Spike-Timing-Dependent Plasticity and Short-Term Plasticity Jointly Control the Excitation of Hebbian Plasticity without Weight Constraints in Neural Networks

Figure 6

Distribution of the weights and release probabilities of neurons at Poisson inputs with mean rate 40 Hz. Each subfigure in the figure depicts the distribution of the weight algorithm and the mean release probability at the given synapse of postsynaptic neuron A and postsynaptic neuron B at Poisson inputs with mean firing rate 40 Hz. For example, the leftmost top subfigure shows the variation of the mean release probability and the weight distribution at the first synapse of the ten synapses. As shown in the figure, the network, both neuron A and neuron B, spent around 150 bins to adjust to the external Poisson inputs and subsequently reach the stability. gives the distribution of the weights of the synaptic connections from presynaptic neuron A to postsynaptic neuron B. Similarly, gives the distribution of the weights of the synaptic connections from presynaptic neuron B to postsynaptic neuron A. is the distribution of the mean release probability of transmitters of presynaptic neuron B and is the distribution of mean release probability of transmitters of presynaptic neuron A at postsynaptic connections of neuron B. Moreover, the slopes of the mean release probabilities, and , were determined using linear regression analysis and give the slope of mean release probability of from bin 1 to 150 and bin 150 to 200, respectively; Similarly, and give the mean release probability of from bin 1 to bin 150 and from 150 to 200, respectively.