Sinz F.,University of Tubingen |
Bethge M.,University of Tubingen |
Bethge M.,Bernstein Center for Computational Neuroscience
PLoS Computational Biology | Year: 2013
Divisive normalization in primary visual cortex has been linked to adaptation to natural image statistics in accordance to Barlow's redundancy reduction hypothesis. Using recent advances in natural image modeling, we show that the previously studied static model of divisive normalization is rather inefficient in reducing local contrast correlations, but that a simple temporal contrast adaptation mechanism of the half-saturation constant can substantially increase its efficiency. Our findings reveal the experimentally observed temporal dynamics of divisive normalization to be critical for redundancy reduction. © 2013 Sinz, Bethge.
Shandilya S.G.,Max Planck Institute for Dynamics and Self-Organization |
Shandilya S.G.,Yale University |
Timme M.,Max Planck Institute for Dynamics and Self-Organization |
Timme M.,Bernstein Center for Computational Neuroscience |
Timme M.,University of Gottingen
New Journal of Physics | Year: 2011
Inferring the network topology from dynamical observations is a fundamental problem pervading research on complex systems. Here, we present a simple, direct method for inferring the structural connection topology of a network, given an observation of one collective dynamical trajectory. The general theoretical framework is applicable to arbitrary network dynamical systems described by ordinary differential equations. No interference (external driving) is required and the type of dynamics is hardly restricted in any way. In particular, the observed dynamics may be arbitrarily complex; stationary, invariant or transient; synchronous or asynchronous and chaotic or periodic. Presupposing a knowledge of the functional form of the dynamical units and of the coupling functions between them, we present an analytical solution to the inverse problem of finding the network topology from observing a time series of state variables only. Robust reconstruction is achieved in any sufficiently long generic observation of the system. We extend our method to simultaneously reconstructing both the entire network topology and all parameters appearing linear in the system's equations of motion. Reconstruction of network topology and system parameters is viable even in the presence of external noise that distorts the original dynamics substantially. The method provides a conceptually new step towards reconstructing a variety of real-world networks, including gene and protein interaction networks and neuronal circuits. © IOP Publishing Ltd and Deutsche Physikalische Gesellschaft.
Pamir E.,Bernstein Center for Computational Neuroscience
Learning & memory (Cold Spring Harbor, N.Y.) | Year: 2011
Conditioned behavior as observed during classical conditioning in a group of identically treated animals provides insights into the physiological process of learning and memory formation. However, several studies in vertebrates found a remarkable difference between the group-average behavioral performance and the behavioral characteristics of individual animals. Here, we analyzed a large number of data (1640 animals) on olfactory conditioning in the honeybee (Apis mellifera). The data acquired during absolute and differential classical conditioning differed with respect to the number of conditioning trials, the conditioned odors, the intertrial intervals, and the time of retention tests. We further investigated data in which animals were tested for spontaneous recovery from extinction. In all data sets we found that the gradually increasing group-average learning curve did not adequately represent the behavior of individual animals. Individual behavior was characterized by a rapid and stable acquisition of the conditioned response (CR), as well as by a rapid and stable cessation of the CR following unrewarded stimuli. In addition, we present and evaluate different model hypotheses on how honeybees form associations during classical conditioning by implementing a gradual learning process on the one hand and an all-or-none learning process on the other hand. In summary, our findings advise that individual behavior should be recognized as a meaningful predictor for the internal state of a honeybee--irrespective of the group-average behavioral performance.
Battaglia D.,Max Planck Institute for Dynamics and Self-Organization |
Battaglia D.,Bernstein Center for Computational Neuroscience |
Hansel D.,University of Paris Descartes |
Hansel D.,Hebrew University of Jerusalem
PLoS Computational Biology | Year: 2011
Visually induced neuronal activity in V1 displays a marked gamma-band component which is modulated by stimulus properties. It has been argued that synchronized oscillations contribute to these gamma-band activity. However, analysis of Local Field Potentials (LFPs) across different experiments reveals considerable diversity in the degree of oscillatory behavior of this induced activity. Contrast-dependent power enhancements can indeed occur over a broad band in the gamma frequency range and spectral peaks may not arise at all. Furthermore, even when oscillations are observed, they undergo temporal decorrelation over very few cycles. This is not easily accounted for in previous network modeling of gamma oscillations. We argue here that interactions between cortical layers can be responsible for this fast decorrelation. We study a model of a V1 hypercolumn, embedding a simplified description of the multi-layered structure of the cortex. When the stimulus contrast is low, the induced activity is only weakly synchronous and the network resonates transiently without developing collective oscillations. When the contrast is high, on the other hand, the induced activity undergoes synchronous oscillations with an irregular spatiotemporal structure expressing a synchronous chaotic state. As a consequence the population activity undergoes fast temporal decorrelation, with concomitant rapid damping of the oscillations in LFPs autocorrelograms and peak broadening in LFPs power spectra. We show that the strength of the inter-layer coupling crucially affects this spatiotemporal structure. We predict that layer VI inactivation should induce global changes in the spectral properties of induced LFPs, reflecting their slower temporal decorrelation in the absence of inter-layer feedback. Finally, we argue that the mechanism underlying the emergence of synchronous chaos in our model is in fact very general. It stems from the fact that gamma oscillations induced by local delayed inhibition tend to develop chaos when coupled by sufficiently strong excitation. © 2011 Battaglia and Hansel.
Clemens J.,Princeton University |
Kramer S.,Humboldt University of Berlin |
Ronacher B.,Bernstein Center for Computational Neuroscience
Proceedings of the National Academy of Sciences of the United States of America | Year: 2014
Decision-making processes, like all traits of an organism, are shaped by evolution; they thus carry a signature of the selection pressures associated with choice behaviors. The way sexual communication signals are integrated during courtship likely reflects the costs and benefits associated with mate choice. Here, we study the evaluation of male song by females during acoustic courtship in grasshoppers. Using playback experiments and computational modeling we find that information of different valence (attractive vs. nonattractive) is weighted asymmetrically: while information associated with nonattractive features has large weight, attractive features add little to the decision to mate. Accordingly, nonattractive features effectively veto female responses. Because attractive features have so little weight, the model suggests that female responses are frequently driven by integration noise. Asymmetrical weighting of negative and positive information may reflect the fitness costs associated with mating with a nonattractive over an attractive singer, which are also highly asymmetrical. In addition, nonattractive cues tend to be more salient and therefore more reliable. Hence, information provided by them should be weighted more heavily. Our findings suggest that characterizing the integration of sensory information during a natural behavior has the potential to provide valuable insights into the selective pressures shaping decision-making during evolution.