Two experiments tested a prediction derived from the recent finding that the Oppel-Kundt illusion – the overestimation of a filled extent relative to an empty one – was much attenuated when the empty part of a bipartite row of dots was vertical and the filled part horizontal, suggesting that the Horizontal-vertical illusion – the overestimation of vertical extents relative to horizontal ones – only acted on the empty part of an Oppel-Kundt figure. Observers had to bimanually indicate the sizes of the two parts of an Oppel-Kundt figure, which were arranged one above the other with one part vertical and the other part tilted -45°, 0°, or 45°. Results conformed to the prediction but response bias was greater when observers had been instructed to point to the extents’ endpoints than when instructed to estimate the extents’ lengths, suggesting that different concepts and motor programs had been activated.

两项实验测试了根据最近的发现得出的预测，即当两部分点行的空白部分垂直且填充部分水平时，奥佩尔-昆特错觉（相对于空白部分高估填充范围）会大大减弱，这表明水平-垂直错觉——相对于水平范围的垂直范围的高估——仅作用于奥佩尔-昆特图形的空白部分。观察者必须双手指示 Oppel-Kundt 图形的两个部分的尺寸，这些部分上下排列，一部分垂直，另一部分倾斜 -45°、0° 或 45°。结果符合预测，但当观察者被指示指向范围的端点时，反应偏差比被指示估计范围的长度时更大，这表明不同的概念和运动程序已被激活。