fig3

Neurochemical memory diversification towards liquid reinforcement learning

Figure 3. PPF behaviors of (A1) the pristine devices, and those after the treatment of H2S, and (A2) the pristine devices, and those after the treatment of H2O2; The inset shows the PPF indices for different pulse intervals (1 s, 15.17 mW·cm-2 light pulses). SIDP emulation of the device after the treatment of (B1) H2S and (B2) H2O2 with light intensities ranging from 2.013 mW·cm-2 to 82.38 mW·cm-2 (1 s light pulses); SDDP emulation of the device after treatment of (C1) H2S and (C2) H2O2 with pulse durations ranging from 0.5 s to 8 s (15.17 mW·cm-2 light pulses); SNDP emulation of the device after treatment of (D1) H2S and (D2) H2O2 with 2 to 10 pulses (15.17 mW·cm-2 light pulses); SRDP emulation of the device after treatment of (E1) H2S and (E2) H2O2 with light pulse intervals ranging from 10 s to 2 s (1 s, 15.17 mW·cm-2 light pulses); (F1) H2S-mediated synaptic plasticity; (F2) H2O2-mediated synaptic plasticity (1 s, 15.17 mW·cm-2 light pulses). ΔPSC: Postsynaptic current response; PPF: paired-pulse facilitation; H2S: hydrogen sulfide; H2O2: hydrogen peroxide; SIDP: spike-intensity-dependent plasticity; SDDP: spike-duration-dependent plasticity; SNDP: spike-number-dependent plasticity; SRDP: spike-rate-dependent plasticity.