Systematic Literature Review of Neuromorphic Computing: A PRISMA-Guided Analysis

Abstract

Neuromorphic computing, inspired by the architecture and functionality of the human brain, has emerged as a transformative paradigm in computational intelligence, enabling low-power, high-efficiency, and real-time data processing for next-generation artificial intelligence (AI) systems. This study presents a comprehensive Systematic Literature Review (SLR) of neuromorphic computing using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) framework to identify, evaluate, and synthesize existing research from 2020 to 2025. Following a rigorous selection and quality assessment process, 62 peer-reviewed publications were analysed, covering key developments in neuromorphic hardware, algorithms, sensors, and applications across domains such as robotics, edge computing, and neuroscience. The review reveals a growing shift toward event-driven processing, brain-inspired learning algorithms, and hybrid analog-digital systems. Key challenges identified include the lack of standardized benchmarks, limitations in hardware-software co-design, and complexity of integration with traditional systems. This SLR highlights future research directions in scalable neuromorphic architectures, energy-efficient learning models, and cross-disciplinary applications. By mapping current progress and gaps, this study aims to provide a foundational reference for researchers, developers, and policymakers in shaping the evolution of neuromorphic computing technologies.

Keywords: neuromorphic, computing, algorithms, brain, complexity