摘要
We take an adaptive leaky integrate-and-fire neuron model to explore the effect of non-Poisson neurotransmitter on stochastic resonance and its signal-to-noise ratio(SNR)gain.Event triggered algorithm is adopted to speed up the simulating process.It is revealed that both the output SNR and the SNR gain can be monotonically improved when increasing the shape parameter for Gamma distribution.Particularly,for large signal coupling strength,the 1:1 stochastic phase locking induced by Gamma noise is responsible for the frequency matching stochastic resonance,and the output signal-to-noise ratio can surpass the input signal-to-noise ratio,which is significantly different with Poisson case,while for extremely weak signal coupling strength,the SNR gain peak,which is far larger than unity,is due to noise induced resonance.The observations are meaningful in understanding the neural processing mechanisms from a more realistic viewpoint of synaptic modeling.
我们采用自适应漏电积分-放电模型来研究非泊松递质对随机共振及其信噪比增益的影响,并运用事件驱动算法加速模拟过程、研究结果表明,输出信噪比和信噪比增益都会随着伽马分布的形状参数的增加而增加.特别地,当输入信号幅值较大时,由Gamma噪声诱导的1:1随机锁像揭示出此时发生的是满足频率匹配关系的随机共振现象,并且输出信噪比可以超过输入信噪比,这与Poisson情形显著不同;而当输入信号幅值极弱时,信噪比增益会远远大于1,这是由于发生了噪声诱导的随机共振现象.这些观察结果对于从更现实的突触建模角度理解神经信息处理机制是有意义的.
作者
Yanmei Kang
Yuxuan Fu
Yaqian Chen
康艳梅;付宇轩;陈亚倩(Department of Applied Mathematics,School of Mathematics and Statistics,Xi'an Jiaotong University,Xi'an 710049,China)
基金
the Non Poisson Modeling of Neuron Synaptic Input and Critical Dynamics for Cortical Networks(Grant No.11772241).
