Most of our knowledge about the functional organization of neuronal systems is based on the analysis of the firing patterns of individual neurons that have been recorded one by one in succession. This approach permits assessment of event-related variations in discharge rate, but it precludes detection of any covariations in the amplitude or timing of distributed responses if these covariations result from internal neuronal interactions rather than from time locking to stimulus or motor events. As it is likely that internal coordination of distributed responses is functionally as relevant as stimulus-induced coordination, multielectrode recordings are increasingly being used to analyze internally generated covariations of firing patterns. More than a decade ago, we used this method to reveal that neurons in the visual cortex tend to synchronize their discharges with a precision in the millisecond range when activated with a single contour (Gray and Singer, 1987, Soc. Neurosci., abstract; 35Barlow H.B. Critical limiting factors in the design of the eye and visual cortex.Proc. R. Soc. Lond. B Biol. Sci. 1981; 212: 1-34Crossref PubMed Google Scholar), whereas they fail to do so when activated by different contours moving in different directions (37Bauer H.-U. Pawelzik K. Alternating oscillatory and stochastic dynamics in a model for a neuronal assembly.Physica D. 1993; 69: 380-393Crossref Google Scholar, 26Bair W. Koch C. Temporal precision of spike trains in extrastriate cortex of the behaving monkey.Neural Comput. 1996; 8: 44-66Crossref Google Scholar). In addition, these stimulus-induced, context-dependent synchronization phenomena were found to be associated with a conspicuous oscillatory modulation of cell firing in a frequency range between 30 and 50 Hz, the so-called γ frequency range. Two aspects make this synchronization interesting. First, it results from internal coordination of spike timing and is not simply caused by stimulus-locked changes in discharge rate. Second, synchronization probability changes in a systematic way when the perceptual coherence of stimulus constellations is modified. Thus, this type of synchrony is not a trivial reflection of anatomical connectivity such as shared input through bifurcating axons, but instead results from context-dependent, dynamic interactions within the cortical network. The evidence for an internal coordination of spike timing raises the question of whether it serves a function in cortical processing or whether it is merely an epiphenomenon. The goal of this paper is to review theoretical arguments and data that relate to this issue. The first part examines how well the nervous system, and in particular the cerebral cortex, can distinguish between synchronous and asynchronous responses, and whether any significance is attributed to precisely synchronized discharge patterns when these are coordinated by external events, e.g., by the synchronous onset of sensory stimuli. In the second part, data are reviewed from experiments that were designed to examine putative functions of internally generated synchronization. As the assessment of internally generated, non-stimulus-locked temporal relations requires the joint evaluation of responses from more than one neuron, only experiments that permit simultaneous measurements of responses from multiple units are considered. These include multielectrode recordings from multiple individual cells, but also measurements of local field potentials (LFPs) and electroencephalographic (EEG) or magnetoencephalographic (MEG) recordings. The signals of these latter methods reflect the average activity of large cell populations. Because this activity leads to measurable signal fluctuations only if it is sufficiently synchronized, these global recordings provide valuable information about the temporal relations between responses. Discussions about the putative functional role of synchrony focus on the question of whether it can serve as a mechanism to bind distributed neuronal activity. To provide an adequate background for the examination of this question, I shall first deal with some general, implementation-independent aspects of binding operations. As the Gestalt psychologists pointed out, our cognitive systems have the tendency to interpret objects and events as related if they are contiguous in space or time, or if they exhibit similarities in certain feature domains. Thus, contours that touch one another, have similar contrast, or move with the same speed in the same direction (common fate) are more likely to be perceived as components of the same object than spatially distant contours or contours that have no features in common. Likewise, events that coincide in time are interpreted with greater probability as related than events separated in time. At early stages of sensory processing, spatial relations and relations in feature space are represented by the amplitude and the topological relations of activation foci in