Neuromorphic Reservoir Computing with LIF Neurons in Nengo: Study on Energy-Efficient Spiking Architectures

Abstract—Neuromorphic computing models the computational processes of the human brain to enable low-power, eventdriven information processing. This educational research paper
presents a comprehensive study of reservoir computing implemented using Leaky Integrate-and-Fire (LIF) neurons in the
Nengo simulator. We explore the theoretical basis of LIF models,
system design, and implementation, followed by extensive experiments analyzing the effect of key parameters such as reservoir
size, connection probability, input rate, and integration window.
Realistic data tables and energy modeling are provided to
demonstrate trade-offs between accuracy and energy efficiency.
Beyond technical depth, we reflect on educational outcomes,
methodological insights, and broader societal implications. The
study combines theoretical rigor with an accessible format suited
for undergraduate research and transfer portfolios