A synaptic learning rule for exploiting nonlinear dendritic computation.

A synaptic learning rule for exploiting nonlinear dendritic computation.
复制标题

利用非线性树状运算的突触学习法则。

DOI:
10.1016/j.neuron.2021.09.044
复制
发表时间:
2021-12-15
期刊:
影响因子:
16.2
通讯作者:
Häusser M
Häusser M
中科院分区:
医学1区
文献类型:
--
作者:
Bicknell BA;Häusser M

文献摘要

参考文献

被引文献

相似文献

大脑中的信息处理取决于分布在神经元树突中的突触输入的整合。树突整合是一个分层过程,被认为相当于多层网络的整合,可能赋予单个神经元大量的计算能力。然而,神经元是否能够学会利用树突特性来实现这种潜力尚不清楚。在这里,我们从树突电缆理论中开发了一种学习规则,并用它来研究详细的锥体神经元模型的处理能力。我们表明,可以学习使用突触输入模式的空间或时间特征的计算,甚至可以协同组合来解决典型的非线性特征绑定问题。学习规则的电压依赖性驱动协同突触参与树突非线性,而尖峰时间依赖性塑造阈下电位的时间进程。因此,树突输入输出关系可以通过突触可塑性灵活调整,从而允许单个神经元优化实现非线性函数。来自电缆理论的学习规则用于生物物理模拟 锥体细胞 I/O 功能可以通过突触可塑性进行计算优化 主动和被动树突机制增强输入模式辨别能力 单个神经元只需通过调整突触权重即可学习网络级计算 Bicknell 和 Häusser 开发了一种用于研究单神经元计算和学习的理论方法。通过推导可塑性规则,优化调整相互作用突触的强度以控制体细胞尖峰,他们表明神经元可以学习利用树突的生物物理特性来执行非线性计算。
Information processing in the brain depends on the integration of synaptic input distributed throughout neuronal dendrites. Dendritic integration is a hierarchical process, proposed to be equivalent to integration by a multilayer network, potentially endowing single neurons with substantial computational power. However, whether neurons can learn to harness dendritic properties to realize this potential is unknown. Here, we develop a learning rule from dendritic cable theory and use it to investigate the processing capacity of a detailed pyramidal neuron model. We show that computations using spatial or temporal features of synaptic input patterns can be learned, and even synergistically combined, to solve a canonical nonlinear feature-binding problem. The voltage dependence of the learning rule drives coactive synapses to engage dendritic nonlinearities, whereas spike-timing dependence shapes the time course of subthreshold potentials. Dendritic input-output relationships can therefore be flexibly tuned through synaptic plasticity, allowing optimal implementation of nonlinear functions by single neurons. A learning rule derived from cable theory is used in biophysical simulations Pyramidal cell I/O functions can be optimized for computation by synaptic plasticity Active and passive dendritic mechanisms enhance input pattern discrimination Single neurons can learn network-level computations simply by tuning synaptic weights Bicknell and Häusser develop a theoretical approach for investigating single-neuron computation and learning. By deriving a plasticity rule that optimally adjusts the strengths of interacting synapses to control somatic spiking, they show that neurons can learn to harness the biophysical properties of their dendrites to perform nonlinear computations.
DOI: 10.1038/s41586-020-2649-2
发表时间: 2020-09
期刊: Nature
影响因子: 64.8
作者:
Harris CR;Millman KJ;van der Walt SJ;Gommers R;Virtanen P;Cournapeau D;Wieser E;Taylor J;Berg S;Smith NJ;Kern R;Picus M;Hoyer S;van Kerkwijk MH;Brett M;Haldane A;Del Río JF;Wiebe M;Peterson P;Gérard-Marchant P;Sheppard K;Reddy T;Weckesser W;Abbasi H;Gohlke C;Oliphant TE
通讯作者: Oliphant TE
DOI: 10.1038/s41467-017-00740-z
发表时间: 2017-09-26
影响因子: 16.6
作者:
Bono J;Clopath C
通讯作者: Clopath C
DOI: 10.1126/science.1189664
发表时间: 2010-09-24
期刊: Science (New York, N.Y.)
影响因子: --
作者:
Branco T;Clark BA;Häusser M
通讯作者: Häusser M
DOI: 10.1016/j.neuron.2012.05.015
发表时间: 2012-07-26
期刊: NEURON
影响因子: 16.2
作者:
Gidon, Albert;Segev, Idan
通讯作者: Segev, Idan
DOI: 10.1038/71125
发表时间: 2000-01-01
影响因子: 25
作者:
Archie, KA;Mel, BW
通讯作者: Mel, BW