ordered maps. Temporal relations, however, are represented by the relative timing of responses. In order to accomplish perceptual grouping, the distributed responses of feature-selective cells need to be bound together at some stage of processing. Evidence indicates that this is achieved in two complementary ways. One strategy is binding of responses by convergence of axonal projections. Axons of cells whose responses should be bound are made to converge onto a common target cell at the next-higher processing level. If the threshold of this binding unit is appropriately adjusted, its response signals the specific conjunction of features to which the feeding cells are tuned. We shall address this grouping strategy as “binding by convergence” or “binding by conjunction cells.” This coding principle is also known as “labeled line coding” because the responses of a given unit have a fixed label attached to them; they always signal the same conjunction of input signals. The complementary strategy for response binding relies on dynamic selection and grouping of responses. Here, responses are bound by jointly enhancing their saliency relative to other, nonbound responses. Enhanced responses have a stronger impact on downstream processes than nonenhanced responses and therefore dominate subsequent computations. Thus, the results of these computations will reflect the specific configuration of features to which cells with enhanced responses are tuned. We shall address this selection and grouping strategy as “dynamic binding” and the associated coding principle as “relational coding” or “assembly coding,” because here the information about a particular conjunction is contained in the dynamically adjustable configuration of the enhanced responses of distributed neurons (for reviews of the extensive literature on labeled line and assembly coding, see 97Chino Y.M. Smith E.L. Wada H. Ridder W.H. Langston A.L. Lesher G.A. Disruption of binocularly correlated signals alters the postnatal development of spatial properties in cat striate cortical neurons.J. Neurophysiol. 1991; 65: 841-859PubMed Google Scholar, and other contributions in this issue of Neuron). The topology of connections in cortical networks reflects these two grouping strategies and accounts well for perceptual grouping according to contiguity in euclidian and feature space. The first strategy is implemented by the highly complex recombination of feedforward cortico-cortical connections. It leads to a large variety of conjunction-specific neurons, the complexity of which increases as one proceeds along the processing hierarchy. The second strategy appears to rely on two other classes of cortico-cortical connections (reviewed by 95Cheng K. Hasegawa T. Saleem K.S. Tanaka K. Comparison of neuronal selectivity for stimulus speed, length, and contrast in the prestriate visual cortex areas V4 and MT of the macaque monkey.J. Neurophysiol. 1994; 71: 2269-2280PubMed Google Scholar, 78Burr D. Acuity for apparent vernier offset.Vision Res. 1979; 19: 835-837Crossref PubMed Scopus (40) Google Scholar): (1) reciprocal connections that link cells situated within the same cortical area, as well as cells distributed across different areas but occupying the same level in the processing hierarchy; and (2) feedback connections that reciprocate the feedforward connections. Together, these reciprocal cortico-cortical connections constitute the large majority of synaptic inputs to cortical cells. For the intraareal connections in primary visual cortex, it is established that they preferentially couple neurons that are nearby or that code for similar features (33Barbas H. Pandya D.N. Architecture and intrinsic connections of the prefrontal cortex in the rhesus monkey.J. Comp. Neurol. 1989; 286: 353-375Crossref PubMed Google Scholar, 90Cave K.R. Kosslyn S.M. Varieties of size-specific visual selection.J. Exp. Psychol. Gen. 1989; 118: 148-164Crossref PubMed Google Scholar, 91Cave K.R. Wolfe J.M. Modeling the role of parallel processing in visual search.Cogn. Psychol. 1990; 22: 225-271Crossref PubMed Google Scholar). Hence, if these neurons are coactivated, they are more likely to have the saliency of their responses enhanced jointly by cooperative interactions than are neurons that are far apart or tuned to very dissimilar features. As a consequence, responses to contours that are contiguous in euclidian and/or feature space have an enhanced probability of being processed jointly and, thus, of being bound together. In contrast to the numerous experimental and theoretical studies devoted to the analysis of grouping operations in the domain of spatial features, comparatively few studies have been devoted to the questions of (1) to what extent temporal contiguity of stimuli is exploited for perceptual grouping and (2) through which neuronal mechanisms such grouping could be achieved. Most of the evidence regarding perceptual grouping by temporal cues comes from recent psychophysical