中文摘要：

该研究选取大学生为被试，采用最小变化法，分别测量了干扰线对数、干扰线与主线之间的问隔、主线之间的间隔三个变量导致Zollner错觉产生的绝对阈限。每个变量中主线与水平面的夹角分为45°和90°两种条件。结果发现，若要产生Zollner错觉：1）干扰线对数的阈限分别为5对、7对；2）干扰线与主线之间的间隔阈限分别为6．58像素、4．48像素，3）主线之间的间隔阈限分别为124．44像素、131．73像素。此外还发现：4）在Zollner错觉形成的影响因素中，从大到小依次是：干扰线对数、主线之间的间隔、干扰线与主线之间的间隔；5）45°情况下比90°情况下更容易产生错觉；6）在Zollner错觉知觉中，期望误差严重。

英文摘要：

It is quite natural that we erroneously perceive what we see and we can hardly correct the image even if we have noticed that we are wrong, which is how visual illusions occur. Since last century, researchers have already studied the relationship between figure images and human behavior, the perceptional processing that affects visual illusion, and the neuron system that determines it. Ironical- ly, few of them considered the trigger condition or elements required when we want illusions to occur, and few of them focused on the fa- mous Zollner illusion even if they discovered how the optical features influence the amount of illusion. So what we are curious about is the condition that determines it. In other words, we are to measure the absolute threshold of the Zollner illusion. In terms of the meth- od, since our vision system is trying to modify the image in our eyes in order to be more accurate, the methods used in the past have se- rious problems because participants had to adjust to the stimuli figure for a quite long time. Therefore, the minimal - change method is much better for threshold measuring where participants should respond in a fixed short period of time. In Experiment 1, the variable is the quantity of interfering lines. In Experiment 2, the variable is the gap between the main line and the interfering lines. In Experiment 3, the variable is the distance between the two main lines. Among every three figures, two are signal （the true Zrllner illusion whose main lines are parallel） and one is noise （the false Zollner illusion whose main lines seem paral- lel but actually are unparallel）. Participants need to answer whether it looks parallel or not according to their very first sensation. By comparing the three experiments, we found that figures with 45 degrees orientation was more falsely perceived, which shows the same results as the past findings and fMRI support. In addition, anticipation error causes significant differences between increasing and decreasing blocks in all experiments.