Besides failing for the reasons Brette gives, codes fail to help us understand brain function because codes imply algorithms that compute outputs without reference to the signals' meanings. Algorithms cannot be found in the brain, only manipulations that operate on meaningful signals and that cannot be described as computations, that is, sequences of predefined operations.

Background: Propofol produces memory impairment at concentrations well below those abolishing consciousness. Episodic memory, mediated by the hippocampus, is most sensitive. Two potentially overlapping scenarios may explain how gamma-aminobutyric acid receptor type A (GABA(A)) potentiation by propofol disrupts episodic memory-the first mediated by shifting the balance from excitation to inhibition while the second involves disruption of rhythmic oscillations. We use a hippocampal network model to explore these scenarios. The basis for these experiments is the proposal that the brain represents memories as groups of anatomically dispersed strongly connected neurons. Methods: A neuronal network with connections modified by synaptic plasticity was exposed to patterned stimuli, after which spiking output demonstrated evidence of stimulus-related neuronal group development analogous to memory formation. The effect of GABA(A) potentiation on this memory model was studied in 100 unique networks. Results: GABA(A) potentiation consistent with moderate propofol effects reduced neuronal group size formed in response to a patterned stimulus by around 70%. Concurrently, accuracy of a Bayesian classifier in identifying learned patterns in the network output was reduced. Greater potentiation led to near total failure of group formation. Theta rhythm variations had no effect on group size or classifier accuracy. Conclusions: Memory formation is widely thought to depend on changes in neuronal connection strengths during learning that enable neuronal groups to respond with greater facility to familiar stimuli. This experiment suggests the ability to form such groups is sensitive to alteration in the balance between excitation and inhibition such as that resulting from administration of a gamma-aminobutyric acid-mediated anesthetic agent.

Neurons send signals to each other by means of sequences of action potentials (spikes). Ignoring variations in spike amplitude and shape that are probably not meaningful to a receiving cell, the information content, or entropy of the signal depends on only the timing of action potentials, and because there is no external clock, only the interspike intervals, and not the absolute spike times, are significant. Estimating spike train entropy is a difficult task, particularly with small data sets, and many methods of entropy estimation have been proposed. Here we present two related model-based methods for estimating the entropy of neural signals and compare them to existing methods. One of the methods is fast and reasonably accurate, and it converges well with short spike time records; the other is impractically time-consuming but apparently very accurate, relying on generating artificial data that are a statistical match to the experimental data. Using the slow, accurate method to generate a best-estimate entropy value, we find that the faster estimator converges to this value more closely and with smaller data sets than many existing entropy estimators.

The visual system uses continuity as a cue for grouping oriented line segments that define object boundaries in complex visual scenes. Many studies support the idea that long-range intrinsic horizontal connections in early visual cortex contribute to this grouping. Top-down influences in primary visual cortex (V1) play an important role in the processes of contour integration and perceptual saliency, with contour-related responses being task dependent. This suggests an interaction between recurrent inputs to V1 and intrinsic connections within V1 that enables V1 neurons to respond differently under different conditions. We created a network model that simulates parametrically the control of local gain by hypothetical top-down modification of local recurrence. These local gain changes, as a consequence of network dynamics in our model, enable modulation of contextual interactions in a task-dependent manner. Our model displays contour-related facilitation of neuronal responses and differential foreground vs. background responses over the neuronal ensemble, accounting for the perceptual pop-out of salient contours. It quantitatively reproduces the results of single-unit recording experiments in V1, highlighting salient contours and replicating the time course of contextual influences. We show by means of phase-plane analysis that the model operates stably even in the presence of large inputs. Our model shows how a simple form of top-down modulation of the effective connectivity of intrinsic cortical connections among biophysically realistic neurons can account for some of the response changes seen in perceptual learning and task switching.

The singing of juvenile songbirds is highly variable and not well stereotyped, a feature that makes it difficult to analyze with existing computational techniques. We present here a method suitable for analyzing such vocalizations, windowed spectral pattern recognition (WSPR). Rather than performing pairwise sample comparisons, WSPR measures the typicality of a sample against a large sample set. We also illustrate how WSPR can be used to perform a variety of tasks, such as sample classification, song ontogeny measurement, and song variability measurement. Finally, we present a novel measure, based on WSPR, for quantifying the apparent complexity of a bird's singing.