studies on vision. These studies indicate that spatially distributed contour elements are bound perceptually and interpreted as elements of a coherent figure if they appear or change synchronously, while elements that follow different time courses are perceived as unrelated (55Borisyuk R.M. Borisyuk G.N. Kazanovich Y.B. The synchronization principle in modelling of binding and attention.Membr. Cell Biol. 1998; 11: 753-761PubMed Google Scholar, 4Abeles M. Prut Y. Bergman H. Vaadia E. Synchronization in neuronal transmission and its importance for information processing.Prog. Brain Res. 1994; 102: 395-404Crossref PubMed Google Scholar, 113Crewther D. Crewther S. Neural site of strabismic amblyopia in cats spatial frequency deficit in primary cortical neurons.Exp. Brain Res. 1990; 79: 615-622Crossref PubMed Google Scholar, 54Borg-Graham L.J. Monier C. Fregnac Y. Visual input evokes transient and strong shunting inhibition in visual cortical neurons.Nature. 1998; 393: 369-373Crossref PubMed Scopus (377) Google Scholar). The temporal resolution of this grouping mechanism is surprisingly high. Temporal offsets between the respective appearances of figure and ground elements of <10 ms still support perceptual grouping (55Borisyuk R.M. Borisyuk G.N. Kazanovich Y.B. The synchronization principle in modelling of binding and attention.Membr. Cell Biol. 1998; 11: 753-761PubMed Google Scholar). Because responses of neurons in the visual cortex follow the time course of the stimuli that evoke them, the results of these studies suggest that synchronous responses are bound perceptually while asynchronous responses are interpreted as unrelated. It is noteworthy that the short offsets that support this segmentation are not perceptible, suggesting a dissociation between the perceptibility of small differences in the time course of stimuli on the one hand and the effect of such differences on perceptual grouping on the other. This dissociation may have to do with the fact that, in vision, the temporal cues supporting grouping as a function of stimulus synchrony are mediated mainly by the magnocellular pathway, while the other, nontemporal grouping cues are mediated by both the magno- and the parvocellular pathway (56Bosking W.H Zhang Y. Schofield B. Fitzpatrick D. Orientation selectivity and the arrangement of horizontal connections in tree shrew striate cortex.J. Neurosci. 1997; 17: 2112-2127PubMed Google Scholar). These two pathways interact at multiple levels but subserve somewhat different functions. The magnocellular pathway is exquisitely sensitive to temporal features and can signal stimulus transients with high temporal resolution, while the parvocellular pathway operates with low temporal but high spatial and spectral resolution (for a review of the extensive literature, see 56Bosking W.H Zhang Y. Schofield B. Fitzpatrick D. Orientation selectivity and the arrangement of horizontal connections in tree shrew striate cortex.J. Neurosci. 1997; 17: 2112-2127PubMed Google Scholar). This functional dichotomy is relevant in the present context because it is a likely basis for the ability of the visual system to use both temporal and spatial cues in parallel for perceptual grouping. If within the same matrix of line elements one figure is defined by the synchronous onset of elements (temporal cue), and another spatially overlapping figure by differences in the orientation of the respective line elements (spatial cue), either the temporally or the spatially defined figure is perceived depending on the relative saliency of the two cues (temporal offset versus orientation difference) (56Bosking W.H Zhang Y. Schofield B. Fitzpatrick D. Orientation selectivity and the arrangement of horizontal connections in tree shrew striate cortex.J. Neurosci. 1997; 17: 2112-2127PubMed Google Scholar). This suggests that spatial and temporal grouping cues are processed in parallel and, if they conflict, the less salient cue is disregarded. This ability of the visual system to rely on either spatial or temporal cues, if the two cues are in conflict, is of considerable functional relevance. On the one hand, it permits binding of nontemporal features that are related but attached to temporally dispersed elements. On the other hand, it allows segregation of features that are unrelated but attached to temporally contiguous elements. The ability to base perceptual grouping on either spatial or temporal cues is also likely the cause of an apparent conflict between the psychophysical studies that support grouping based on temporal cues (see above) and a study that denies such a mechanism. 