Baston Chiara, Ursino Mauro
Annu Int Conf IEEE Eng Med Biol Soc. 2015;2015:6505-8. doi: 10.1109/EMBC.2015.7319883.
Basal Ganglia (BG) are implied in many motor and cognitive tasks, such as action selection, and have a central role in many pathologies, primarily Parkinson Disease. In the present work, we use a recently developed biologically inspired BG model to analyze how the dopamine (DA) level can affect the temporal response during action selection, and the capacity to learn new actions following rewards and punishments. The model incorporates the 3 main pathways (direct, indirect and hyperdirect) working in BG functioning. The behavior of 2 alternative networks (the first with normal DA levels, the second with reduced DA) is analyzed both in untrained conditions, and during training performed in different epochs. The results show that reduced DA causes delayed temporal responses in the untrained network, and difficult of learning during training, characterized by the necessity of much more epochs. The results provide interesting hints to understand the behavior of healthy and dopamine depleted subjects, such as parkinsonian patients.
基底神经节（BG）参与许多运动和认知任务，如动作选择，并且在许多疾病（主要是帕金森病）中起着核心作用。在本研究中，我们使用最近开发的受生物启发的BG模型来分析多巴胺（DA）水平如何影响动作选择过程中的时间响应，以及在奖励和惩罚后学习新动作的能力。该模型纳入了在BG功能中起作用的3条主要通路（直接、间接和超直接）。在未训练条件下以及在不同时期进行训练期间，分析了2个替代网络（第一个具有正常DA水平，第二个具有降低的DA）的行为。结果表明，DA水平降低会导致未训练网络中的时间响应延迟，以及训练期间学习困难，其特征是需要更多的训练时期。这些结果为理解健康和多巴胺缺乏的受试者（如帕金森病患者）的行为提供了有趣的线索。
