中文摘要：

婴儿的动、名词词类学习存在跨文化差异,但是很少研究从注意偏好角度解释这类差异。本研究利用习惯化范式考察汉语婴儿在6-8个月和17-19个月时对事件中人物,动作和物体的区分。结果发现,6-8个月的婴儿仅能区分动作变化,对人物和物体无法区分,而17-19个月的婴儿对三类变化均可以区分。本研究提供了婴儿早期注意偏好发展的实验依据,同时为儿童早期单词获得提供了新的理论基础。

英文摘要：

Word learning requires establishing a mapping between words and concepts.Yet,even before this mapping process,children need to attend to and identify conceptual units from perceptual inputs.From an associative perspective,attention is the initial step of early word learning,and differences in attentional patterns could play a role in language acquisition and development.English learners typically show vocabularies that are dominated by nouns,while studies reveal that verbs are acquired early and in large quantities by Mandarin Chinese learners.However,little research has examined whether English and Mandarin leamers＇early comprehension and production of nouns and verbs could be attributed to attentional patterns.Thus,the interest for the present study is whether we will find fundamental differences in attentional patterns to complex scenes across developmental spans.In particular,we are interested in how the process of word learning may interact with attention among children who have not yet begun this process（i.e.6 month olds） vs.children who are in the middle of an explosive phase of early vocabulary development（i.e.18-month-olds）. In this study,we explore Mandarin-learning infants＇abilities to discriminate among persons,actions,and objects in both silent and word-learning environments.Infants were randomly assigned to a habituation scene,in which a videotaped event of a young woman performing a novel action on a novel object was presented to infants in a habituation paradigm.Once they were bored,the infants were presented with four test trials,in which none（control trial） or one of the three elements changed（person Change,action change,object change） with the other two held consistent with the habituation.The infants＇looking time was calculated on-line by a live experimenter who was blind to the stimuli appearing on the screen and coded the infant＇s eye movement through a curtain hole. These studies aim to further investigate the cognitive mechanism underlying word-to-world mapping.Some