45Best J. Reuss S. Dinse H.R.O. Lamina-specific differences of visual latencies following photic stimulation in the cat striate cortex.Brain Res. 1986; 385: 356-360Crossref PubMed Google Scholar found that perception of figures defined by spatial cues is not impaired if false temporal conjunctions are introduced at random by presenting selected elements of the figure synchronously with elements of the background. Here, the temporal cues did not define a figure and hence may have been simply discarded through competition. In summary, the results of psychophysical studies suggest the following conclusions. First, information about the temporal parameters of stimuli is transmitted over several processing stages with a precision in the millisecond range. Second, asynchronies among spatially distributed responses of <10 ms are exploitable for perceptual grouping. Third, the mechanism that evaluates temporal relations among responses for perceptual grouping interprets synchronous responses as related and segregates them from responses that are temporally offset. Fourth, temporal and nontemporal grouping cues are evaluated in parallel, the former being conveyed mainly by the magnocellular pathway. These findings raise the question of how temporal grouping cues are evaluated at the neuronal level. In analogy to grouping mechanisms for nontemporal features (see above), it would suffice that the synchronous responses to simultaneously appearing or simultaneously changing stimuli are more salient, i.e., have a stronger joint impact on cells at subsequent processing stages than the asynchronous responses to temporally dispersed stimuli. Two nonexclusive scenarios may be considered. First, synchronously active cells might cooperate particularly effectively through cortico-cortical connections and thereby increase their discharge rate. Second, synchronous responses might by themselves and without further amplification have a stronger impact on cells at subsequent processing stages than cells responding to temporally offset stimuli. In both cases, two prerequisites need to be fulfilled. First, timing of discharges must be preserved across polysynaptic transmission chains with a precision in the millisecond range. Second, neurons must be able to differentiate between synchronous and asynchronous input. Synchronous excitatory postsynaptic potentials (EPSPs) must be more efficient than temporally dispersed EPSPs, and dispersions of <10 ms must already make a significant difference. Contrary to what one should expect from the long time constants of synaptic integration in central neurons (see, e.g., 92Chance F. Nelson S. Abbott L. Complex cells as cortically amplified simple cells.Nat. Neurosci. 1999; 2: 277-282Crossref PubMed Scopus (119) Google Scholar), cortical networks can operate with amazing temporal precision. In the auditory cortex of mammals, the spiking patterns of single-cell responses to species-specific calls reproduce with millisecond precision from trial to trial (15Alonso J.-M. Usrey W.M. Reid R.C. Precisely correlated firing in cells of the lateral geniculate nucleus.Nature. 1996; 383: 815-819Crossref PubMed Scopus (270) Google Scholar, 44Bergen J.R. Julesz B. Parallel versus serial processing in rapid pattern discrimination.Nature. 1983; 303: 696-698Crossref PubMed Google Scholar). Comparable accuracy is found in song birds for auditory neurons responding to songs and for central motor neurons controlling the vocalization patterns (119Das A. Gilbert C.D. Topography of contextual modulations mediated by short-range interactions in primary visual cortex.Nature. 1999; 399: 655-661Crossref PubMed Scopus (190) Google Scholar, 20Arguin M. Cavanagh P. Joanette Y. Visual feature integration with an attention deficit.Brain Cogn. 1994; 24: 44-56Crossref PubMed Scopus (24) Google Scholar). In cat visual cortex, neurons faithfully follow flicker frequencies of up to 50 Hz and on occasion even up to 100 Hz (81Buzsaki G. The hippocampo-neocortical dialogue.Cereb. Cortex. 1996; 6: 81-92Crossref PubMed Scopus (366) Google Scholar). Highly synchronous oscillatory discharges of retinal responses that reach oscillation frequencies of up to 100 Hz are also transmitted reliably from the retina to primary visual cortex (76Bülthoff H.H. Edelman S.Y. Tarr M.J. How are three-dimensional objects represented in the brain?.Cereb. Cortex. 1995; 3: 247-260Crossref Google Scholar; Castelo-Branco et al., 1998a; Herculano et al., 1999; 77Buračas G. Zador A. DeWeese M. Albright T. Efficient discrimination of temporal patterns by motion-sensitive neurons in primate visual cortex.Neuron. 1998; 20: 959-969Abstract Full Text Full Text PDF PubMed Scopus (214) Google Scholar) (Figure 1). Even neurons in the medial temporal cortex (MT/V5) of macaque monkeys, which are at least four synaptic stages away from the retina, signal the time structure of temporally modulated visual stimuli with a precision in the millisecond range (Buracas et al., 1998). Further indication of high temporal fidelity in neuronal transmission comes from evidence that neurons distributed within and across cortical areas, and even across the cerebral hemispheres and subcortical structures, can synchronize their spike discharges on the basis of oscillations in the γ frequency range (reviewed by 97Chino Y.M. Smith E.L. Wada H. Ridder W.H. Langston A.L. Lesher G.A. Disruption of binocularly correlated signals alters the postnatal development of spatial properties in cat striate cortical neurons.J. Neurophysiol. 