|
|
|
| Study of Specific-Stimulus Information for Transmission of Speech Signals in Cochlea Neural Networks |
| WANG Fengjiao, REN Yuhao, ZHAO Jin, DUAN Fabing
|
| Institute of Complexity Science, Qingdao University, Qingdao 266071, China |
|
|
|
|
Abstract For decoding information contained in the cochlea neural networks responses to speech signals, it is interesting to address which parts of input stimuli are more efficient. In this paper, the stimulus-specific information associated with a particular stimulus will be adopted to study the decrease of average uncertainties, and its calculation method is developed. We use a leaky integrate-and-fire model to capture the responses of cochlea neurons to the input speech signal, and calculate the stimulus-specific information caused by each speech signal part. It is shown that the weighted average of stimulus-specific information over the stimulus ensembles yields the mutual information, and the stimulus-specific information is also useful in clearly indentifying the stimuli that are significantly efficient to the cochlea neural network. Moreover, the stimulus-specific information can not only determine which signal component mainly contributes to the mutual information, but also confirms the availability of internal noise in the neural networks. There is a non-monotonic relationship between the noise intensity and the maximum stimulus-specific information. These results indicate that the applicability of the integrate-and-fire neuron model for current cochlear implant decoding technology deserves to be further investigated.
|
|
Received: 25 December 2014
Published: 22 June 2026
|
|
|
|
|
|
[1] Borst A, Theunissen F E. Information theory and neural coding[J]. Nature Neuroscience, 1999, 2(11): 947-957.
[2] Arcas B A Y, Fairhall A L, Bialek W. What can a single neuron compute[J]. Advances in Neural Information Processing Systems, 2000, 13(1):75-81.
[3] DeWeese M R, Meister M. How to measure the information gained from one symbol[J]. Network: Computation in Neural Systems. Neural Syst, 1999, 10(4): 325-340.
[4] Butts D A. How much information is associated with a particular stimulus?[J]. Network: Computation in Neural Systems, 2003, 14: 177-187.
[5] Butts D A, Goldman M S. Tuning curves, neuronal variability and sensory coding[J]. PLOS Biology, 2006, 4(4): 639-646.
[6] Montgomery N, Wehr M. Auditory cortical neurons convey maximal stimulus-specific information at their best frequency[J]. The Journal of Neuroscience, 2010, 30(40): 13362-13366.
[7] Stocks N G. The application of suprathreshold stochastic resonance to cochlear implant coding[J]. Flucatuation and Noise Letters, 2002, 2(3): 169-181.
[8] Stocks N G. Suprathreshold stochastic resonance in multilevel threshold systems[J]. Physical Review Letters, 2000, 84(11): 2310-2313.
[9] 祁明,许丽艳,季冰,等. 周期性语音信号传输的超阈值随机共振研究[J]. 复杂系统与复杂性科学,2013, 3(10): 31-36.
Qi Ming, Xu Liyan, Ji Bing, et al. Suprathreshould stochastic resonance phenomenon of periodic voice signal transmission[J]. Complex Systems and Complexity Science, 2013, 3(10): 31-36.
[10] Chacron M J, Longtin A, Pakdaman K. Chaotic firing in the sinusoidally forced leaky integrate-and-fire model with threshold fatigue[J]. Physica D: Nonlinear Phenomena, 2004, 192(1/2): 138-160
[11] Barbi M, Chillemi S,Garbo A D. The leaky integrate-and-fire with noise: a useful tool to investigate SR[J]. Chaos, Solitons & Fractals, 2000, 11(12): 1849-1853.
[12] Cover T M, Thomas J A. Elements of Information Theory[M]. New York: Wiley, 1991:13-37.
[13] 杨一威,徐月晋,廖吉昌,等.人工耳蜗的膜电位积分放电刺激方案及其数字信号处理[J].南方医科大学学报,2012, 32 (10):1435-1439.
Yang Yiwei, Xu Yuejin, Miu Jichang, et al. Digital signal processing of a novel neuron discharge model stimulation strategy for cochlear implants[J]. Journal of Southern Medical University, 2012, 32 (10):1435-1439.
[14] Yarrow S, Challis E, Series P. Fisher and Shannon inforamtion in finite neural populations[J]. Neural Computation, 2012, 24(7): 1740-1780. |
|
|
|
