Department of Psychology, New York University, New York, New York, USA.
Nat Neurosci. 2011 Sep 11;14(10):1250-2. doi: 10.1038/nn.2904.
Although the human amygdala and striatum have both been implicated in associative learning, only the striatum's contribution has been consistently computationally characterized. Using a reversal learning task, we found that amygdala blood oxygen level-dependent activity tracked associability as estimated by a computational model, and dissociated it from the striatal representation of reinforcement prediction error. These results extend the computational learning approach from striatum to amygdala, demonstrating their complementary roles in aversive learning.
虽然人类的杏仁核和纹状体都与联想学习有关，但只有纹状体的贡献得到了一致的计算描述。使用反转学习任务，我们发现杏仁核的血氧水平依赖活动追踪了计算模型估计的关联性，并将其与纹状体强化预测误差的表示分离。这些结果将计算学习方法从纹状体扩展到杏仁核，证明了它们在厌恶学习中的互补作用。
