Neural Mechanisms of Selective Attention

描述（由申请人提供）：在拥挤的视觉场景中，需要注意力来选择性地处理来自相关刺激的信息，并过滤掉不相关的干扰。因此，对人类和动物的研究表明，当注意力被引导到它们的感受野时，几种不同大脑结构中的神经元表现出增强的反应。然而，这些无处不在的注意效应实际上是如何通过这些结构之间的相互作用产生的，特别是当注意线索来自认知因素和任务要求时，仍然很少有人了解。为了设计有效的神经假体或治疗注意力障碍患者，我们需要在系统层面上更好地理解注意力。前额叶皮层、后顶叶皮层和枕被认为是向调节注意力的视觉皮层提供反馈的三个关键结构。在计划的实验中，我们将比较这三个结构中的每一个在视觉皮层V4区提供反馈的作用。V4区在视觉信息沿着腹侧流传递的过程中起着核心作用，腹侧流是物体识别的基础，我们之前已经证明，V4神经元的反应受到注意力的调节，V4区的损伤会导致注意力障碍。我们最近还发现，前额叶皮层的神经元具有足够短的注意力延迟，足以介导V4神经元反应的一些注意力效应，并且前额叶神经元与V4细胞具有一致的活动。这种相干活动在很宽的频率范围内时移约10 ms，这可能是允许功能相互作用的关键时间。在目标1中，我们将把来自后顶叶皮层的记录添加到前额叶记录中，以比较这两个结构在调节V4活动中的作用。枕为来自前额叶和顶叶皮质的信号影响V4提供了另一种解剖学途径。因此，在目标2中，我们将同时记录枕和V4区，以测试枕神经元的特性是否与这种反馈作用一致。然而，神经生理学记录只能提供活动相关性的证据，而不是因果关系。因此，在目标3中，我们将补充前额叶和顶叶皮层和枕的活动抑制的神经记录，以测试因果假说的作用，从每个结构的反馈调制V4反应。对于这些实验，我们将使用我们和我们的合作者最近开发的技术，使用质子泵Arch-T来抑制神经活动，该质子泵由慢病毒递送。有了Arch-T，我们将能够在关键时间点的几十毫秒内抑制活动，以获得对V4反馈的新机制见解。总的来说，我们希望这些研究能给我们提供迄今为止最好的解释，说明多个大脑结构之间的相互作用如何导致有效的视觉处理。

DESCRIPTION (provided by applicant): In crowded visual scenes, attention is needed to selectively process information from relevant stimuli and to filter out irrelevant distracters. Accordingly, studies in humans and animals have shown that neurons in several different brain structures show enhanced responses when attention is directed into their receptive fields. However, there is still little understanding of how these ubiquitous attentional effects are actually generated through the interactions among these structures, particularly when the attentional cues arise from cognitive factors and task demands. To design an effective neural prosthesis or to treat people with attentional disorders, we need a better understanding of attention at the systems level. Three key structures thought to provide feedback to visual cortex that mediates attentional effects are the prefrontal cortex, the posterior parietal cortex and the pulvinar. In the planned experiments, we will compare the roles of each of these three structures in providing feedback to area V4 in visual cortex. Area V4 plays a central role in the relay of visual information along the ventral stream that underlies object recognition, and we have previously shown that V4 neuronal responses are modulated by attention and that damage to V4 causes attentional impairments. We have also recently shown that neurons in prefrontal cortex have attentional latencies that are short enough to mediate some of the attentional effects on V4 neuronal responses, and that prefrontal neurons have coherent activity with cells in V4. This coherent activity is time-shifted across a wide range of frequencies by about 10 ms, which may be the critical time to allow for functional interactions. In Aim 1, we will add recordings from the posterior parietal cortex to the prefrontal recordings, to compare the roles of these two structures in modulating activity in V4. The pulvinar provides an alternative anatomical route for signals from prefrontal and parietal cortex to influence V4. Therefore, in Aim 2, we will record simultaneously in the pulvinar and area V4, to test whether pulvinar neuronal properties are consistent with this feedback role. However, neurophysiological recordings alone can only provide evidence for correlations in activity, not for causality. Therefore, in Aim 3, we will supplement the neural recordings with suppression of activity in prefrontal and parietal cortex and pulvinar, to test causal hypotheses about the role of feedback from each structure in modulating V4 responses. For these experiments, we will use techniques that we and our collaborators have recently developed for the optogenetic suppression of neural activity using the proton pump, Arch-T, which is delivered by lentivirus. With Arch-T we will be able to suppress activity with resolution in the tens of milliseconds, at critical time points, to gain new mechanistic insights into the feedback to V4. In total, we expect these studies to give us the best account so far of how the interactions among multiple brain structures leads to effective visual processing with attention.