Funamizu Akihiro
Institute for Quantitative Biosciences, University of Tokyo, Laboratory of Neural Computation, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-0032, Japan.
Department of Life Sciences, Graduate School of Arts and Sciences, University of Tokyo, Meguro-ku, Tokyo 153-8902, Japan.
iScience. 2021 Jul 9;24(8):102826. doi: 10.1016/j.isci.2021.102826. eCollection 2021 Aug 20.
In perceptual decision-making, prior knowledge of action outcomes is essential, especially when sensory inputs are insufficient for proper choices. Signal detection theory (SDT) shows that optimal choice bias depends not only on the prior but also the sensory uncertainty; however, it is unclear how animals integrate sensory inputs with various uncertainties and reward expectations to optimize choices. We developed a tone-frequency discrimination task for head-fixed mice in which we randomly presented either a long or short sound stimulus and biased the choice outcomes. The choice was less accurate and more biased toward the large-reward side in short- than in long-stimulus trials. Analysis with SDT found that mice did not use a separate, optimal choice threshold in different sound durations. Instead, mice updated one threshold for short and long stimuli with a simple reinforcement-learning rule. Our task in head-fixed mice helps understanding how the brain integrates sensory inputs and prior.
在感知决策中，对动作结果的先验知识至关重要，尤其是当感官输入不足以做出恰当选择时。信号检测理论（SDT）表明，最优选择偏差不仅取决于先验知识，还取决于感官不确定性；然而，尚不清楚动物如何将具有各种不确定性的感官输入与奖励期望相结合以优化选择。我们为头部固定的小鼠开发了一种音调频率辨别任务，在此任务中我们随机呈现长或短的声音刺激，并使选择结果产生偏差。与长刺激试验相比，在短刺激试验中，选择的准确性更低，且更偏向于大奖励一侧。运用信号检测理论进行分析发现，小鼠在不同声音持续时间内并未使用单独的最优选择阈值。相反，小鼠通过一个简单的强化学习规则更新了针对短刺激和长刺激的一个阈值。我们针对头部固定小鼠的任务有助于理解大脑如何整合感官输入和先验知识。