BACKGROUND:The understanding of how general anesthetics act on individual cells and on global brain function has increased significantly during the last decade. What remains poorly understood is how anesthetics act at intermediate scales. Several major theories emphasize the importance of neuronal groups, sets of strongly connected neurons that fire in a time-locked fashion, in all aspects of brain function, particularly as a necessary substrate of consciousness. The authors have undertaken computer modeling to determine how ã-aminobutyric acid receptor type A (GABAA) receptor potentiating agents such as propofol may influence the dynamics of neuronal group formation and ongoing activity.METHODS:A computer model of a cortical network with connections modified by synaptic plasticity was examined. At baseline, the model spontaneously formed neuronal groups. Direct effects of GABAA receptor potentiation and indirect effects on input drive were then examined to study their effects on this process.RESULTS:Potentiation of GABAA inhibition and input drive reduction reduced the firing frequency of inhibitory and excitatory neurons in a dose-dependent manner. The diminution in spiking rates led to dramatic reductions in the firing frequency of neuronal groups. Simulated electroencephalographic output from the model at baseline exhibits gamma and theta rhythmicity. The direct and indirect GABAA effects reduce the amplitude of these underlying rhythms and modestly slow the gamma rhythm.CONCLUSIONS:GABAA facilitation both directly and indirectly inhibits the ability of neurons to form groups spontaneously. A lack of group formation is consistent with some theories of anesthetic-induced loss of memory formation and consciousness.

We review a concept of the most primitive, fundamental function of the vertebrate CNS, generalized arousal (GA). Three independent lines of evidence indicate the existence of GA: statistical, genetic, and mechanistic. Here we ask, is this concept amenable to quantitative analysis? Answering in the affirmative, four quantitative approaches have proven useful: ( i ) factor analysis, ( ii ) information theory, ( iii ) deterministic chaos, and ( iv ) application of a Gaussian equation. It strikes us that, to date, not just one but at least four different quantitative approaches seem necessary for describing different aspects of scientific work on GA.

Entropy rate quantifies the change of information of a stochastic process (Cover & Thomas, 2006). For decades, the temporal dynamics of spike trains generated by neurons has been studied as a stochastic process (Barbieri, Quirk, Frank, Wilson, & Brown, 2001; Brown, Frank, Tang, Quirk, & Wilson, 1998; Kass & Ventura, 2001; Metzner, Koch, Wessel, & Gabbiani, 1998; Zhang, Ginzburg, McNaughton, & Sejnowski, 1998). We propose here to estimate the entropy rate of a spike train from an inhomogeneous hidden Markov model of the spike intervals. The model is constructed by building a context tree structure to lay out the conditional probabilities of various subsequences of the spike train. For each state in the Markov chain, we assume a gamma distribution over the spike intervals, although any appropriate distribution may be employed as circumstances dictate. The entropy and confidence intervals for the entropy are calculated from bootstrapping samples taken from a large raw data sequence. The estimator was first tested on synthetic data generated by multiple-order Markov chains, and it always converged to the theoretical Shannon entropy rate (except in the case of a sixth-order model, where the calculations were terminated before convergence was reached). We also applied the method to experimental data and compare its performance with that of several other methods of entropy estimation.

We investigated the effects of β-estradiol on the locomotor behavior of female mice in a radial maze. Data comprising the total distance traveled during each arm entry were obtained from video records of six consecutive daily recording sessions. Distributions of these data were bimodal for both ovariectomized control and β-estradiol-treated ovariectomized subjects. Data were fit with the sum of two gamma probability distributions. Three parameters of the analytic fits were useful for quantifying the effect of β-estradiol on locomotor behavior: (i) the sampling distance (median of the total distance traveled during each arm entry in the short-distance peak of a bimodal distribution), (ii) the committed distance (median of the total per-arm-entry distance traveled in the long-distance peak), and (iii) the partition distance (distance represented by the minimum between the two peaks). Analysis showed that for sampling-distance arm entries β-estradiol typically had little if any significant effect on female locomotor behavior, whereas it significantly increased the total distance traveled during committed-distance arm entries on the first 2 days of exposure to the empty maze. β-Estradiol also increased the ability of females to discriminate between empty maze arms and arms that contained intact or castrated male mice and partially prevented loss of this capacity after removal of the males.

A recent theoretical emphasis on complex interactions within neural systems underlying consciousness has been accompanied by proposals for the quantitative characterization of these interactions. In this article, we distinguish key aspects of consciousness that are amenable to quantitative measurement from those that are not. We carry out a formal analysis of the strengths and limitations of three quantitative measures of dynamical complexity in the neural systems underlying consciousness: neural complexity, information integration, and causal density. We find that no single measure fully captures the multidimensional complexity of these systems, and all of these measures have practical limitations. Our analysis suggests guidelines for the specification of alternative measures which, in combination, may improve the quantitative characterization of conscious neural systems. Given that some aspects of consciousness are likely to resist quantification altogether, we conclude that a satisfactory theory is likely to be one that combines both qualitative and quantitative elements.

