Population coding

  • We study population coding models to gain insight into the computational properties of neural ensembles and to understand how interactions between neurons affect the neural representation of the visual world.
    For example, interactions with a certain structure, so-called limited range noise correlations, can impair the representational accuracy of a neural population. We have shown that this is only the case if the tuning functions of all neurons are homogenous, i.e. identical copies of each other. In contrast, such noise correlations may even improve the coding properties of a code in a heterogenous population (Ecker et al. 2011). Our theoretical work on noise correlations is informed by experimental studies on the structure of noise correlations in awake animals (Ecker et al. 2010), where we have shown that the average level of noise correlations is much lower than previously thought.
    In addition, we try to improve our understanding of the tools used to study neural population coding. We have shown that Fisher information, one of the most widely used tools for measuring the quality of a code, has severe short-comings in particular when applied to coding in short time windows (Berens et al. 2011).
    population coding
    A, Fisher information as a function of population size in a homogeneous population of neurons (black line, independent population; colored lines, correlated populations; see legend in C). B, Same as in A but for heterogeneous population of neurons (κ = 0.25). C, Fisher information relative to independent population for a homogeneous population. D, Same as in C but for heterogeneous population of neurons. Reproduced from Ecker et al. (2011).
  • We aim towards bringing population coding modeling closer together with data analysis. To this end, we have organized a Research Topic in Frontiers in Computational Neuroscience to invite leading researchers in the field to explore possible connections. A total of 15 papers provide an up-to-date overview over the state-of-the art techniques to link population coding and statistical data analysis.

    Selected References

    M. Bethge
    Efficient Population Coding
    Encyclopedia of Computational Neuroscience, Springer New York, 2014
    #population coding
    URL, DOI, PDF, BibTex

    L. A. Gatys, A. S. Ecker, T. Tchumatchenko, and M. Bethge
    Synaptic unreliability facilitates information transmission in balanced cortical populations
    Physical Review E, 91(6), 62707, 2015
    #synaptic noise, #balanced state, #neural population coding
    Code, URL, DOI, PDF, BibTex

    A. S. Ecker, G. H. Denfield, M. Bethge, and A. S. Tolias
    On the Structure of Neuronal Population Activity under Fluctuations in Attentional State
    Journal of Neuroscience, 36(5), 1775-1789, 2016
    #attention, #gain modulation, #noise correlations, #population coding
    Code, URL, DOI, PDF, BibTex

    S. Gerwinn, J. Macke, and M. Bethge
    Bayesian inference for generalized linear models for spiking neurons
    Frontiers in Computational Neuroscience, 4, 2010
    #bayesian inference, #generalized linear model, #spiking neurons
    Code, URL, DOI, PDF, BibTex

    J. H. Macke, S. Gerwinn, L. White, M. Kaschube, and M. Bethge
    Gaussian process methods for estimating cortical maps
    NeuroImage, 56(2), 570-581, 2010
    #gaussian process
    Code, URL, DOI, PDF, BibTex

    A. S. Ecker, P. Berens, A. S. Tolias, and M. Bethge
    The effect of noise correlations in populations of diversely tuned neurons
    The Journal of Neuroscience, 31(40), 14272-14283, 2011
    #noise correlations, #population coding, #fisher information, #orientation
    Code, URL, DOI, PDF, BibTex

    A. S. Ecker, P. Berens, G. A. Keliris, M. Bethge, N. K. Logothetis, and A. S. Tolias
    Decorrelated Neuronal Firing in Cortical Microcircuits
    Science, 327(5965), 584-587, 2010
    #noise correlations, #multi-tetrode recordings, #v1, #macaque
    Code, URL, DOI, PDF, Dataset, BibTex

    P. Berens, A. S. Ecker, S. Gerwinn, A. S. Tolias, and M. Bethge
    Reassessing optimal neural population codes with neurometric functions
    Proceedings of the National Academy of Sciences of the United States of America, 108(11), 4423-4428, 2011
    #fisher information, #population coding, #mean squared error, #discrimination error, #neurometric function
    Code, URL, PDF, BibTex
    University of Tuebingen BCCN CIN MPI