traditional view of neural assemblies. Common-spike spectrotemporal receptive fields (STRFs) were obtained for common-spike activity and all-spike activity within a cluster. Common-spike STRFs had higher signal-to-noise ratio than all-spike STRFs and showed generally spectral and temporal sharpening. The coincident and noncoincident output of the clusters could potentially act in parallel and may serve different modes of stimulus coding.

Analyzing functional connectivity using a network likelihood model of ensemble neural spiking activity

Analyzing the dependencies between spike trains is an important step in understanding how neurons work in concert to represent biological signals. Usually this is done for pairs of neurons at a time using correlation-based techniques. Chornoboy, Schramm, and Karr (1988) proposed maximum likelihood methods for the simultaneous analysis of multiple pair-wise interactions among an ensemble of neurons. One of these methods is an iterative, continuous-time estimation algorithm for a network likelihood model formulated in terms of multiplicative conditional intensity functions. We devised a discrete-time version of this algorithm that includes a new, efficient computational strategy, a principled method to compute starting values, and a principled stopping criterion. In an analysis of simulated neural spike trains from ensembles of interacting neurons, the algorithm recovered the correct connectivity matrices and interaction parameters. In the analysis of spike trains from an ensemble of rat hippocampal place cells, the algorithm identified a connectivity matrix and interaction parameters consistent with the pattern of conjoined firing predicted by the overlap of the neurons' spatial receptive fields. These results suggest that the network likelihood model can be an efficient tool for the analysis of ensemble spiking activity.

On the binding of successive sounds: perceiving shifts in nonperceived pitches

It is difficult to hear out individually the components of a "chord" of equal-amplitude pure tones with synchronous onsets and offsets. In the present study, this was confirmed using 300-ms random (inharmonic) chords with components at least 1/2 octave apart. Following each chord, after a variable silent delay, listeners were presented with a single pure tone which was either identical to one component of the chord or halfway in frequency between two components. These two types of sequence could not be reliably discriminated from each other. However, it was also found that if the single tone following the chord was instead slightly (e.g., 1/12 octave) lower or higher in frequency than one of its components, the same listeners were sensitive to this relation. They could perceive a pitch shift in the corresponding direction. Thus, it is possible to perceive a shift in a nonperceived frequency/pitch. This paradoxical phenomenon provides psychophysical evidence for the existence of automatic "frequency-shift detectors" in the human auditory system. The data reported here suggest that such detectors operate at an early stage of auditory scene analysis but can be activated by a pair of sounds separated by a few seconds.

Cross-correlation and joint spectro-temporal receptive field properties in auditory cortex

Recordings were made from the right primary auditory cortex in 17 adult cats using two eight-electrode arrays. We recorded the neural activity under spontaneous firing conditions and during random, multi-frequency stimulation, at 65 dB SPL, from the same units. Multiple single-unit (MSU) recordings (281) were stationary through 900 s of silence and during 900 s of stimulation. The cross-correlograms of 545 MSU pairs with peak lag times within 10 ms from zero lag time were analyzed. Stimulation reduced the correlation in background activity, and as a result, the signal-to-noise ratio of correlated activity in response to the stimulus was enhanced. Reconstructed spectro-temporal receptive fields (STRFs) for coincident spikes showed larger STRF overlaps, suggesting that coincident neural activity serves to sharpen the resolution in the spectro-temporal domain. The cross-correlation for spikes contributing to the STRF depended much stronger on the STRF overlap than the cross-correlation during either silence or for spikes that did not contribute to the STRF (OUT-STRF). Compared with that for firings during silence, the cross-correlation for the OUT-STRF spikes was much reduced despite the unchanged firing rate. This suggests that stimulation breaks up the large neural assembly that exists during long periods of silence into a stimulus related one and maybe several others. As a result, the OUT-STRF spikes of the unit pairs, now likely distributed across several assemblies, are less correlated than during long periods of silence. Thus the ongoing network activity is significantly different from that during stimulation and changes afterng arousal during stimulation.

