Uncovering the nature of object representations in the human dorsal visual cortex

人类的视觉系统使我们能够识别物体(例如，这是一个杯子)，并以惊人的轻松和准确的方式对它们进行操作(例如，抓住杯子)。几十年来，人们普遍认为，这些不同的功能依赖于不同的皮质通路。具体地说，腹侧通路有助于物体识别，而背侧通路有助于视觉引导动作。然而，最近的证据挑战了这种二元区分，并表明背部“动作”路径中的区域也衍生出在物体感知中发挥作用的形状表征。尽管最近取得了这些进展，但背侧通路所产生的形状表征的性质仍然不清楚。阐明这些表征的一个主要挑战是它们受到各种因素的调节，例如(A)来自其他大脑区域的输入，(B)任务要求和(C)刺激的物理特征。因此，这项研究的首要目标是通过思考不同因素对这些表征的调节程度来阐明背部形状表征的本质。在两个互补的项目中，我们将通过利用各种最先进的行为和神经成像技术来解决这些问题。第一个项目将检验背侧通路是否计算出独立的形状表征，即使腹侧通路的输入最小化。为此，我们将利用明确设计的实验技术，将形状加工偏向背侧通路。主要的预测是，尽管两条通路之间存在相互联系，但背侧通路获得了独立的形状表征。第二个项目将探讨任务需求对背部形状表征的调节程度。为此，将采用需要处理不同形状属性的各种任务，而视觉输入将保持不变。我们预测，尽管背侧通路对形状信息自动敏感，但背侧表征将受到任务需求的调节，特别是对涉及空间加工的任务。总而言之，这项研究计划有可能阐明形状感知的神经基础，形状感知是一种基本的知觉能力。此外，这些洞察力可能有助于揭示先天性和获得性疾病的潜在机制，在这些疾病中，视觉感知受到损害。

The human visual system enables us to recognize objects (e.g., this is a cup) and act upon them (e.g., grasp the cup) with astonishing ease and accuracy. For decades, it was widely accepted that these different functions rely on separated cortical pathways. In particular, the ventral pathway subserves object recognition, while the dorsal pathway subserves visually guided actions. However, recent evidence challenges this binary distinction and suggests that regions in the dorsal "action" pathway also derive shape representations that play a role in object perception. Despite this recent progress, the nature of shape representations derived by the dorsal pathway is still unclear. One major challenge in elucidating these representations is that they are modulated by a variety of factors, such as (a) the inputs from other brain regions, (b) task demands and (c) the physical features of the stimulus. Thus, the overarching goal of this research program is to shed light on the nature of dorsal shape representations by contemplating the extent to which different factors modulate these representations. In two complementary projects, we will address these questions, by utilizing a variety of state-of-the-art behavioural and neuroimaging techniques. The first project will examine whether independent shape representations are computed by the dorsal pathway, even when inputs from the ventral pathway are minimized. To this end, we will utilize experimental techniques which are designed explicitly to bias shape processing to the dorsal pathway. The main prediction is that despite the interconnections between the two pathways, the dorsal pathway derives independent shape representations. The second project will probe the degree to which task demands modulate dorsal shape representations. To that purpose, a variety of tasks, which require the processing of different shape attributes, will be employed, while the visual input will be kept constant. We predict that despite an automatic sensitivity of the dorsal pathway to shape information, dorsal representations will be modulated by task demands, particularly for tasks that involve spatial processing. Together, this research program has the potential to elucidate the neural underpinnings of shape perception, a fundamental perceptual ability. Moreover, these insights might shed light on the underlying mechanisms of congenital and acquired disorders in which visual perception is impaired.