Fig. 4: Co-dependent inhibitory synaptic plasticity: EI balance without firing rate setpoint.
a, Schematic of the simulations used in c and d. A postsynaptic neuron receives 800 excitatory and 200 inhibitory synapses that undergo plasticity (*). b, Schematic of changes in synaptic weight, Δw, as a function of the postsynaptic neuron’s firing rate for spike-based models with stable setpoints. Top, firing rate setpoint from ESP is higher than the one from ISP. Bottom, firing rate setpoint from ISP is higher than the one from ESP. The interval between the setpoint is defined as Δr. c, Combination of excitatory9 and inhibitory7 spike-based rules. Top, firing rate of a postsynaptic neuron receiving excitatory and inhibitory inputs. Red and blue lines indicate the firing rate setpoints imposed by the excitatory9 and inhibitory7 spike-based learning rules, respectively. The parameters of the learning rules were chosen so that the setpoints coincide during the first and third quarters of the simulation. During the second and fourth quarters of the simulation, the setpoint imposed by the excitatory spike-based learning rule is increased and decreased, respectively. Middle, ratio between excitatory and inhibitory currents. Bottom, average excitatory (red) and inhibitory (blue) synaptic weights of input neurons normalized by their initial value. d, Same as c for the combination of excitatory spike-based9 and co-dependent inhibitory synaptic learning rules. The blue line in the middle panel indicates the balance setpoint imposed by the co-dependent inhibitory synaptic plasticity rule.
