希伯来大学的研究团队发现，单个人类皮层神经元拥有远比之前认识到的更强大的计算能力[1]。这项研究由教授Idan Segev和Mickey London领导，相关成果已发表在《美国国家科学院院刊》2026年第123卷第28期[1]。研究人员通过创建新的测量方法，结合先进计算机建模和人工智能技术，确定了单个脑细胞的计算复杂性[1]。
长期以来，科学家将神经元视为简单的开关，但这一发现颠覆了这一认识。Segev教授表示："人们经常认为神经元是一个简单的开关，开或关。我们展示的是单个人类神经元本身是一个非凡复杂的计算设备。"[1]研究表明，人类皮层神经元的计算能力可与深度神经网络相媲美[1]。这意味着每个神经元都能执行复杂的信息处理任务，而非仅是简单的开关。这一发现提示，人脑的高级认知能力可能并非仅源于神经元数量和连接方式，而是由单个神经元本身的复杂性驱动[1]。
Researchers at the Hebrew University have discovered that individual human cortical neurons possess significantly greater computational capabilities than previously understood, functioning as sophisticated biological computing devices rather than simple on-off switches.[1] The findings, published in the Proceedings of the National Academy of Sciences in 2026 (Volume 123, Issue 28), reveal that the advanced cognitive abilities of the human brain may stem from the inherent complexity of individual neurons themselves, rather than solely from their quantity and interconnections.[1]
The research team, led by professors Idan Segev and Mickey London, employed novel measurement techniques combined with advanced computational modeling and artificial intelligence to assess the computational complexity of individual brain cells.[1] According to Segev, "People often think of neurons as a simple switch, on or off. What we have shown is that a single human neuron itself is a remarkably complex computing device."[1] The computational power of human cortical neurons is comparable to that of deep neural networks, suggesting that individual neurons can perform intricate information-processing tasks previously attributed to larger neural circuits.[1]