Meltzoff A N
Department of Psychology, University of Washington, Seattle 98195.
Ann N Y Acad Sci. 1990;608:1-31; discussion 31-7. doi: 10.1111/j.1749-6632.1990.tb48889.x.
This chapter began with a query about whether there was any content to an enterprise called "developmental cognitive science," and if so, whether the findings could inform work in adult cognition and neuropsychology. Both questions can now be answered in the affirmative. Evidence has been marshaled from infant studies concerning five topics of enduring interest in the cognitive and neuro-sciences: cross-modal integration, imitation, the coordination of perception and action, memory, and representation. The data show that young human infants can detect equivalences between information picked up by different sensory modalities. This was demonstrated both in tactual-visual perception of objects and auditory-visual perception of speech. Results also show that perception and production are intertwined literally from the earliest phases of infancy, with 4-month-olds demonstrating vocal imitation and newborns reproducing elementary gestures they saw an adult perform. There seems to be a transparency between the perceptual and motor systems, and it is conceivable that they may draw on the same internal code. Infants' proclivity to imitate was used to investigate early memory. It was found that young infants were not constrained to immediate mimicry, but could imitate after significant delays. The findings support the inference that infants, perhaps as early as birth, have a functioning memory system that cannot be reduced to "habit formation" or an exclusively "procedural memory." It was proposed instead that there is a kernel of some higher level memory system right from the earliest phases of human infancy. This does not imply that there is no development in the representational world of infants. Data were reviewed suggesting that there is a watershed transformation in childhood cognition at about 18 months of age. However, this is not a change from a stage in which there was a purely sensorimotor or habit-based system. Rather the development was characterized as a shift from using empirical or experience-based representations to using hypothetical representations, which concern possible realities. This developmental shift allows children to project into the future "what must be" and deduce from the past "what must have been," in advance of, and sometimes in the absence of, strictly perceptual evidence. This capacity provides the underpinnings for the conduct of science itself. Its origins are to be found in infancy.
名为“发展认知科学”的学科是否有实质内容？如果有，其研究结果能否为成人认知和神经心理学领域的工作提供参考？现在，这两个问题的答案都是肯定的。从婴儿研究中整理出的证据涉及认知科学和神经科学长期关注的五个主题：跨模态整合、模仿、感知与行动的协调、记忆和表征。数据表明，人类幼儿能够察觉到通过不同感官模态获取的信息之间的等效性。这在物体的触觉 - 视觉感知以及语音的听觉 - 视觉感知中都得到了证明。研究结果还表明，从婴儿期的最早阶段起，感知和行为就紧密相连，4个月大的婴儿就能进行声音模仿，新生儿也能重现他们看到成人做出的基本手势。感知系统和运动系统之间似乎存在一种透明度，并且可以想象它们可能使用相同的内部代码。婴儿的模仿倾向被用于研究早期记忆。研究发现，幼儿并不局限于即时模仿，而是能够在显著延迟后进行模仿。这些发现支持了这样的推断：婴儿可能早在出生时就拥有一个运作的记忆系统，该系统不能简单归结为“习惯形成”或单纯的“程序性记忆”。相反，有人提出，从人类婴儿期的最早阶段起就存在某种更高层次记忆系统的核心。这并不意味着婴儿的表征世界没有发展。回顾的数据表明，在大约18个月大时，儿童认知会发生一个分水岭式的转变。然而，这并非是从一个纯粹基于感觉运动或习惯的系统阶段转变而来。相反，这种发展的特点是从使用基于经验或体验的表征转变为使用涉及可能现实的假设性表征。这种发展转变使儿童能够在没有严格感知证据的情况下，甚至有时在没有感知证据的情况下，提前推测未来“必然是什么”，并从过去推断“必然已经是什么”。这种能力为科学本身的开展提供了基础。其根源可以追溯到婴儿期。