1991; 65: 841-859PubMed Google Scholar; for more recent findings, see 58Bouyer J.J. Montaron M.F. Rougeul A. Fast fronto-parietal rhythms during combined focused attentive behaviour and immobility in cat cortical and thalamic localizations.Electroencephalogr. Clin. Neurophysiol. 1981; 51: 244-252Abstract Full Text PDF PubMed Scopus (100) Google Scholar, 10Aertsen A. Diesmann M. Gewaltig M.-O. Propagation of synchronous spiking activity in feedforward neural networks.J. Physiol. (Paris). 1996; 90: 243-247Crossref PubMed Scopus (63) Google Scholar, 29Bair W. Cavanaugh J.R. Movshon J.A. Reconstructing stimulus velocity from neuronal responses in area MT.in: Mozer M.C. Jordan M.I. Petsche T. Advances in Neural Information Processing Systems. MIT Press, Cambridge, MA1997Google Scholar, 69Britten K.H. Newsome W.T. Shadlen M.N. Celebrini S. Movshon J.A. A relationship between behavioral choice and the visual responses of neurons in macaque MT.Vis. Neurosci. 1996; 13: 87-100Crossref PubMed Google Scholar). This implies that dispersion of spike timing must have remained below the duration of a half cycle, i.e., below about 10 ms for cortical interactions and below 5 ms for retino-cortical transmission. The conspicuous spike patterns described by 79Bush P. Sejnowski T. Inhibition synchronizes sparsely connected cortical neurons within and between columns in realistic network models.J. Comput. Neurosci. 1996; 3: 91-110Crossref PubMed Scopus (93) Google Scholar in the prefrontal cortex point in the same direction. As proposed by 1Abeles M. Role of cortical neuron integrator or coincidence detector?.Israel J. Med. Sci. 1982; 18 (a): 83-92PubMed Google Scholar and recently again by 93Chelazzi L. Miller E.K. Duncan J. Desimone R.C. A neural basis for visual search in inferior temporal cortex.Nature. 1993; 363: 345-347Crossref PubMed Scopus (556) Google Scholar, one possible way to achieve such high temporal precision in neuronal signaling despite “slow” neurons is synchronization of discharges across parallel channels, a special form of population coding. In the proposed models, this is achieved by cross-coupling parallel channels through diverging and converging axon collaterals. As demonstrated recently (3Abeles M Corticonics. Cambridge University Press, Cambridge1991Crossref Google Scholar, 18Anderson C.H. Van Essen D.C. Shifter circuits a computational strategy for dynamic aspects of visual processing.Proc. Natl. Acad. Sci. USA. 1987; 84: 6297-6301Crossref PubMed Google Scholar), such synfire chains (1Abeles M. Role of cortical neuron integrator or coincidence detector?.Israel J. Med. Sci. 1982; 18 (a): 83-92PubMed Google Scholar) have the interesting property that the synchronization of discharges across parallel channels does not decrease from one synaptic level to the next but may even increase if coupling is appropriately adjusted. The reason for this preservation of precision is that synchronized EPSP barrages are more effective in triggering postsynaptic spikes than temporally dispersed inputs (see below). The synchronous EPSP barrages generated in such synfire chains elicit postsynaptic spikes with minimal latency jitter and hence can transmit the temporal signature of stimuli with high precision over many synaptic stages. Note that what matters for this temporal precision in transmission is the rise time of the compound EPSPs rather than the passive membrane time constant of the integrating neurons. If synchronized responses are grouped because they are more salient than nonsynchronized responses, neurons evaluating temporal grouping cues must respond differently to precisely synchronized and temporally dispersed barrages of EPSPs, and—as suggested by psychophysics—dispersions of <10 ms must be detectable. Again, at first glance, the long time constants of neuronal membranes seem incompatible with such differential sensitivity to coincident and dispersed input, but experimental observations suggest the contrary. In hippocampal cultures, most of the spontaneously occurring spikes are triggered by synchronously arriving EPSPs rather than by the smaller and more numerous temporally dispersed EPSPs, suggesting a privileged role of synchronized activity in synaptic transmission (100Cobb S.R. Buhl E.H. Halasy K. Paulsen O. Somogyi P. Synchronization of neuronal activity in hippocampus by individual GABAergic interneurons.Nature. 1995; 378: 75-78Crossref PubMed Google Scholar). The same conclusion is suggested by in vivo intracellular recordings from pyramidal cells of the monkey motor cortex (71Brody C.D. Disambiguating different covariation types.Neural Comput. 