Impairment of eyeblink conditioning in GluRdelta2-mutant mice depends on the temporal overlap between conditioned and unconditioned stimuli

Mice lacking the glutamate receptor subunit delta2 (GluRdelta2) are deficient in cerebellar long-term depression (LTD) at the parallel fibre-Purkinje cell synapses. We conducted delay and trace eyeblink conditioning with these mice, using various temporal intervals between the conditioned stimulus (CS) and unconditioned stimulus (US). During trace conditioning in which a stimulus-free trace interval (TI) of 250, 100 or 50 ms intervened between the 352-ms tone CS and 100-ms US, GluRdelta2-mutant mice learned as successfully as wild-type mice. Even in the paradigm with TI = 0 ms, in which the end of CS and onset of US are simultaneous, there was no difference between the GluRdelta2-mutant and wild-type mice in their acquisition of a conditioned response. However, in the delay paradigm in which the 452-ms CS overlapped temporally with the coterminating 100-ms US, GluRdelta2-mutant mice exhibited severe learning impairment. The present study together with our previous work [Kishimoto, Y., Kawahara, S., Suzuki, M., Mori, H., Mishina, M. & Kirino, Y. (2001) Eur. J. Neurosci., 13, 1249-1254], indicates that cerebellar LTD-independent learning is possible in paradigms without temporal overlap between the CS and US. On the other hand, GluRdelta2 and cerebellar LTD are essential for learning when there is CS-US temporal overlap, suggesting that the cerebellar neural substrates underlying eyeblink conditioning may change, depending on the temporal overlap of the CS and US.

Neuronal mechanisms of executive control by the prefrontal cortex

Executive function is considered to be a product of the coordinated operation of various processes to accomplish a particular goal in a flexible manner. The mechanism or system responsible for the coordinated operation of various processes is called executive control. Impairments caused by damage to the prefrontal cortex are often called dysexecutive syndromes. Therefore, the prefrontal cortex is considered to play a significant role in executive control. Prefrontal participation to executive control can be partly explained by working memory that includes mechanisms for temporary active storage of information and processing stored information. For the prefrontal cortex to exert executive control, neuronal mechanisms for temporary storage of information and dynamic and flexible interactions among them are necessary. In this article, we present the presence of dynamic and flexible changes in the strength of functional interaction and extensive functional interactions among temporal information-storage processes in the prefrontal cortex. In addition, recent imaging studies show dynamic changes in functional connectivity between the prefrontal cortex and other cortical and subcortical structures depending upon the characteristics or the temporal context of the task. These observations indicate that the examination of dynamic and flexible modulation in neuronal interaction among prefrontal neurons as well as between the prefrontal cortex and other cortical and subcortical areas is important for explaining how the prefrontal cortex exerts executive control.

Improvements to the sensitivity of gravitational clustering for multiple neuron recordings

We outline two improvements to the technique of gravitational clustering for detection of neuronal synchrony, which are capable of improving the method's detection of weak synchrony with limited data. The advantages of the enhancements are illustrated using data with known levels of synchrony and different interspike interval distributions. The novel simulation method described can easily generate such test data. An important dependence of the sensitivity of gravitational clustering to the interspike interval distribution of the analysed spike trains is described.

How do cell assemblies encode information in the brain?

The present review discusses why cell-assembly coding, i.e. ensemble coding by functionally connected neurons, is a tenable view of the brain's neuronal code and how it operates in the working brain. The cell-assembly coding has two major properties, i.e., partial overlapping of neurons among assemblies and connection dynamics within and among the assemblies. The former is the ability of one neuron to participate in different types of information processing. The latter is the capability for functional synaptic connections, detected by activity correlations of the neurons, to change among different types of information processing. An example of a series of experiments which detected these two major properties is then given. Several relevant points concerning the detection of the actual dynamics of cell-assembly coding are also enumerated. They include the dependence of the type of cell-assembly coding on types of information-processing in different structures of the brain, sparse coding by distributed overlapped assemblies, and coincidence detection as a role of individual neurons to bind distributed neurons into cell assemblies.

