Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Efficient auditory coding


The auditory neural code must serve a wide range of auditory tasks that require great sensitivity in time and frequency and be effective over the diverse array of sounds present in natural acoustic environments. It has been suggested1,2,3,4,5 that sensory systems might have evolved highly efficient coding strategies to maximize the information conveyed to the brain while minimizing the required energy and neural resources. Here we show that, for natural sounds, the complete acoustic waveform can be represented efficiently with a nonlinear model based on a population spike code. In this model, idealized spikes encode the precise temporal positions and magnitudes of underlying acoustic features. We find that when the features are optimized for coding either natural sounds or speech, they show striking similarities to time-domain cochlear filter estimates, have a frequency-bandwidth dependence similar to that of auditory nerve fibres, and yield significantly greater coding efficiency than conventional signal representations. These results indicate that the auditory code might approach an information theoretic optimum and that the acoustic structure of speech might be adapted to the coding capacity of the mammalian auditory system.

This is a preview of subscription content, access via your institution

Relevant articles

Open Access articles citing this article.

Access options

Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Figure 1: Representing a natural sound with the use of spikes. A brief segment of the word ‘canteen’ (input) is represented as a spike code (top).
Figure 2: Efficient codes for natural sounds predict revcor filter shapes and population characteristics.
Figure 3: Human speech is adapted to the mammalian cochlear code.
Figure 4: Fidelity–rate curves for Fourier, wavelet and spike codes.


  1. Barlow, H. B. in Sensory Communication (ed. Rosenbluth, W. A.) 217–234 (MIT Press, Cambridge, Massachusetts, 1961)

    Google Scholar 

  2. Atick, J. J. Could information-theory provide an ecological theory of sensory processing. Network 3, 213–251 (1992)

    Article  Google Scholar 

  3. Simoncelli, E. & Olshausen, B. Natural image statistics and neural representation. Annu. Rev. Neurosci. 24, 1193–1216 (2001)

    CAS  Article  Google Scholar 

  4. Laughlin, S. B. & Sejnowski, T. J. Communication in neuronal networks. Science 301, 1870–1874 (2003)

    ADS  CAS  Article  Google Scholar 

  5. Lewicki, M. S. Efficient coding of natural sounds. Nature Neurosci. 5, 356–363 (2002)

    CAS  Article  Google Scholar 

  6. Shamma, S., Chadwick, R., Wilbur, W. J., Morrish, K. A. & Rinzel, J. A biophysical model of cochlear processing: Intensity dependence of pure tone responses. J. Acoust. Soc. Am. 78, 1612–1621 (1986)

    Article  Google Scholar 

  7. Yang, X., Wang, K. & Shamma, S. Auditory representations of acoustic signals. IEEE Trans. Inf. Theory 38, 824–839 (1992)

    Article  Google Scholar 

  8. Lewicki, M. S. & Sejnowski, T. J. in Advances in Neural Information Processing Systems (eds Kearns, M. J., Solla, S. A. & Cohn, D. A.) vol. 11, 730–736 (MIT Press, Cambridge, Massachusetts, 1999)

    Google Scholar 

  9. Lewicki, M. S. in Probabilistic Models of the Brain: Perception and Neural Function (eds Rao, R. P. N., Olshausen, B. A. & Lewicki, M. S.) 241–255 (MIT Press, Cambridge, Massachusetts, 2002)

    Google Scholar 

  10. Smith, E. C. & Lewicki, M. S. Efficient coding of time-relative structure using spikes. Neural Comput. 17, 19–45 (2005)

    Article  Google Scholar 

  11. Davis, G., Mallat, S. & Avellaneda, M. Adaptive greedy approximations. Construct. Approx. 13, 57–98 (1997)

    MathSciNet  Article  Google Scholar 

  12. Mallat, S. G. & Zhang, Z. Matching pursuits with time–frequency dictionaries. IEEE Trans. Signal Process. 41, 3397–3415 (1993)

    ADS  Article  Google Scholar 

  13. de Ruyter van Steveninck, R. & Bialek, W. Realtime performance of a movement sensitive neuron in the blowfly visual system: Coding and information transfer in short spike sequences. Proc. R. Soc. Lond. B 234, 379–414 (1988)

    ADS  Article  Google Scholar 

  14. Olshausen, B. A. in Probabilistic Models of the Brain: Perception and Neural Function (eds Rao, R. P. N., Olshausen, B. A. & Lewicki, M. S.) 257–272 (MIT Press, Cambridge, Massachusetts, 2002)

    Google Scholar 

  15. Emmons, L. H., Whitney, B. M. & Ross, D. L. Sounds of the Neotropical Rainforest Mammals [audio CD] (Library of Natural Sounds, Cornell Laboratory of Ornithology, Ithaca, New York, 1997).

  16. deBoer, E. & deJongh, H. On cochlear encoding: Potentialities and limitations of the reverse correlation technique. J. Acoust. Soc. Am. 63, 115–135 (1978)

    ADS  CAS  Article  Google Scholar 

  17. Carney, L. H. Sensitivities of cells in the anteroventral cochlear nucleus of cat to spatiotemporal discharge patterns across primary afferents. J. Neurophysiol. 64, 437–456 (1990)

    CAS  Article  Google Scholar 

  18. Recio-Spinoso, A., Temchin, A. N., van Dijk, P., Fan, Y.-H. & Ruggero, M. A. Wiener-kernel analysis of responses to noise of chinchilla auditory-nerve fibers. J. Neurophysiol. 93, 3635–3648 (2005)

    Article  Google Scholar 

  19. Irino, T. & Patterson, R. A level-dependent auditory filter: the gammachirp. J. Acoust. Soc. Am. 101, 764–774 (1997)

    Article  Google Scholar 

  20. Carney, L. H., McDuffy, M. J. & Shekhter, I. Frequency glides in the impulse responses of auditory-nerve fibers. J. Acoust. Soc. Am. 105, 2384–2391 (1999)

    ADS  CAS  Article  Google Scholar 

Download references


E.C.S. was supported by a National Institutes of Health training grant. This material is based on work supported by a National Science Foundation grant to M.S.L. Empirical data in this paper were acquired from Boston University's Earlab, an online, freely accessible auditory database ( Author Contributions M.S.L. and E.C.S. developed the model, analysed the results and wrote the paper together; E.C.S. designed and ran the simulations.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Michael S. Lewicki.

Ethics declarations

Competing interests

Reprints and permissions information is available at The authors declare no competing financial interests.

Supplementary information

Supplementary Discussion

This file contains additional discussion on the non-linear nature of the spike coding algorithm and comparison of learned kernels and auditory revcor filters.

Supplementary Methods

This file contains additional details of the methods used in this study.

Supplementary Figure 1

Alternate efficient codes emerge when training on other sound ensembles.

Supplementary Figure 2

A quantitative comparison of the correlation between cat revcor filters and different codes.

Supplementary Figure 3

Characterization of spike code kernel functions by reverse correlation.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Smith, E., Lewicki, M. Efficient auditory coding. Nature 439, 978–982 (2006).

Download citation

  • Received:

  • Accepted:

  • Issue Date:

  • DOI:

This article is cited by


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing