Center for Information and Neural Networks (CiNet), National Institute of Information and Communications Technology, and Osaka University, Yamadaoka 1-4, Suita, Osaka 565-0871, Japan.
Center for Information and Neural Networks (CiNet), National Institute of Information and Communications Technology, and Osaka University, Yamadaoka 1-4, Suita, Osaka 565-0871, Japan; Dynamic Brain Network Laboratory, Graduate School of Frontier Bioscience, Osaka University, Yamadaoka 1-3, Suita, Osaka 565-0871, Japan; Department of Brain Physiology, Graduate School of Medicine, Osaka University, Yamadaoka 1-3, Suita, Osaka 565-0871, Japan.
Curr Biol. 2018 Jul 23;28(14):2250-2262.e3. doi: 10.1016/j.cub.2018.05.056. Epub 2018 Jul 5.
Errors in reaching drive trial-by-trial adaptation to compensate for the error. Parietal association areas are implicated in error coding, but whether the parietal error signals directly drive adaptation remains unknown. We first examined the activity of neurons in areas 5 and 7 while two monkeys performed rapid target reaching to clarify whether and how the parietal error signals drive adaptation in reaching. We introduced random errors using a motor-driven prism device to augment random motor errors in reaching. Neurons in both regions encoded information on the target position prior to reaching and information on the motor error after reaching. However, post-movement microstimulation caused trial-by-trial adaptation to cancel the motor error only when it was delivered to area 5. By contrast, stimulation to area 7 caused trial-by-trial adaptation so that the reaching endpoint was adjusted toward the target position. We further hypothesized that area 7 would encode target error that is caused by a target jump during the reach, and our results support this hypothesis. Area 7 neurons encoded target error information, but area 5 neurons did not encode this information. These results suggest that area 5 provides signals for adapting to motor errors and that area 7 provides signals to adapt to target errors.
在逐次试验中，错误会导致适应以补偿错误。顶叶联合区与错误编码有关，但顶叶的错误信号是否直接驱动适应尚不清楚。我们首先在两只猴子进行快速目标抓取时检查了区域 5 和 7 中的神经元活动，以阐明顶叶的错误信号是否以及如何驱动抓取中的适应。我们使用电机驱动的棱镜装置引入随机错误，以增加抓取中的随机运动错误。在抓取之前，两个区域中的神经元都对目标位置进行了编码，在抓取之后对运动错误进行了编码。然而，只有当刺激施加到区域 5 时，运动后微刺激才会导致逐次试验的适应，从而消除运动错误。相比之下，刺激区域 7 会导致逐次试验的适应，从而使抓取的终点向目标位置调整。我们进一步假设区域 7 将编码在抓取过程中目标跳跃引起的目标错误，我们的结果支持这一假设。区域 7 神经元编码了目标错误信息，但区域 5 神经元没有编码该信息。这些结果表明，区域 5 提供了适应运动错误的信号，而区域 7 提供了适应目标错误的信号。
