Artificial Neural Networks Illuminate Peripersonal Neuron Functions

Artificial neural networks (ANNs) have recently shown promise in elucidating the functions of peripersonal neurons, which play a critical role in how humans and other primates perceive and interact with their immediate surroundings. Research conducted by a collaborative team from the Chinese Academy of Sciences and the Italian Institute of Technology (IIT), among others, has developed a novel theoretical framework for understanding these neurons, as documented in a study published in the prestigious journal Nature Neuroscience on June 18, 2025.

Peripersonal space, defined as the area immediately surrounding the body, is essential for understanding interactions between individuals and their environments. The study's lead author, Dr. Giandomenico Iannetti, a senior researcher at the Italian Institute of Technology, elaborated on the significance of this area, stating, "Our journey into this field began truly serendipitously, during unfunded experiments done purely out of curiosity." This exploration led to discoveries about the hand-blink reflex, which is inherently linked to the positioning of one’s hand relative to their eye, further hinting at the complex neural mechanisms underpinning spatial awareness.

The researchers utilized ANNs trained through reinforcement learning to simulate how peripersonal neurons assess potential actions based on the expected rewards or punishments. Rory John Bufacchi, the first author of the study and a researcher at the IIT, explained the methodology: "In simple terms, we built computer simulations of simplified 'animals' that learned through trial and error how to choose actions based on the rewards those actions would yield over time." This innovative approach enabled the team to develop what they termed an "egocentric value map," which provides a predictive model of the peripersonal space.

Their findings indicated that these artificial neurons created receptive fields centered around various body parts, mimicking the behavior of biological peripersonal neurons. This modular structure allows for rapid adaptation to environmental changes, a feature that is vital for both biological organisms and artificial intelligence (AI) agents. The results also revealed that these receptive fields expanded in response to faster-moving stimuli and higher-value objects, reinforcing the notion that peripersonal neurons are intricately linked to the value of potential actions in one’s immediate vicinity.

The implications of this research extend beyond theoretical neuroscience. The framework developed by Iannetti, Bufacchi, and their colleagues holds potential applications in fields such as neuroprosthetics, robotics, and human–robot interaction. Dr. Iannetti noted, "These findings could help robots simulate egocentric value maps, allowing for more adaptive and context-specific representations of human interaction distances, thus improving collaboration between humans and robots."

Moving forward, the research team plans to refine their model by incorporating parameters for sensory uncertainty and cognitive modeling of the environment. "We aim to collaborate across labs to model richer, more fine-grained neuronal data," Bufacchi stated, emphasizing the commitment to advancing this field of study.

This groundbreaking research not only enhances our understanding of the neural underpinnings of spatial awareness in primates but also paves the way for innovative applications in technology and robotics. As the study progresses, it will be crucial to monitor how these findings might transform our interactions with AI and robotic systems, ultimately leading to more intuitive and effective human-machine interfaces.