Building large networks of biological neurons

We envision the building of many realistic cortical neurons on a single Integrated Ciruit (IC). This goal requires efficiently utilizing the similarilities between silicon and biological physics. We present recent results building Silicon models of biological channel utilizing the physics connections to MOSFET devices and the gating effect modulating its channel. Using this approach, we present recent results on voltage-clamp measurements for Silicon Sodium and Potassium Channels, biologically realistic action potentials from these channels, models of programmable and learning synapses with biological responses, and active models of dendritic cables. These results, combined with recent advances in reconfigurable analog IC approaches, currently allows the detailed implementation of a pyramidal cell in a few mm
²
in 0.35μm CMOS. Using modern CMOS processes, one could envision 1000s of realistic neurons on an IC with millions of state variables (computing nodes) on a single IC, accelerating the day we can build synthetic systems to represent significant cortical regions.