Asch S E, Witkin H A
Swarthmore College.
J Exp Psychol Gen. 1992 Dec;121(4):407-18; discussion 404-6.
To overcome certain disadvantages of the tilted-mirror situation as a technique for separating visual and postural determinants of the perceived upright, a new procedure was developed. The S was presented with a small tilted room, on the back wall of which was a rod which he had to adjust to the true upright. In a first test, judgments of the rod were obtained with body upright, and under three different conditions of the field: (a) The S, standing at a distance from the tilted scene, viewed it through a tube which restricted his view to the interior of the scene. (b) The S stood directly in front of the scene without a tube. (c) The S stood at a distance from the scene, without a tube, so that he saw not only the tilted scene but the outer upright room as well. Under all three conditions the perceived vertical and horizontal were displaced significantly in the direction of the axes of the tilted scene. When an outer upright field was present, as in the third condition, the effect of the tilted scene upon the perceived upright diminished. In another experiment, judgments of the rod were obtained with body tilted, both to the same side as the field and to the opposite side. It was found that tilting the body resulted in an increased tendency to accept the tilted field as a basis for judging the upright. Not only were the perceived vertical and horizontal displaced further in the direction of the tilted scene, but a number of Ss perceived the tilted scene as fully upright. In another experiment there was found a tendency for the tilted scene to right itself with prolonged observation. In some cases the righting was complete, so that at the end of the observation period, the tilted scene was perceived to be fully upright. Under all conditions, striking individual differences were found in the extent to which the perceived upright was affected by the surrounding tilted field. There is evidence of considerably consistency in a person's performance under the different conditions employed.
为克服倾斜镜法在分离视觉和姿势因素以确定感知垂直方向时存在的某些缺点，开发了一种新程序。让被试处于一个小型倾斜房间中，房间后墙上有一根杆，他必须将其调整到真正的垂直方向。在第一个测试中，在身体垂直的情况下，以及在三种不同的视野条件下获得对杆的判断：(a) 被试站在离倾斜场景一定距离处，通过一根管子观察，该管子将他的视野限制在场景内部。(b) 被试直接站在场景前，不使用管子。(c) 被试站在离场景一定距离处，不使用管子，这样他不仅能看到倾斜场景，还能看到外部的垂直房间。在所有三种条件下，感知到的垂直和水平方向都明显向倾斜场景的轴方向偏移。当存在外部垂直视野时，如在第三种条件下，倾斜场景对感知垂直方向的影响会减小。在另一个实验中，在身体向与视野相同的一侧倾斜以及向相反一侧倾斜的情况下获得对杆的判断。发现身体倾斜会导致更倾向于将倾斜视野作为判断垂直方向的基础。不仅感知到的垂直和水平方向在倾斜场景方向上进一步偏移，而且一些被试将倾斜场景视为完全垂直。在另一个实验中发现，随着观察时间延长，倾斜场景有自行扶正的趋势。在某些情况下扶正完成，因此在观察期结束时，倾斜场景被视为完全垂直。在所有条件下，发现个体在感知垂直方向受周围倾斜视野影响的程度上存在显著差异。有证据表明，在采用的不同条件下，一个人的表现具有相当的一致性。
