The shunting inhibition was demonstrated using the unidirectional two-compartment neuron configuration. Of the 256 synapse circuits, 192 were configured to be excitatory (connected via terminal Isyn_exc), 1 as shunting inhibitory (GABAa-type connected via terminal Isyn), and 1 was configured as the leak resistor Rleak (connected via terminal Isyn). The remaining 62 synapse circuits were connected via terminal Isyn, but were not activated in this demonstration. The synaptic efficacies of all of the synapse circuits were set to the maximum. The resting membrane potential of the soma was set at approximately 600 mV. The synaptic reversal potential (Esyn) for the shunting inhibitory synapse circuit was set to 590 mV. For the synapse circuit configured as the resistor Rleak, the value of Eleak was set at 590 mV. These values were set based on the relative difference of general electrophysiological values measured from neuronal cells. The average resting membrane potential in neuronal cells is about −70 mV. In our chip, the maximum and minimum voltage supplies were 1 V and 0 V. Furthermore, the soma circuit is designed utilizing PMOS transistors’ characteristics and its spiking behavior is opposite to the convention (See Section 2.3). Due to this, its resting membrane potential is close to 1 V instead of 0 V. It was set to 600 mV in the unidirectional two-compartment configuration, as this value is ideal for the operation of both synapse and soma circuits. Additionally, shunting inhibitory synapses have a reversal potential close to the resting membrane potential (−70 mV), and on average, this value is slightly higher than the resting membrane potential [36,37]. However, as the polarity of the current in our soma circuit is opposite to that of the conventional direction, the value of the synaptic reversal potential (590 mV) was fixed slightly lower than the resting membrane potential. Upon activation, an excitatory synapse generates an EPSP that is shunted if a shunting inhibitory synapse circuit is simultaneously activated. This is because the shunting inhibitory synapse circuit turns on in the right region of the I–V plot in Figure 9B (Esyn > Vmem), and shunts the EPSP as desired. To demonstrate the shunting inhibition in the circuit experiments, the learning circuitry was deactivated, and the following runs were performed. Initially, only one excitatory synapse circuit was activated by an input spike. In the second run, only the shunting inhibitory synapse circuit was activated, and in the third run, both the excitatory and shunting inhibitory synapse circuits were simultaneously activated. The dendritic membrane potentials for all three cases are plotted in Figure 10A. In the first run, the dendritic membrane potential was strongly depolarized. As expected, in the second run, there was no major change in the dendritic membrane potential. In the third run, the EPSC induced by the excitatory synapse circuit slightly depolarized Vden. As expected, the EPSP was shunted by the shunting inhibitory synapse circuit, i.e., Vden did not depolarize as strongly as in the first case. Each of the three runs above was performed 10 times, and it was observed that shunting inhibition in the second run reduced the amplitude of the EPSP in the first run by an average value of 34.6%, with a standard deviation of 1.65%. Two additional runs were performed using additional excitatory synapse circuits with the same circuit parameters. First, the minimum number of excitatory synapse circuits (four) required to generate a spike were activated synchronously, and next, the same four excitatory synapse circuits were activated along with the single shunting inhibitory synapse circuit. The dendritic (orange and red traces) and somatic (blue and green traces) membrane potentials for both runs are plotted in Figure 10B. As expected, the soma did not spike in the second run, owing to simultaneous activation of the shunting inhibitory synapse. These runs were performed 15 times, and the probability of blocking the soma’s spike by the shunting inhibition was observed as 100%. Synchronous activation of five excitatory synapse circuits was required to overpower the inhibition of a single synapse and for the soma to generate a spike. Instead of using all 192 synapse circuits for the demonstration, we chose fewer circuits, this was done to show visible dendritic depolarization with both single and multiple (four) excitatory synapse circuits. By the appropriate configuration of the circuit’s voltage parameters (that control the amplitude and time constant of excitatory and shunting inhibitory synapse current), the number of synapse circuits can be chosen as desired. The spiking threshold of the soma circuit was approximately 575 mV.