Action-suppression in the basal ganglia: mediated by a neural implementation of the Race-model?

在这个项目提案中，我的目的是了解基底神经节在决策中的作用。特别是，我研究的神经算法的抑制不必要的行动。实验研究检查动作抑制，例如在简单的任务中，如果发生某种感觉刺激，动物必须阻止或取消特定的动作。已经成功地用所谓的Race模型来模拟干扰抑制。这项建议的主要研究目标是测试的假设，在基底神经节的动作抑制是由种族模型的神经实现介导的。为了验证这一假设，我将创建一个新的计算模型的基底神经节，实现种族模型的行动抑制，这使我们能够确定的基本机制的决策方面的整合行动促进和行动抑制信号在基底神经节。此外，使用新模型，我计划研究预防学习，即学习在响应不再得到奖励时停止响应。特别是，我假设在基底神经节的间接途径是一个至关重要的神经回路的条件反射的灭绝。我想在Berke博士（密歇根大学）的实验室执行这些项目，因为我可以访问现有的和新获得的实验数据集来设计和约束模型。此外，在与实验室的密切互动中，我可以通过分析电生理数据来测试计算模型产生的预测，我还将根据模型结果提出新的实验。该项目的结果与了解正常和病理行为有关，例如赌博，抽动秽语综合征和帕金森病。

In this project proposal, I aim at understanding the role of the basal ganglia in decision-making. In particular, I study the neural algorithm underlying the suppression of unwanted actions. Experimental studies examine action-suppression e.g. in simple tasks where the animal has to prevent or cancel specific actions if a certain sensory stimulus occurs. Action-suppression has been successfully modeled with the so-called Race-model. The main research objective of this proposal is to test the hypothesis that action-suppression in the basal ganglia is mediated by a neural implementation of the Race-model. To test this hypothesis, I will create a novel computational model of the basal ganglia that implements the Race-model for action-suppression, which allows us to identify fundamental mechanisms of decision-making in terms of the integration of action-facilitating and action-suppressing signals in the basal ganglia. Furthermore, using the new model, I plan to study extinction-learning, i.e. learning to stop responding when the response is no longer rewarded. In particular, I hypothesize that the indirect pathway in the basal ganglia is a crucial neural circuit for the extinction of a conditioned response. I would like to perform these projects in the lab of Dr. Berke (University of Michigan) because I will have access to existing and newly acquired experimental data sets to design and to constrain the models. Furthermore, in close interaction with the lab, I can test predictions arising from the computational models through analysis of the electrophysiological data, and I will also suggest new experiments based on the model results. The results of this project are relevant to understand both normal and pathological behavior, e.g. in gambling, Tourette syndrome, and Parkinson Disease.