Table 2 Behavioral measures obtained with HA, SD, RD regimes.

From: FNS allows efficient event-driven spiking neural network simulations based on a neuron model supporting spike latency

\(Network \,\, configuration\)

\(Neuron \,\, model\)

\(max_{PSTH} , dt=0.01\)

\(c_{hi-PSTH} , dt=0.01\)

\(FRd_{0.01 \rightarrow 0.1}\)

HA (F.R.\(\approx 10\;\text{Hz}\))

AEIF

66.99 (1.76)

0 (0)

n.r.

IAF

65.34 (2.66)

0 (0)

n.r.

IAF_ps

63.32 (0.72)

0 (0)

n.r.

LIFL

68.11 (1.94)

0 (0)

–

SD (F.R.\(\approx 1 \;\text{Hz}\))

AEIF

160.49 (95.15)

0.88 (0.99)

n.r.

IAF

243.31 (70.4)

6.00 (2.56)

\(13.53\%\)

IAF_ps

97.25(81.98)

0.25 (0.7)

n.r.

LIFL

171.07 (114)

1.12 (0.81)

–

RD (F.R.\(\approx 50 \;\text{Hz}\))

AEIF

2925.5 (61.21)

48.85 (0.73)

n.r.

IAF

2956.48 (27.3)

49.4 (0.48)

n.r.

IAF_ps

2803.65 (11.255)

49.8 (0.51)

n.r.

LIFL

3020 (25.50)

50 (0.41)

–

  1. Both mean and standard deviation are reported for \(max_{PSTH}\) and \(c_{hi-PSTH}\). The index FRd was not computed for the LIFL model since this simulation did not involve time steps; for the other models, only values \(\ge 2\) have been considered relevant and then reported (n.r. instead).
Back to article page