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: Eighteenth Annual Computational Neuroscience Meeting: CNS*2009

Open Access Open Badges Poster presentation

Emergent pitch perception using short term plasticity

Tjeerd olde Scheper

Author Affiliations

Department of Integrative Neurophysiology, Center for Neurogenomics and Cognitive Research, VU University, 1081 HV, Amsterdam, The Netherlands

Department of Computing, Oxford Brookes University, Oxford, OX33 1HX, UK

BMC Neuroscience 2009, 10(Suppl 1):P197  doi:10.1186/1471-2202-10-S1-P197

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

Published:13 July 2009

© 2009 Scheper; licensee BioMed Central Ltd.


In pitch perception, the perceived frequency is the summation of the simple and complex sounds that form the complex harmonics. The phenomenon of the perceived missing fundamental frequency can be described as the frequency that is then perceived but is not actually present in the sound [1]. The mechanism in the cochlear nucleus that seems to be responsible is still unclear, however using simple phenomenologcial models we show that short-term plasticity is capable of solving this problem.

Short-term depression

It has been proposed that short-term depression may be involved in synaptic filtering and adaptation [2]. Using a modified Hindmarsh-Rose model [3], we demonstrate that a simple network of two mutually inhibitory neurons with short term depression, which are periodically driven with different frequencies, show the missing fundamental frequency as an emergent property. To achieve this, a novel post-synaptic potential model has been devised based on simplified Bessel equations. Short-term depression is then included using a mechanism described by Tsodyks et al. [4,5].

Emergent dynamic adaptation

The emerging frequency in the two-neuron model is the summation of the two input frequencies based on the principle of the least common multiple of the periods. The model can be extended to the least common multiple of three or more input frequencies without any apparent constraints, allowing the readily summation of many periodic signals. This demonstrates that dynamic adaptation made possible by the short-term depression is very effective at filtering and signal addition. The described mechanism is relatively insensitive to noise and wholly insensitive to phase shifts. Introducing arbitrary differences in phase in the input periods has no effect on the correctly emerging frequency. The summation of the input frequencies also allows other interesting phenomena such as the multiplexing of completely different signals at alternating timeframes.


  1. Balenzuela P, Garcia-Ojalvo J: Neural mechanism for binaural pitch perception via ghost stochastic resonance.

    Chaos 2005, 15:23903. PubMed Abstract | Publisher Full Text OpenURL

  2. Abbott LF, Regehr WG: Synaptic computation.

    Nature 2004, 431:796-803. PubMed Abstract | Publisher Full Text OpenURL

  3. Pinto RD, Varona P, Volkovskii AR, Szucs A, Abarbanel HDI, Rabinovich MI: Synchronous behavior of two coupled electronic neurons.

    Physical Review E 2000, 62:2644-2656. Publisher Full Text OpenURL

  4. Tsodyks M, Pawelzik K, Markram H: Neural Networks with Dynamic Synapses.

    Neural Computation 1998, 10:821-835. PubMed Abstract | Publisher Full Text OpenURL

  5. Tsodyks M, Uziel A, Markram H: Synchrony generation in recurrent networks with frequency-dependent synapses.

    J Neuroscience 2000, 20:1-5. OpenURL