Phillips Webb, Santos Laurie R
Department of Psychology, Yale University, USA.
Cognition. 2007 Mar;102(3):455-63. doi: 10.1016/j.cognition.2006.01.009. Epub 2006 Mar 3.
How do we come to recognize and represent different kinds of objects in the world? Some developmental psychologists have hypothesized that learning language plays a crucial role in this capacity. If this hypothesis were correct, then non-linguistic animals should lack the capacity to represent objects as kinds. Previous research with rhesus monkeys (Macaca mulatta) has shown that this species can successfully individuate different kinds of objects - monkeys who saw one kind of object hidden inside a box searched longer after finding a different kind of object. However, in these studies and the infant studies on which they were based, the objects to be individuated differed both in kind and in properties. Thus, subjects in these experiments may not be representing the kinds of objects per se, but instead only their immediate perceptual properties. Here, we show that rhesus monkeys successfully individuate different kinds of objects even when their perceptual properties are held constant. Although these data provide the best evidence to date that language is not necessary to represent kinds, we discuss our findings in terms of possible associative hypotheses as well.
我们如何开始识别并表征世界上不同种类的物体呢?一些发展心理学家假设,学习语言在这种能力中起着关键作用。如果这个假设是正确的,那么非语言动物应该缺乏将物体表征为种类的能力。先前对恒河猴(猕猴)的研究表明,这个物种能够成功地区分不同种类的物体——看到一种物体藏在盒子里的猴子在找到不同种类的物体后会搜索更长时间。然而,在这些研究以及它们所基于的婴儿研究中,要区分的物体在种类和属性上都有所不同。因此,这些实验中的受试者可能并不是在表征物体本身的种类,而只是它们直接的感知属性。在这里,我们表明,即使恒河猴的感知属性保持不变,它们也能成功地区分不同种类的物体。尽管这些数据提供了迄今为止关于表征种类不需要语言的最佳证据,但我们也从可能的联想假设的角度讨论了我们的发现。