1999; 11 (a): 1527-1535Crossref PubMed Google Scholar). Likewise, simultaneous recordings from coupled neuron triplets along thalamo-cortical (7Adelson E.H. Lightness perception and lightness illusions.in: Gazzaniga M.S. The Cognitive Neurosciences. MIT Press, Cambridge, MA1999Google Scholar, 114Crick F. Function of the thalamic reticular complex the searchlight hypothesis.Proc. Natl. Acad. Sci. USA. 1984; 81: 4586-4590Crossref PubMed Google Scholar) and intracortical pathways (6Adelson E.H. Perceptual organization and the judgment of brightness.Science. 1993; 262: 2042-2044Crossref PubMed Google Scholar) in the visual system have revealed that EPSPs synchronized within intervals below 2 ms are more effective than EPSPs dispersed over longer intervals. Evidence on a more global level for the enhanced saliency of synchronized activity has been obtained with simultaneous recordings from several sites (area 18 and the posterior mediolateral suprasylvian sulcus [PMLS]) in cat visual cortex and retinotopically corresponding sites in the superior colliculus (10Aertsen A. Diesmann M. Gewaltig M.-O. Propagation of synchronous spiking activity in feedforward neural networks.J. Physiol. (Paris). 1996; 90: 243-247Crossref PubMed Scopus (63) Google Scholar). The impact of a particular group of cortical cells on target cells in the colliculus increased dramatically when the cortical cells' discharge was synchronized with other cortical cell groups projecting to the same collicular site (Figure 2). Enhanced saliency of synchronized responses can also be inferred from the tight correlation between perception and the strength of neuronal response synchronization observed in experiments on binocular rivalry in cats (Fries et al., 1997a) and human subjects (109Corbetta M. Miezin F.M. Shulman G.L. Petersen S.E. A PET study of visuospatial attention.J. Neurosci. 1993; 13: 1202-1226PubMed Google Scholar, 110Corbetta M. Shulman G.L. Miezin F.M. Petersen S.E. Superior parietal cortex activation during spatial attention shifts and visual feature conjunction.Science. 1995; 270: 802-805Crossref PubMed Google Scholar) (see also below). Finally, simulation studies also indicate that neurons with conventional integrate-and-fire properties can be quite sensitive to the temporal dispersion of synaptic input. Large-scale simulations of biologically inspired thalamocortical networks revealed that neurons exhibited a strong tendency to engage in oscillatory firing patterns and to synchronize their responses. When synchrony was artificially disrupted by introducing a jitter in spike timing, transmission across polysynaptic pathways was drastically reduced (62Braitenberg V. Schuz A Anatomy of the Cortex. Springer, Berlin1991Crossref Google Scholar, 63Braun J. Julesz B. Dividing attention at little cost detection and discrimination tasks.Percept. Psychophys. 1998; 60: 1-23Crossref PubMed Google Scholar). There are several mechanisms, some of which have been identified only recently, that make synchronously arriving EPSPs more efficient than temporally dispersed EPSPs. First, because of their exponential decay, simultaneous EPSPs summate more effectively than temporally dispersed EPSPs, and there is some evidence for supralinear summation due to voltage-gated dendritic conductances. Second, firing threshold is sensitive to the rising slope of the depolarization and lowers for fast-rising depolarizations (C. M. Gray, personal communication). Third, the effect of EPSPs is dramatically enhanced when these coincide with a back-propagating dendritic spike and hence with the input that generated this spike (51Blake R. Yang Y. Spatial and temporal coherence in perceptual binding.Proc. Natl. Acad. Sci. USA. 1997; 94: 7115-7119Crossref PubMed Scopus (29) Google Scholar). All three mechanisms are sensitive to dispersions of EPSPs in the range of a few milliseconds. In conclusion, both psychophysical and physiological evidence indicates that neuronal networks are exquisitely sensitive to temporal relations among discharges in input connections, assigning particular significance to coincident, i.e., synchronous input. Synchronicity serves as a tag of relatedness most likely because it causes an increase in the saliency of the synchronized responses, which in turn favors their joint evaluation (binding) at subsequent processing stages. Conditions are thus comparable to those in which figure elements have higher contrast than do elements of the background: in that case, too, responses become grouped according to saliency. The only difference is that, in the case of enhanced contrast, saliency is increased because of higher discharge rates rather than synchronization. Internally generated response synchrony closely resembles that induced by synchronously presented stimuli with respect to both its temporal precision and its magnitude (81Buzsaki G. The hippocampo-neocortical dialogue.Cereb. Cortex. 1996; 6: 81-92Crossref PubMed Scopus (366) Google Scholar). This raises the question of whether the internally synchronized discharges affect processing in the same way as externally induced synchrony. If so, interna
更多
原文链接
分享
引用
