三维大小错觉加工的时程机制

Size perception in a 3D scene is known to be affected by distance perception. Two objects with the same physical size appear to have very different sizes if they are perceived to be at different distances. A couple of areas, such as primary visual cortex area (V1), have been suggested to be involved in the experience of the 3D size illusion. However, when our brain start to represent and processing the 3D size illusion remains largely unknown. In the current project, we are planning to use the methods of psychophysics and event-related potentials (ERP) to investigate the time course of brain activity related to the 3D size illusion processing triggered by two types of depth cue (a monocular 3D background and binocular convergence angles). Psychophysically, we measured the magnitude of the 3D size illusion by asking participates to adjust the size of a ball on a uniform background to match the perceived size of a ball (either near or far) in a 3D scene. Electrophysiologically, we acquired EEG data while participates were asked to detect a small dot occurred on the edge of balls with a small probability. In the monocular cue studies, we will manipulate the position and contrast of the ball to construct four experimental conditions (high contrast far, low contrast far, high contrast near and low contrast near condition), which would enable us to isolate the ERP for the balls only by subtracting the ERPs elicited by the low contrast stimuli from those elicited by the corresponding high contrast stimuli. Apart from the traditional ERP analysis, we also planned a dynamic causal modeling (DCM) based on the source reconstruction of the ERP data to explore how connections between V1 and other higher-level visual areas change with the 3D size illusion. Our research can thus enable us to understand the time course of the 3D size processing as well as the dynamic functional connections between V1 and other higher-level visual areas underline the 3D size illusion.

对三维空间中客体的大小知觉，很大程度上受到对该客体空间距离知觉的影响。有着相同尺寸的两个客体，随着对其距离知觉的不同会产生截然不同的大小知觉。尽管有研究表明一些功能区域如初级视觉皮层区域，同三维大小错觉有着密切的联系，但迄今为止，对人类大脑在何时如何表征及加工这种三维大小错觉的问题还知之甚少。本项目通过巧妙的设计排除线索背景不匹配的混淆后，利用心理物理法和高时间分辨率的ERP技术结合不同层次的知觉深度线索（单眼和双眼深度线索）来系统地考察三维大小错觉加工的时间进程机制。并进一步通过基于源重建的模型分析技术，对三维大小错觉加工涉及的脑功能区域以及这些功能区域之间的相互关系，在三维大小错觉加工中的作用进行初步的探讨。研究结果将帮助我们了解大脑对三维大小错觉加工的时间过程（即，大脑什么时候在什么区域开始表征和加工三维大小错觉），并初步建立这一加工过程的认知脑机制模型。

人类在三维空间中对物体大小的知觉通常受限于对物体空间距离深度的判断。两个即使在物理上完全一样的物体，若被知觉为不同的距离，常在知觉大小上被判定为具有很大的差距。过往研究发现大脑的初级视觉皮层区域（V1）同三维大小错觉有密切的联系。然迄今为止，V1究竟在何时参与到三维大小错觉中仍是一个亟待解决的问题。在当前项目中我们利用了高时间分辨率的事件相关电位技术结合心理物理法来试图探究这一问题。行为上，我们通过让被试调节控制球的大小来匹配位于不同空间距离的球的大小。脑电上，我们在被试检测一个在球上随机出现的（20%概率）光点的任务中采集了64通道的脑电信号。重要的是，我们系统地操控了球的对比度（高对比度 vs. 低对比度）和位置（远处 vs. 近处）两个因素，并利用减法逻辑巧妙的去除了背景所诱发的ERP，从而很好的分离出了球刺激诱发的ERP。结果发现，相较位于近处的位置，位于远处的物理上完全一样的球刺激在其呈现后80ms 左右诱发了更大的被定为到初级视觉皮层区域的C1成分。随后的控制实验更是发现，这一C1成分的调节效应随着被试主观大小错觉量的变化而变化，当我们去除了背景信息后，伴随着行为错觉效应的消失，C1的调节效应也消失了。这些结果表明初级视觉皮层早在刺激呈现后月80 ms左右就参与到了对大小错觉的加工中。