Fig. 1

(a) A population of \(N=3\) EIF neurons receiving common \(\xi_{c}\) and independent inputs \(\xi_{i}\). The voltages of the neurons evolve according to Eq. (1). Parameters: \(\tau _{m} = 5\mbox{ ms}\), \(\varDelta_{T} = 3\mbox{ mV}\), \(V_{T} = 20\mbox{ mV}\), \(V_{S} = -53\mbox{ mV}\), \(V_{R} = -60\mbox{ mV}\), \(\tau_{\mathrm{ref}} = 3\mbox{ ms}\). We tune the noise amplitude so that when the DC component of the input is \(\gamma= -60\mbox{ mV}\), the neurons fire at 10 Hz; this yields \(\sigma= 6.23\mbox{ mV}\). The resulting firing is strongly irregular, with the coefficient of variation of the ISI distribution being 0.91. (b) Cartoon of the binning process: spikes recorded from each of the EIF neurons in a bin contribute towards the population spike count. More than one spike occurring from the same neuron within a single bin is treated as a single event. This happens less than 0.4 % of the time in our numerical simulations with \(\mu=0.1\) and \(\rho=0.1\) (input parameters \(\gamma=-60\mbox{ mV}\), \(\sigma =6.23\mbox{ mV}\), \(\lambda=0.30\))