Correlations between neural discharges are related to receptive field properties in cat primary auditory cortex

The functional role of correlated neural activity in auditory cortex for the processing of sounds was explored by investigating whether and how cross-correlation parameters are related to receptive field similarities of neurons. Multi-unit activity was recorded simultaneously from several sites of isofrequency domains in primary auditory cortex. At each site various receptive field properties were determined. From the discharges of pairs of clusters, normalized cross-correlation histograms (CCH) were calculated for extended periods of spontaneous activity and for periods with noise-burst stimulation. In both conditions, most CCHs exhibited a symmetrical positivity near the origin of the CCH, a few to several tens of milliseconds wide. Cross-correlation histograms were characterized with two parameters: the correlation strength, which was estimated from the peak correlation, and the correlation width, i.e. the time period of correlated firing, which was measured as the width of the positivity at half height. It was found that correlation strength increased and correlation width narrowed with increasing similarity of the receptive fields of two clusters. These relationships were observed both in the acoustically-driven and spontaneous conditions. Specifically, correlation strength was most strongly associated with similarity in binaural interaction and in temporal response properties such as response onset, response offset and the temporal pattern of the response. Correlation width was predominantly associated with similarity in characteristic frequency, bandwidth and intensity threshold. Results suggest that correlated activity, reflecting potential mechanisms involved in the neural computation in auditory cortex, provides a means to evaluate the properties of the functional organization of auditory cortex. Systematic relationships were found between correlation properties and the receptive field-based organization of cortical processing, suggesting that similar general mechanisms are utilized in many parts of the sensory cortex. In particular, the magnitude and/or the time period of synchronized firing of neurons is increased if the receptive field properties of the involved neurons are similar.

The search for cell assemblies in the working brain

This paper discusses why population ensemble coding by multiple neurons is a tenable view of the brain's basic neuronal code. The discussion is based on features of neuronal activity in working brains of behaving animals. The key concept to elucidate population ensemble coding is the 'cell assembly', i.e. overlapped populations of neurons with flexible functional connections within and among the populations. Recent examples of experimental approaches which indicate the cell-assembly coding of memory in the working brain are given. These experiments used a strategy that reveals two main properties of cell assemblies; the overlapping of neurons and the dynamic changes of synaptic connections in processing different kinds of memory. Several possible features of cell-assembly coding that might be explored in future experimental research are enumerated.

Population coding by cell assemblies--what it really is in the brain

One of the central questions in neuroscience concerns the basic code for information processing in the brain. Much experimental evidence and theoretical consideration have suggested that single-neuron coding is no longer a tenable hypothesis. The present review explains why population neuronal coding is valid and discusses how it is carried out in the brain. The main context is experimental access to real features of the coding in working brains as deduced from experimental research. Several recent studies recording neuronal activities from behaving animals have shown that ensemble activity of neurons represents specific information, indicating the reality of population coding by many neurons. The key concept which can integrate the experimental evidence is the 'cell assembly', i.e., overlapped populations of neurons with flexible functional connections within and among the populations. Correlated activity among the neurons constructs the functional connection. In order to see features of the cell-assembly coding, two main properties of cell assemblies in processing several different kinds of information must be investigated, that is, the overlapping of neurons and the dynamics of synaptic connections. This manner of coding can provide both the experimental and theoretical framework to detect the real dynamic features of information processing by the brain.

A comparison of tone-evoked response properties of 'cluster' recordings and their constituent single cells in the auditory cortex

This study examined relationships between some acoustic response properties of 'cluster' recordings (CL) and their constituent single cells (SU) in the auditory cortex obtained from 22 clusters comprised of 63 responsive single units mainly in the anterior tonotopic field of the waking guinea pig. Response parameters included characteristic frequency (CF), threshold (Th) at CF, bandwidth 10 (BW10) and 30 (BW30) dB above Th. Clusters and single units were classified by their pattern of discharges as either 'onset' or 'sustained' response types. Comparison of CL and their constituent SU revealed differences in one or more response parameters in all CL. The CFs of onset CL were generally the same as the CFs of their constituent onset SU in contrast to sustained CL for which greater differences were observed in CF. The Th of all CL differed from that of some of their cells. The BW of approximately 50% of CL differed from their SU. The findings indicate that cluster recordings are often not good predictors of the response parameters of all of their constituent neurons.

