Chernyshev Boris V, Pronko Platon K, Stroganova Tatiana A
Laboratory of Cognitive Psychophysiology,National Research University Higher School of Economics,Moscow,Russia.
MEG Center,Moscow State University of Psychology and Education,Moscow,Russia.
Vis Neurosci. 2016 Jan;33:E007. doi: 10.1017/S0952523816000031.
Detection of illusory contours (ICs) such as Kanizsa figures is known to depend primarily upon the lateral occipital complex. Yet there is no universal agreement on the role of the primary visual cortex in this process; some existing evidence hints that an early stage of the visual response in V1 may involve relative suppression to Kanizsa figures compared with controls. Iso-oriented luminance borders, which are responsible for Kanizsa illusion, may evoke surround suppression in V1 and adjacent areas leading to the reduction in the initial response to Kanizsa figures. We attempted to test the existence, as well as to find localization and timing of the early suppression effect produced by Kanizsa figures in adult nonclinical human participants. We used two sizes of visual stimuli (4.5 and 9.0°) in order to probe the effect at two different levels of eccentricity; the stimuli were presented centrally in passive viewing conditions. We recorded magnetoencephalogram, which is more sensitive than electroencephalogram to activity originating from V1 and V2 areas. We restricted our analysis to the medial occipital area and the occipital pole, and to a 40-120 ms time window after the stimulus onset. By applying threshold-free cluster enhancement technique in combination with permutation statistics, we were able to detect the inverted IC effect-a relative suppression of the response to the Kanizsa figures compared with the control stimuli. The current finding is highly compatible with the explanation involving surround suppression evoked by iso-oriented collinear borders. The effect may be related to the principle of sparse coding, according to which V1 suppresses representations of inner parts of collinear assemblies as being informationally redundant. Such a mechanism is likely to be an important preliminary step preceding object contour detection.
诸如卡尼兹三角之类的虚幻轮廓（ICs）的检测主要依赖于枕外侧复合体，这是已知的。然而，关于初级视觉皮层在此过程中的作用尚无普遍共识；一些现有证据表明，与对照组相比，V1区视觉反应的早期阶段可能对卡尼兹三角有相对抑制作用。负责卡尼兹错觉的同向性亮度边界可能会在V1区和相邻区域引起周边抑制，从而导致对卡尼兹三角的初始反应减少。我们试图测试在成年非临床人类参与者中，卡尼兹三角所产生的早期抑制效应的存在情况，以及其定位和时间。我们使用了两种大小的视觉刺激（4.5°和9.0°），以便在两个不同的离心率水平上探究这种效应；刺激在被动观看条件下呈现在中央。我们记录了脑磁图，它比脑电图对源自V1区和V2区的活动更敏感。我们将分析限制在枕叶内侧区域和枕极，以及刺激开始后的40 - 120毫秒时间窗口内。通过应用无阈值聚类增强技术并结合置换统计，我们能够检测到反向IC效应——与对照刺激相比，对卡尼兹三角的反应相对抑制。当前的发现与涉及同向性共线边界引起的周边抑制的解释高度相符。这种效应可能与稀疏编码原则有关，根据该原则，V1区抑制共线集合内部部分的表征，因为它们在信息上是冗余的。这样一种机制很可能是物体轮廓检测之前的一个重要初步步骤。
