Figure 1. Additive STDP learning window.

(A) Examples of two different synapses between the pre-synaptic neuron, 1, and the post-synaptic neuron, 2. The left one has a short dendritic delay and the right one has similar axonal and dendritic delays. (B) Examples of spike pairs. Top: Spike times are given at the somas of each of the neurons in each case in B. Middle: Pre- and post-synaptic spike times for the synapse with a short dendritic delay in B. Bottom: Pre- and post-synaptic spike times for the synapse with similar axonal and dendritic delays in B. (C) The learning window, , used in this study that describes how the change in synaptic weight depends upon the difference in time between the pre- and post-synaptic spikes at the synapse. The form of this is described in Equation (4) with parameter values shown in Table 1. This window was used in an additive STDP learning rule along with two rate-based terms as described in Equation (3). The changes in synaptic strength due to the synaptic spike pairs (shown in B) for each of these two cases is shown by the red and green vertical lines. This shows that as the dendritic delay is increased, or the axonal delay decreased, the for the spike pairs is shifted to the left on the learning window (the opposite occurs for increasing the axonal delay, or decreasing the dendritic delay).