Visual Integration and Memory Process

众所周知，我们的眼睛每秒大约移动三到四次。从扫视抑制的事实来看，可以说我们得到了3到4张关于外部世界略有不同方面的静止图片。即使在这样一个动态的情况下，我们也能感知到稳定的世界，这一事实告诉我们，我们的认知系统以某种方式整合了跨眼跳的视觉信息。我们的工作假设是：有一个视觉缓冲区，输入的视觉输入被整合在一起，在我们对世界的知识的帮助下，构建一个稳定的视觉世界。我们准备了一套设备，在计算机显示器上呈现连续的视觉刺激，同时记录眼睛的位置。在一个初步研究中，一条水平线段a被呈现在注视点的左侧，持续时间较短。眼球扫视后，视线移到前视点a左侧的另一个视点，在a所在的同一区域出现垂直线段b。观察者的任务是用数字估计第二段b在a上的哪一点看起来交叉。结果发现，在光线较暗的房间条件下，这种估计是基于节段的物理位置而不是视网膜位置。必须考察周围视觉环境的影响。相关实验表明，对两个相对移动的物体中的一个进行注视可以增强对非注视物体运动的感知。

It is known that we move our eyes about three or four times per second. From the fact of saccadic suppression, it can be said we are getting 3 or 4 still pictures of somewhat different aspeccts of the outer world. The fact that we perceive stable world even in such a dynamic situation tells that our cognitive system somehow integrate visual information across saccades. Our working hypothesis is this: There is a visual buffer in which incoming visual inputs are integrated to construct a stable visual world being aided by our knowledge of the world.A set of equipments was prepared to present successive visual stimuli on a computer display and record eye positions simultaneously.In a pilot study, a horizontal line segment a was presented to the left of a fixation point for a short duration. After a saccade which moved eyes to another fixation point which located at the left of the former segment a, a vertical line segment b appeared at the same area where a was located. Observer's task was to estimate numerically at which point on a the second segment b appeared crossed. It was found the estimation was based on physical instead of retinotopic positions of the segmen ts in a light room condition. The effect of surrounding visual context must be examined.A related experiment revealed that fixation on one of the two objects moving relatively to each other enhanced perceiving of movement of the nonfixated object.