Synchronization dynamics on the picosecond time scale in coupled Josephson junction neurons

Area of Science:

• Physics
• Neuroscience
• Computer Science

Background:

• Conventional digital computation faces physical limits in speed and energy efficiency.
• Neuromorphic computing aims to emulate brain function for enhanced performance.
• Superconducting circuits offer potential for high-speed, low-power computation.

Purpose of the Study:

• To fabricate and test a neuromorphic circuit using superconducting Josephson junctions.
• To model the behavior of mutually coupled excitatory neurons.
• To investigate synchronization states and their control in a Josephson junction neural circuit.

Main Methods:

• Fabrication of a neuromorphic circuit with Josephson junctions modeling neurons, axons, and synapses.
• Experimental testing of the circuit's response under varying synaptic parameters (delay and strength).
• Analysis of neuronal firing synchronization states (in-phase and antiphase) and phase-flip bifurcations.

Main Results:

• The Josephson junction circuit successfully modeled two mutually coupled excitatory neurons.
• Observed neuronal desynchronization and synchronization in in-phase and antiphase states.
• Demonstrated toggling between states via synaptic parameter alteration, exhibiting phase-flip bifurcations.
• Achieved firing synchronization calculations over 70,000 times faster than conventional digital methods.
• Reported low energy dissipation of 10^-17 J/spike.

Conclusions:

• Josephson junction neurons represent a viable approach for advancing neuronal computation.
• The developed circuit shows promise for applications in high-speed, low-energy neuromorphic computing.
• Proof-of-concept experiments validate the potential of superconducting circuits for brain-inspired computing.