We employ computer simulations to explore the effect of different temporal patterns of afferent impulses on the evoked discharge of a model cerebellar Purkinje cell. We show that the frequency and temporal correlation of impulses across afferent fibers determines which of four regimes of discharge activity is evoked. In the uncorrelated, here Poissonian, case, (i) cell discharge is determined by the total stimulation rate and temporal patterns of discharge are the same for different combinations of afferent fiber number and mean impulse rate per fiber giving the same total stimulation. Alternatively, if temporal correlations are present in the stimulus, (ii) for stimulation frequencies of 4 to at least 64 Hz there is a narrow range of afferent fiber number for which every stimulus pulse (composed of a single impulse on each afferent fiber) evokes a single action potential. In this case cell discharge is frequency locked to the stimulus with a concomitant reduction in discharge variability. (iii) For lower fiber numbers and thus discharge frequencies lower than the locking frequency, the variability of cell discharge is typically independent of afferent impulse timing, whereas, (iv) at higher fiber numbers and thus higher discharge frequencies, the reverse is true. We conclude that in case (iii) the cell acts as an integrator and discharge is determined by the stimulation rate, whereas in case (iv) the cell acts as a coincidence detector and the timing of discharge is determined by the temporal pattern of afferent stimulation. We discuss our results in terms of their significance for neuronal activity at the network level and suggest that the reported effects of varying stimulus timing and afferent convergence can be expected to obtain also with other principal cell types within the central nervous system.

To better understand the role of timing in the function of the nervous system, we have developed a methodology that allows the entropy of neuronal discharge activity to be estimated from a spike train record when it may be assumed that successive interspike intervals are temporally uncorrelated. The so-called interval entropy obtained by this methodology is based on an implicit enumeration of all possible spike trains that are statistically indistinguishable from a given spike train. The interval entropy is calculated from an analytic distribution whose parameters are obtained by maximum likelihood estimation from the interval probability distribution associated with a given spike train. We show that this approach reveals features of neuronal discharge not seen with two alternative methods of entropy estimation. The methodology allows for validation of the obtained data models by calculation of confidence intervals for the parameters of the analytic distribution and the testing of the significance of the fit between the observed and analytic interval distributions by means of Kolmogorov-Smirnov and Anderson-Darling statistics. The method is demonstrated by analysis of two different data sets: simulated spike trains evoked by either Poissonian or near-synchronous pulsed activation of a model cerebellar Purkinje neuron and spike trains obtained by extracellular recording from spontaneously discharging cultured rat hippocampal neurons.

