Email updates

Keep up to date with the latest news and content from BMC Neuroscience and BioMed Central.

This article is part of the supplement: Sixteenth Annual Computational Neuroscience Meeting: CNS*2007

Open Access Poster presentation

Bayesian binning for maximising information rate of rapid serial presentation for sensory neurons

Dominik Endres* and Peter Földiák

Author Affiliations

School of Psychology, University of St. Andrews, Fife, KY16 9JP, UK

For all author emails, please log on.

BMC Neuroscience 2007, 8(Suppl 2):P151  doi:10.1186/1471-2202-8-S2-P151


The electronic version of this article is the complete one and can be found online at:


Published:6 July 2007

© 2007 Endres and Földiák; licensee BioMed Central Ltd.

Poster presentation

Understanding the response properties of single neurons is seriously limited by the available experimental time and the rate [bit/s] at which information can be gained from the neurons. A substantial improvement in the latter can be achieved by speeding up the presentation of stimuli.

We show how the novel technique of Bayesian Binning [1] can be used to find the optimal stimulus presentation rate of a continuous sequence of stimuli.

This method applied to neurons in high-level visual cortical area STSa gives optimal presentation rates of approximately 56 ms/stimulus (18 stimuli/s) which is significantly faster than conventional presentation rates, allowing a better sampling of stimulus space. We relate these results to findings obtained with the Bayesian Bin Classification method [2,3], which can be used to select the optimal time window for the analysis of the continuous response stream. Both methods will soon be freely available as standalone command-line applications or Matlab/Octave plugins.

The optimal window duration is equal to the stimulus duration near the best presentation rate. Interestingly, this duration also corresponds to the peak of spike efficiency [bit/spike] of a rate code whose firing rates match those found in visual neurons (area STSa).

References

  1. Endres D, Földiák P: Bayesian bin distribution inference and mutual information.

    IEEE Trans Inform Theory 2005, 51(11):3766-3779. Publisher Full Text OpenURL

  2. Endres D: Bayesian and information-theoretic tools for neuroscience. [http://hdl.handle.net/10023/162] webcite

    PhD thesis. University of St. Andrews, digital research repository; 2007. OpenURL

  3. Endres D, Földiák P: Exact Bayesian bin classification: a fast alternative to Bayesian classification and its application to neural response analysis.

    2007, in press.