9-26个月婴儿上位、基本和下位水平类别学习的发展顺序

Order of Acquisition of Superordinate-, Basic-, and Subordinate-level Categories on 9-26 Month's Infant

研究探索了9-26个月婴儿在上位水平类别、基本水平类别和下位水平类别的发展顺序。采用触摸屏式的序列触摸任务,36名婴儿参加实验,分9-14月、15-20月、21-26月三个年龄组。以平均序列长度作为组分析的因变量指标,以Dixon等(1998;2007)开发的Touchstat V3.0软件统计的归类者概率作为个别分析的因变量指标,结果表明,三个年龄组的婴儿,都能显著地区分基本水平的类别;15-20个月和21-26个月婴儿,能够显著区分上位水平的类别;21-26个月的婴儿,能够区分下位水平的类别,说明了婴儿不同类别水平概念的学习,是按照基本水平(L2),然后上位水平(L1),最后才是下位水平(L3)的顺序发展的,即L2→L1→L3。

Categorizing is an essential and important cognitive achievement. Object categories refer to shared representations of like but discriminable objects. Categories are especially valuable in infancy and early childhood, because the ability to categorize enable them to respond anew to each novel entity they experience. Category learning in infants usually refers to 1- to 3- year olds’ categorizing. Having a look at the previous study on in infancy, there can be summarized two views ： one stressed on perceptual-based category learning, the other stressed on categorization inclusiveness hierarchy. The latter one emphasizes hierarchical inclusiveness, which classifies category levels to superordinate level （L1）, basic level （L2）, and subordinate level （L3）. We are interested in which level infants first categorize on. Different methods have been devised for testing infants ages, and different result has gained. Here we improve the former researcher’s design and adopt some new ideas and technique to study the categorizing ability of infants of different ages on different levels and then draw the developmental trajectory of infant categorization. Three levels of category inclusiveness in 4 object domains （animals, vehicles, fruit, and furniture） were examined and assessed using a sequential touching procedure in thirty 9-14 months, 15-20 months and 21-26 months. For group analysis, we followed the tradition of testing children’s mean run length in each cell against chance （Mandler, 1987） in SPSS. For individual analysis, TouchStat V3.0（Dixon, et al. 2007） is adopted to assess the individual categorizer. The percentage of categorizers is calculated and two styles of categorizers are classified. Results shows： Mean run length at the basic level （L2） exceeded chance for all age groups; significant mean run lengths at the superordinate level （L1） varied by age, with mean run lengths exceeding chance in 15-20 months and 21-26 months; mean run lengths at the subordinate level of categorization （L3） didn’t exceed chance at 9-14 months and 15-20 months, but shows marginal significance which indicates a tendency of forming the categorizing ability. The data of individual analysis also shows the priority of basic level （L2）, for 9-14 months infant, reveals that there are 41.7% of categorizers at the basic level, 25% at superordinate level, and 8.3% at subordinate level. Meanwhile, the gap of these percentages between different level narrows by age. On the whole, the study reveals a developmental trajectory of infants as： Basic level （L2） → Superordinate level （L1） → Subordinate level （L3）.