The Branching Dendrite Secret Behind Human Cognitive Superiority

Key Points:

• Researchers at Hebrew University of Jerusalem have discovered that single human cortical neurons function as highly sophisticated computing devices, with processing power comparable to multi-layered deep artificial neural networks, challenging the traditional view of neurons as simple switches.
• The team developed the “Twin” Imitation Metric, using AI to measure how complex an artificial neural network must be to mimic a single neuron’s input/output functions, revealing that human neurons’ densely branching dendritic trees and unique electrical properties give them a massive computational advantage over other mammals.
• Unlike the previous belief that intelligence primarily depends on brain size and neuron count, this study suggests that the enhanced internal computational depth of individual human neurons plays a crucial role in human cognitive abilities such as language, imagination, and pattern recognition.
• These findings provide a new framework linking neuron structure to computational power, offering insights that could inspire a revolutionary generation of neuromorphic AI systems built from artificial units with intrinsic multi-layered processing capabilities similar to biological neurons.
• Published in PNAS, the research highlights key morpho-electrical differences between human and rodent cortical neurons, such as dendritic branching and synaptic receptor properties, which underlie the superior functional complexity of human neurons and may explain advanced human cognition.