Dynamics of Population Decoding with Strong Inhibition

  • Thomas Trappenberg
Conference paper


Decoding information from a population of noisy neurons can be achieved efficiently with center-surround recurrent networks. Here we study such networks with continuing external input and investigate the dynamics of decoding with varying inhibition strength in the network. We find that the best decoding is achieved at the onset of the memory regime in such networks.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Zemel, R., Dayan, P., Pouget, A.: Probabilistic interpretation of population code. Neural Computation 10 (1998) 403–430.PubMedCrossRefGoogle Scholar
  2. 2.
    Pouget, A., Dayan, P., Zemel, R.: Information processing with population codes. Nature Reviews Neuroscience 1 (2000) 125–132.PubMedCrossRefGoogle Scholar
  3. 3.
    Pouget, A., Zhang, K., Deneve, S., Latham, P.: Statistically efficient estimation using population coding. Neural Computation 10 (1998) 373–401.PubMedCrossRefGoogle Scholar
  4. 4.
    Wu, S., Amari, S., Nakahara, H.: Population coding and decoding in a neural field: a computational study. Neural Computation 14 (2002) 999–1026.PubMedCrossRefGoogle Scholar
  5. 5.
    Wilson, H., Cowan, J.: A mathematical theory of the functional dynamics of cortical and thalamic nervous tissue. Kybernetik 13 (1973) 55–80.PubMedCrossRefGoogle Scholar
  6. 6.
    Grossberg, S.: Contour enhancement, short-term memory, and constancies in reverberating neural networks. Studies in Applied Mathematics 52 (1973) 217–257.Google Scholar
  7. 7.
    Amari, S.: Dynamics of pattern formation in lateral-inhibition type neural fields. Biological Cybernetics 27 (1977) 77–87.PubMedCrossRefGoogle Scholar
  8. 8.
    Deneve, S., Pouget, A., Latham, P.: Divisive normalization, line attractor networks and ideal observers. Advances in Neural Information Processing Systems 11 (1999) 104–110.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Thomas Trappenberg
    • 1
  1. 1.Dalhousie UniversityHalifaxCanada

Personalised recommendations