The three-dimensional structure of 132microglobulin, the light chain of the major histocompatibility complex class I antigens, has been determined by x-ray crystallography. An electron density map of the bovine protein was calculated at a nominal resolution of 2.9 A by using the methods of multiple isomorphous replacement and electron density modification refinement. The molecule is approximately 45 x 25 x 20 A in size. Almost half of the amino acid residues participate in two large .3 structures, one of four strands and the other of three, linked by a central disulfide bond. The molecule thus strongly resembles Ig constant domains in polypeptide chain folding and overall tertiary structure. Amino acid residues that are the same in the sequences of 132-microglobulin and Ig constant domains are predominantly in the interior of the molecule, whereas residues conserved among 182-microglobulins from different species are both in the interior and on the molecular surface. In the crystals studied, the molecule is clearly monomeric, consistent with the observation that 132-microglobulin, unlike Ig constant domains, apparently does not form dimers in vivo but associates with the heavy chains of major histocompatibility complex antigens. Our results demonstrate that, at the level of detailed threedimensional structure, the light chain of the major histocompatibility class I antigens belongs to a superfamily of structures related to the Ig constant domains. 032-microglobulin (,82m) was discovered in the urine of patients with chronic kidney dysfunction (1). It has since been found in a variety of physiological fluids as well as on the surfaces of nearly all cells as the light chain of the major histocompatibility complex (MHC) class I antigens of man (HLA) and other vertebrates (2-5). These antigens play central roles in two widely studied activities involving immune recognition by T cells: the rejection of foreign tissue grafts through direct recognition of foreign MHC antigens and the recognition of viral and other antigens in conjunction with self-MHC antigens (6-8). The MHC class I antigens display an extraordinary polymorphism that is the apparent basis for the diversity and specificity of these recognition events. The heavy chains of these antigens are integral membrane proteins, and their polymorphism is confined to their NH2-terminal 180 amino acid residues, those farthest from the cell surface. In contrast, 832m and the 90 extracellular residues closest to the membrane are highly conserved. Although the function of P32m in these antigens is unknown, there is evidence that its presence is necessary for posttranslational processing and insertion of HLA heavy chains into the membrane (9, 10). It also appears to stabilize the structure of the heavy chain in that its removal causes loss of alloantigenic sites on HLA (11, 12). 832ms from different species have similar chemical structures. Approximately 50% of the residues are identical in the five sequences that are known completely (13-17). 832ms from different species apparently can replace one another in the quaternary structure of the MHC class I antigens (18-20), suggesting that the conservation of sequence reflects strong evolutionary pressure to conserve a functionally important conformation. By amino acid sequence homology, f32m belongs to a "superfamily" of proteins related to the Ig constant domains and believed to have evolved from a common ancestor (21). The observation that the chains of Ig contain regions homologous to one another in amino acid sequence led Edelman to suggest that these regions would be folded into distinct compact domains with similar three-dimensional structures (22). Such domains have subsequently been observed in all Igs whose three-dimensional structures have been determined (23). Recently, several other proteins have been shown to have all or part of their amino acid sequences homologous to those of Ig constant domains. These molecules include the T-cell differentiation antigen Thy-1 (24), the a and f3 chains of the T-cell antigen receptor (25-29), the MHC class I and II antigens (7, 30), and 32m (13-17). The similarities in primary structure have led to the suggestion that the homologous portions of these molecules may resemble Ig domains in detailed three-dimensional structure as well. Chemical studies indicate that in the MHC class I antigen complex, f32m associates noncovalently with the region of the heavy chain that is homologous to the Ig constant domains, the 90 extracellular residues adjacent to the membrane (31). In this case, the mode ofassociation as well as the three-dimensional structure may resemble that of Ig constant domains. Here we report the determination, at 2.9 A nominal resolution, of the three-dimensional structure of (32m. We show that the molecule closely resembles the constant domains of Ig with significant differences only in the polypeptide loops connecting the p structures. Most of the amino acid residues that are identical in the aligned sequences of f32m and Ig constant domains have their side chains in the interior of the molecule, consistent with the common peptide folding. There is much more variation in the residues on the molecular surface, consistent with the different functional roles of these molecules. MATERIALS AND METHODS The preparation and crystallization of 832m from bovine milk and colostrum have been reported previously (32, 33). The molecule crystallizes in the orthorhombic space group P212121 with a = 77.27, b = 47.99, and c = 34.42 A. Heavy-atom derivatives were prepared by soaking crystals in crystallization buffer (0.05 M phosphate/0.02% NaN3, pH 7.80) in which heavy-atom reagents had been dissolved. If the reagent was insoluble in this buffer, crystals were transferred to a solution of 0.02 M Tris N03/0.02% NaN3, pH 7.80, for at least 2 hours and then treated with heavy-atom reagents Abbreviations: f32m, 182-microglobulin; MHC, major histocompatibility complex; m.i.r., multiple isomorphous replacement; HLA, the major histocompatibility antigens of man. 4225 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. 4226 Immunology: Becker and Reeke Table 1. Data collection and reduction Concentration, Soaking No. of Total Unique Bdert Pmaxt Name mM time, days crystals observations reflections R* A2 -2 Native 1 56,102 2892 0.0451 0.0288 Hg(OAc)2§ 0.1 8 3 32,478 3049 0.0637 -0.24 0.0256 Pt(NH3)2(NO2)2 1.0 7 3 31,494 2965 0.0463 -5.66 0.0240 Hg4¶ 12 3 40,078 2996 0.0487 1.31 0.0256 *R = I[I(H) T(H)]/ET(H) for the averaging of symmetry-equivalent reflections, where T is the average intensity of reflection H and I is any measurement of that reflection. tBder = Bdenvative Bnative, the difference between the isotropic temperature factors. tPmax = (sin2O/X2) for the highest resolution data used in any calculation. §The Hg(OAc)2 derivative was prepared in Tris NO3 buffer. IThe preparation of the HgIderivative is discussed in the text. dissolved in the latter buffer. Approximately 90 compounds were tested before three usable derivatives, Hg(OAc)2, Pt(NH3)2(NO2)2, and HgIJ2, were found. The HgI2derivative was discovered in soaking experiments using CH2(HgI)2. In subsequent control experiments, it was found that this derivative was not formed when the soaking solution was shielded from room light. A high-resolution difference electron density map of the site of substitution, using multiple isomorphous replacement (m.i.r.) phases based on the other two derivatives, revealed a clearly tetrahedral moiety. This result suggested that the derivatizing substance was HgI2-, formed by photodecomposition of CH2(HgI)2. This suggestion was confirmed by the observation that crystals treated with pure K2HgI4 have projection diffraction patterns and difference maps identical to those treated with CH2(HgI)2 and light. Three-dimensional diffraction data were collected to a maximum resolution of 2.9 A on 20 screenless oscillation photographs. Graphite monochromatized copper radiation from a rotating-anode generator operated at 40 kV, 60 mA was used. Except where noted, all computation was carried out by use of the ROCKS system of crystallographic computer programs (34). During film scanning, crystal slippage was assessed by comparing each photograph with a plot of the diffraction pattern predicted from the crystal parameters measured in alignment and early data photographs. When necessary, new orientation parameters were calculated from data obtained from the data photographs (35). Data reduction and scaling were performed as described (36). The native diffraction data were placed on an absolute scale and an isotropic temperature factor was estimated by using differential Wilson plot procedures, with the parameters of the refined structure of concanavalin A (37) as the reference. Heavy-atom derivative data were processed similarly, using the native protein as reference. Wilson plots of derivative intensity differences (Ider Inat) were used to assess the highest resolution at which derivative data could be considered isomorphous to native. Data collection and reduction are summarized in Table 1. Heavy atoms were located by inspection of difference Patterson and difference electron density maps. Anomalous difference Patterson maps indicated that the Hg(OAc)2 derivative provided useful anomalous dispersion data and these data were included in the m.i.r. phasing and used to establish the absolute hand of the protein. Heavy-atom parameters were refined by the method of Blow and Matthews (38) in which each derivative was refined separately, using m.i.r. phases calculated from the other two derivatives. After this refinement had converged, the parameters were refined in a final series of joint refinements. Anisotropic temperature parameters were included for the me

