Representational drift: Emerging theories for continual learning and experimental future directions
Representational drift: Emerging theories for continual learning and experimental future directions
复制标题
DOI：
10.1016/j.conb.2022.102609
复制
发表时间：
2022-08-05
影响因子：
5.7
通讯作者：
Harvey, Christopher D.
中科院分区：
文献类型：
--
作者：
Driscoll, Laura N.;Duncker, Lea;Harvey, Christopher D.
最近的研究表明，与感觉、认知和行动相关的神经活动模式通常并不稳定，而是在几天或几周内经历大规模的变化，这种现象称为表征漂移。在这里，我们强调了最近对漂移的观察，漂移如何不太可能被实验混淆所解释，以及大脑如何可能补偿漂移以允许稳定的计算。我们认为，漂移可能在神经计算中发挥重要作用，以允许持续学习，无论是分离和相关的记忆发生在不同的时间。最后，我们提出了一个展望未来的实验方向，需要进一步表征漂移和测试新兴理论漂移的计算中的作用。
Recent work has revealed that the neural activity patterns correlated with sensation, cognition, and action often are not stable and instead undergo large scale changes over days and weeks-a phenomenon called representational drift. Here, we highlight recent observations of drift, how drift is unlikely to be explained by experimental confounds, and how the brain can likely compensate for drift to allow stable computation. We propose that drift might have important roles in neural computation to allow continual learning, both for separating and relating memories that occur at distinct times. Finally, we present an outlook on future experimental directions that are needed to further characterize drift and to test emerging theories for drift's role in computation.