Experience replay is associated with efficient nonlocal learning.
Experience replay is associated with efficient nonlocal learning.
复制标题
DOI：
10.1126/science.abf1357
复制
发表时间：
2021-05-21
期刊：
影响因子：
--
通讯作者：
Dolan RJ
中科院分区：
文献类型：
--
作者：
Liu Y;Mattar MG;Behrens TEJ;Daw ND;Dolan RJ
为了做出有效的决策，我们需要考虑行动和结果之间的关系。这些往往被时间和空间隔开。跨越这些差距的机制仍然未知。一种有希望的假设涉及非本地经验的神经重放。使用直接与间接价值学习分离的任务，结合脑磁图，我们研究了神经重放在人类非局部学习中的作用。收到奖励后，我们发现非本地经验有显着的向后重放，状态间有 160 毫秒的时滞，这与行动价值的有效学习有关。对于对未来行为有更大益处的经历，向后重播以及非本地学习的行为证据更加明显。这些发现支持非本地重放作为解决学习过程中复杂学分分配问题的神经机制。
To make effective decisions we need to consider the relationship between actions and outcomes. These are often separated by time and space. The mechanisms spanning these gaps remain unknown. One promising hypothesis involves neural replay of non-local experience. Using a task segregating direct from indirect value learning, combined with magnetoencephalography, we examined the role of neural replay in human non-local learning. Following reward receipt, we found significant backward replay of non-local experience, with a 160 msec state-to-state time lag, this was linked to efficient learning of action values. Backward replay, and also behavioural evidence of non-local learning, was more pronounced for experiences of greater benefit for future behaviour. These findings support non-local replay as a neural mechanism for solving complex credit assignment problems during learning.
登录
查看更多内容
影响因子：
19.9
作者：
Herweg NA;Solomon EA;Kahana MJ
通讯作者：
Kahana MJ
影响因子：
25
作者：
Doll, Bradley B.;Duncan, Katherine D.;Daw, Nathaniel D.
通讯作者：
Daw, Nathaniel D.
影响因子：
64.5
作者：
Kay, Kenneth;Chung, Jason E.;Frank, Loren M.
通讯作者：
Frank, Loren M.
影响因子：
16.2
作者：
Daw ND;Gershman SJ;Seymour B;Dayan P;Dolan RJ
通讯作者：
Dolan RJ
影响因子：
25
作者：
Girardeau, Gabrielle;Benchenane, Karim;Zugaro, Michael B.
通讯作者：
Zugaro, Michael B.