Excitatory interactions in neuronal networks which include cells of the auditory cortex and the medial geniculate body

The use of the method of cross-correlation analysis to elucidate the interactions between simultaneously recorded neurons from various loci of the auditory cortex (AC) and the medial geniculate body (MGB) has made it possible to identify the following characteristics of the functional organization of the excitatory interactions in the thalamocortical neuronal networks: the interdependant impulse action of neurons located at various loci of the AC and MGB was determined by reciprocal excitatory connections; the efficiency of the connections between neurons of the AC, 400-500 microns apart, and between tonotopically associated neurons of the AC and MGB was approximately identical (associations were identified in 12% of the cases); the "divergent" properties of the MGB (AC) neurons were manifested in the fact that one and same neuron could simultaneously excite both neighboring cells and neurons from one or several loci of the AC (MGB); the "convergent" properties of the AC and MGB neurons were manifested in the fact that cells located at various loci of the AC and MGB simultaneously excited one neuron. The results make it possible to explain the deviations observed in the investigation of RF of neurons of the AC and MGB from the principle of tonotopical organization. It is hypothesized that the character of the organization of the excitatory connections in the thalamocortical networks may promote the creation of the necessary conditions for the modification of the efficiency of synapses between all of the elements of the network during the stimulation of individual elements.

Investigating dynamic aspects of brain function in slice preparations: spatiotemporal stimulus patterns generated with an easy-to-build multi-electrode array

Electrical stimulation of nervous tissue with single stimulating electrodes is a technique widely used for the investigation of nervous system function. While it has proved to be useful in all kinds of experiments, single electrode stimuli are, however, far from being 'natural'. In most parts of the living brain, incoming activity results from the firing of a large number of presynaptic neurons, thus reflecting a complex combination of space and time aspects of neural activity. In this paper, a multi-electrode stimulating system is introduced which allows for the generation of fast space-time stimulus patterns. An example for the application of dynamic input patterns to the cerebellar cortex in vitro is given. The corresponding experiments revealed aspects of cerebellar function which cannot be seen using static or single electrode stimulation.

Statistical analysis of the functional connections between the neurons of the visual and motor cortex in different forms of conditioned reflex behavior

The organization of interneuronal cortical connections was studied in experiments on cats with the development of delayed alimentary instrumental reflexes in response to light. The dynamics of the intra- and interstructural neuronal network at the level of the cortical projection (visual and motor) zones of the cat brain in three forms of behavior was revealed through a cross correlation analysis: the realization of the CR, in the intersignal period with the presence and absence of instrumental movements. The predominance of "informational" or "motivational" neuronal connections was observed depending upon the forms of the behavior.

Organization of network properties of cells in local and distributed neuronal networks of the brain of cats

The network properties of neurons of the visual and motor cortex and of the lateral nucleus of the hypothalamus were investigated on the basis of identified interneuronal interactions, using the cross-correlation method of analysis, in cats with developed alimentary conditioned instrumental reflexes to light. The varied organization of the network properties of cortical neurons in the organization of local and distributed neuronal networks was demonstrated, namely: the predominance of divergent properties over convergent properties for large cells in local networks and the leveling out of these relationships in distributed networks. The neurons of the lateral nucleus of the hypothalamus had an equal representation of convergent and divergent properties in the organization of local and distributed networks. The network properties of neurons of the cortical and subcortical structures were manifested in the background, following the development of conditioned reflexes, and during extinction. Only the small cells of the visual cortex were functionally dependent and changed the relationship of network properties in local networks during the extinction of conditioned reflexes.

Evidence for spatiotemporal firing patterns within the auditory thalamus of the cat

Multiple spike trains were recorded in the auditory thalamus of cats. Each unit was studied before, during and after cooling of the ipsilateral primary auditory cortex, during spontaneous activity and acoustically evoked activity. The search for spatiotemporal firing patterns provided evidence that excess of patterns does exist and that the acoustical stimulation increased their number. Cortical cooling did not affect the probability of finding the firing pattern.

