The human brain is a complex biological machine able to sense, compute and perceive the external environment and stimuli in real-time. As big as a shoe box, and consuming less than a household light bulb, the human brain features a distributed system based on slow and unreliable components. Yet, it is able to recognize a face or a voice faster than modern supercomputers. Therefore, the real-time and low power cognitive processes of our brain have always been the ultimate ambition to build artificial systems for Edge Information-Extraction and Computing, User-Specific Applications such as Healthcare, Autonomous vehicles, Robotics, and the Internet of Things. Studies on the brain lead to models describing its operating and computational principles, which in turn can be used as a guideline to build new devices, circuits and system emulating brain functionality.

Neuromorphic Very Large Scale Integration (VLSI) circuits model neural networks using a synthetic biology approach whereby they attempt to understand the properties of brain-inspired neural networks by building biologically plausible artifacts that reproduce the same physics of the biological systems they model. Neuromorphic circuits can exhibit very slow, biologically plausible, time constants, facilitating the artificial system / real-world interaction. Despite the slow time-constants, the neuromorphic neural processing chips have extremely fast response times, thanks to a distributed memory, which improves the latency with respect to conventional von Neumann architectures. For these reasons, neuromorphic systems can be developed to carry out sensory data analysis and information extraction, as well as to solve problems in noisy and uncertain settings and constraint satisfactory problems. In addition, these systems are able to learn from experience leading to significant progress in the perceptive abilities of e.g. robots, security, and healthcare systems.

Recently, on the device side, further solutions compatible with and complementary to CMOS technology are being researched to implement artificial synapses or neurons. In this respect, emerging devices such as memristive and spintronic devices have recently attracted a lot of attention due to their excellent performance in terms of high scalability, low latency, and low power operation. Furthermore, they have been reported to locally implement spike-based time- and rate-sensitive operations and to support edge-of-chaos dynamics, which are fundamental computing primitives belonging also to biological neurons and synapses.

The leading frontiers in this Research Topic will support neuromorphic researchers in gaining insight into devices, circuits and systems concepts with the final aim of a neuromorphic implementation of sensor, computing and perception.

In this Research Topic, we wish to provide an overview of the avant-garde artificial biologically plausible sensing, computing and perception paradigms and of the technologies enabling biologically plausible neuromorphic systems. Welcome contributions include (but are not limited to):

1. Sensors for biological signals and external environment stimuli
2. Emerging devices, circuits, and systems enabling neuromorphic paradigms
3. Emerging technologies and devices models to emulate synaptic plasticity and learning
4. Biologically plausible models implementable in neuromorphic sensing, computing and perception systems

Keywords: biologically plausible, neuromorphic circuits and systems, memristive and spintronic devices, sensing and computing, perception

Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.