波状多巴胺动力学作为时空信用分配机制，这一成果由美国布朗大学Christopher I. Moore、Michael J. Frank和Arif A. Hamid团队合作取得。 2021年4月15日，国际学术期刊《细胞》在线发表了这一成果。

研究人员揭示了多巴胺轴突中波状多巴胺的时空活动模式，其在背侧纹状体中释放。这些波在激活基序和多巴胺瞬变组织相关纹状体亚区域内的局部簇之间切换。值得注意的是，波浪轨迹是依据任务需求而特定产生的，当奖励取决于动物行为时，它会从背体纹状体向背侧纹状体传播；而当奖励与行为响应无关时，则向相反方向传播。

研究人员研发了一种计算体系结构，在该体系结构中，通过对事件的推断来预测纹状体多巴胺波，并提供一种将信用分配与特化纹状体次区域直接匹配的机制。追求奖励时的多巴胺活性表明了仪器控制的程度，并可以通过与奖励波相互作用以预测对未来行为的调整，这支持了该模型的预测。

据悉，大量证据表明多巴胺通过编码奖励预测错误来影响学习。然而，尚不清楚纹状体是否受控于空间功能特化的多巴胺动力学。

附：英文原文

Title: Wave-like dopamine dynamics as a mechanism for spatiotemporal credit assignment

Author: Arif A. Hamid, Michael J. Frank, Christopher I. Moore

Issue&Volume: 2021-04-15

Abstract: Significant evidence supports the view that dopamine shapes learning by encoding rewardprediction errors. However, it is unknown whether striatal targets receive tailoreddopamine dynamics based on regional functional specialization. Here, we report wave-likespatiotemporal activity patterns in dopamine axons and release across the dorsal striatum.These waves switch between activational motifs and organize dopamine transients intolocalized clusters within functionally related striatal subregions. Notably, wavetrajectories were tailored to task demands, propagating from dorsomedial to dorsolateralstriatum when rewards are contingent on animal behavior and in the opponent directionwhen rewards are independent of behavioral responses. We propose a computational architecturein which striatal dopamine waves are sculpted by inference about agency and providea mechanism to direct credit assignment to specialized striatal subregions. Supportingmodel predictions, dorsomedial dopamine activity during reward-pursuit signaled theextent of instrumental control and interacted with reward waves to predict futurebehavioral adjustments.

DOI: 10.1016/j.cell.2021.03.046

Source: https://www.cell.com/cell/fulltext/S0092-8674(21)00377-9