Abstract: This experiment was performed in two golden monkeys ( Rhinopithecus roxellan-ae), 2 yr. 6 months and 3 yr. 6 months of age respectively. The method was similar to that of experiment ( I ) .At the beginning of the experiment, the subjects were trained to discriminate two triangles that were different in size. Then the generalization-tests with stimuli of variant forms and size were given. The apparatus was a modification of the Wisconsin General Test Apparatus ( WGTA ), but the animals were required to respond in pushing a box with the stimulus in it, rather than displacing the stimulus. The results showed that the animals were able to transfer from the basic training to generalization tests at different levels.

摘要： 人们通常所说的客观事物的大小,是指其相对关系而言。离开客体间的绝对尺寸的比较,也就无所谓大小了。所以没有比较就没有对客体大小属性本身的认识,因而也就不能对它们之间的关系进行概括。格式塔心理家认为在辨别学习中被试是学习了刺激物之间的关系。在特定的试验中,被试不是对绝对刺激值进行反应,而是根据刺激的相对属性,在辨别学习中是如此,在泛化测验中更是如此。因此,被试在这里已经摆脱开具体对象的绝对属性,而将这些属性的关系作为线索,并作出正确的反应。刺激物的绝对属性是可能不断改变的,但它们之间的关系则往往是稳定的,于是被试就能够排除其它因素的干扰而对这种关系作出较稳定的反应。这种关系之所以能同具体的绝对属性具有等值作用,无疑是得力于动物的初级概括能力。西方行为灵长类学家们把灵长动物的这种初级概括叫作“概念的形成”(Brown等,1958; Robinson, 1960; Hicks,1956 ),我们认为这种说法是片面的。