Modality-invariant audio-visual association coding in crow endbrain neurons
Modality-invariant audio-visual association coding in crow endbrain neurons
复制标题
DOI：
10.1016/j.nlm.2016.11.011
复制
发表时间：
2017-01-01
影响因子：
2.7
通讯作者：
Nieder, Andreas
中科院分区：
文献类型：
--
作者：
Moll, Felix W.;Nieder, Andreas
单神经元活动的corvid nidopallium caudolaterale（NCL），假定鸟类功能类似的前额叶皮层，代表协会的听觉与视觉刺激。这对于那些需要依靠视听联系来交流、寻找配偶或逃避捕食者的鸣禽来说具有很高的适应价值。然而，目前还不清楚NCL神经元是否可以代表跨模态协会在模态不变，抽象的方式。为了分离模态依赖性和模态不变的NCL活动，我们训练两只乌鸦在时间记忆延迟期间将听觉样本线索与视觉测试刺激相匹配，反之亦然。在样品介绍，NCL活动选择性地编码协会在一个模态不变的方式。在延迟期间，我们观察到受试者特定的，人群水平的编码偏差的NCL活动。尽管这些偏见，任务相关的信息可以同样很好地解码从任何主题的神经元延迟活动。解码的成功是由许多混合选择性神经元，介导的任务变量的NCL人口水平的高维表示。这些结果平行的哺乳动物PFC的研究结果，建议共同的机制负责跨类群的多模态协会领域的适应性。(C)2016 Elsevier Inc. All rights reserved.
Single neuron activity in the corvid nidopallium caudolaterale (NCL), the supposed avian functional analog of the prefrontal cortex, represents associations of auditory with visual stimuli. This is of high adaptive value for songbirds that need to rely on audio-visual associations to communicate, find a mate or escape predators. However, it remains unclear whether NCL neurons can represent cross-modal associations in a modality invariant, abstract fashion. To dissociate between modality-dependent and modality invariant NCL activity, we trained two crows to match auditory sample cues with visual test stimuli, and vice versa, across a temporal memory delay. During sample presentation, NCL activity selectively encoded associations in a modality invariant fashion. During the delay, we observed subject specific, population-level coding biases in NCL activity. Despite of these biases, task relevant information could be decoded equally well from either subject's neuronal delay activity. Decoding success was facilitated by many mixed selectivity neurons, which mediated high dimensional representations of task variables on the NCL population level. These results parallel findings from the mammalian PFC, suggesting common mechanisms responsible for the adaptability of multimodal association areas across taxa. (C) 2016 Elsevier Inc. All rights reserved.