Physiological differentiation within the auditory part of the thalamic reticular nucleus of the cat

Spike trains of 153 single units were recorded in the caudoventral part of the thalamic reticular nucleus (RE) of 7 nitrous oxide anaesthetized cats. Functional properties defined by spontaneous activity pattern, studied by mean of auto renewal density histograms, were used to subdivide the units into 4 groups. Types I (18%), II (56%) and III (15%) were defined by an increasing bursting activity and Type IV (11%) by firing no bursts spontaneously. The responses to auditory stimuli confirmed that the caudoventral part of RE is tightly related to central auditory pathways. Responses to white noise bursts (200 ms duration) significantly let appear that Type I units responded in a high proportion (greater than 70%) until 80 ms after the stimulus onset, Type II units where mostly affected during the entire stimulus duration, and Type III units showed preferentially late responses. The units responsive to high frequencies (greater than 8 kHz) were mostly located in the dorsal and the units responsive to low frequencies (less than 2 kHz) in the anteroventral sector of auditory RE. However, only a loosely tonotopy is supported by this study. The neuronal circuitry within RE was shown to be stable when white noise bursts were delivered. Cross-correlograms indicated a large proportion of interconnected units (64%) and signs of mutual inhibition between neighboring RE units (11%). The hypothesis is discussed that the auditory RE exerts a fine control on the time-dependent analysis of the incoming auditory input to the cerebral cortex. The complex intranuclear connectivity suggests that the cell types correspond to distinct patterns of functional connections.

Cholinergic modulation of responses to single tones produces tone-specific receptive field alterations in cat auditory cortex

Acetylcholine (ACh), acting via muscarinic receptors, is known to modulate neuronal responsiveness in primary sensory neocortex. The administration of ACh to cortical neurons facilitates or suppresses responses to sensory stimuli, and these effects can endure well beyond the period of ACh application. In the present study, we sought to determine whether ACh produces a general change in sensory information processing, or whether it can specifically alter the processing of sensory stimuli with which it was "paired". To answer this question, we restricted acoustic stimulation in the presence of ACh to a single frequency, and determined single neuron frequency receptive fields in primary auditory cortex before and after this pairing. During its administration, ACh produced mostly facilitatory effects on spontaneous activity and on responses to the single frequency tone. Examination of frequency receptive fields after ACh administration revealed receptive field modifications in 56% of the cells. In half of these cases, the receptive field alterations were highly specific to the frequency of the tone previously paired with ACh. Thus ACh can produce stimulus-specific modulation of auditory information processing. An additional and unexpected finding was that the type of modulation during ACh administration did not predict the type of receptive field modulation observed after ACh administration; this may be related to the physiological "context" of the same stimulus in two different conditions. The implications of these findings for learning-induced plasticity in the auditory cortex is discussed.

Responses of single auditory cortical neurons to tone sequences

The responses of single neurons in the primary and secondary auditory cortex of cat were recorded during the presentation of sequences consisting of five tones of different frequencies. Discharges to tones within these sequences usually (84%) exhibited a dependence on the 'direction' of the sequence (ascending, descending, or mixed frequencies). For sequences consisting of 5 tones of identical frequency (monotone) the response often depended on serial position, including cases in which the neuron only responded to later tones in the sequence. Comparison of responses to heterogeneous and monotone sequences showed that response dependence on serial position was a factor in response dependence on sequence direction. Auditory cortical neurons can exhibit stronger responses to a tone presented in a sequence than to the same tone presented alone. Hence, the responses to tones within sequences may not be highly predictable from the responses to isolated tones.

Neuronal assemblies

This paper examines the concept of neuronal assembly as it has appeared in selected portions of the literature. The context is experimental access to real neuronal assemblies in working brains, as made possible by recent technological progress. One current measure of assembly organization is based on correlation of firing among neurons; recent observations show that such correlations can vary rapidly. In this paper, we demonstrate that dynamic firing correlation can be caused either by dynamic changes in neuronal connection strengths or, alternatively, by the effects of an unobserved (large) pool of other neurons. The static connectivity within the pool appears to be important in determining these effects.