International Journal of Computational Intelligence and ApplicationsVol. 02, No. 02, pp. 241-244 (2002) Book ReviewNo AccessBOOK REVIEW: "SELF-ORGANIZATION IN BIOLOGICAL SYSTEMS" BY S. CAMAZINE, J. DENEUBOURG, N. R. FRANKS, J. SNEYD, G. THERAULAZ AND E. BONABEAUGeorge N. Reeke, Jr.George N. Reeke, Jr.The Rockefeller University, 1230 York Avenue, New York, NY 10021, USA Search for more papers by this author https://doi.org/10.1142/S1469026802000506Cited by:0 Previous AboutSectionsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack CitationsRecommend to Library ShareShare onFacebookTwitterLinked InRedditEmail Remember to check out the Most Cited Articles! Check out these titles in artificial intelligence! FiguresReferencesRelatedDetails Recommended Vol. 02, No. 02 Metrics History PDF download

An analysis of biological selection aimed at deriving a mechanism-independent definition removes Hull et al.'s obligatory requirement for replication of the carriers of information, under conditions, such as those obtaining in the nervous system, where the information content of a carrier can be modified without duplication by an amount controlled by the outcome of interactions with the environment.

We present a simple method for the realistic description of neurons that is well suited to the development of large-scale neuronal network models where the interactions within and between neural circuits are the object of study rather than the details of dendritic signal propagation in individual cells. Referred to as the composite approach, it combines in a one-compartment model elements of both the leaky integrator cell and the conductance-based formalism of Hodgkin and Huxley (1952). Composite models treat the cell membrane as an equivalent circuit that contains ligand-gated synaptic, voltage-gated, and voltage- and concentration-dependent conductances. The time dependences of these various conductances are assumed to correlate with their spatial locations in the real cell. Thus, when viewed from the soma, ligand-gated synaptic and other dendritically located conductances can be modeled as either single alpha or double exponential functions of time, whereas, with the exception of discharge-related conductances, somatic and proximal dendritic conductances can be well approximated by simple current-voltage relationships. As an example of the composite approach to neuronal modeling we describe a composite model of a cerebellar Purkinje neuron.

Neuronal discharge variability has typically been studied as a function of stimulus rate in the context of a Poisson point process. Here we report that novel discharge behavior is elicited in a model cerebellar Purkinje cell when the timing of afferent impulses follows a pulsed Gaussian stimulation paradigm. We show that under these circumstances effective convergence superseded frequency of activation as the independent variable controlling cell discharge. This provides strong evidence that firing rate alone is an insufficient foundation for the neural code.