Woodward S A, Lenzenweger M F, Kagan J, Snidman N, Arcus D
Department of Psychology, Harvard University, 33 Kirkland St., Cambridge, MA 02138, USA.
Psychol Sci. 2000 Jul;11(4):296-301. doi: 10.1111/1467-9280.00259.
Previously, we proposed a theoretical framework that classified infants into qualitative categories of reactivity, rather than on a continuous dimension. The present research used an objective statistical procedure (maximum covariance analysis, or MAXCOV) to determine if a qualitative latent structure, consistent with our theoretical conjectures, would be found to underlie quantitative indices of reactivity to stimuli in a sample of 599 four-month-old infants. Results of the MAXCOV analysis showed clear evidence of a latent discontinuity underlying the behavioral measures of infant reactivity. The base rate of the latent class (or taxon) was estimated at 10%. Infants within the putative high-reactivity taxon, compared with infants not in the taxon, were elevated on measures of behavioral inhibition at 4.5 years. These results provide objective empirical support for a central tenet in our theoretical model by supporting the taxonicity of infant reactivity.
此前,我们提出了一个理论框架,该框架将婴儿按照反应性的定性类别进行分类,而非基于连续维度。本研究采用了一种客观的统计程序(最大协方差分析,即MAXCOV)来确定在599名4个月大婴儿的样本中,是否会发现一种与我们的理论推测一致的定性潜在结构,该结构构成对刺激反应性定量指标的基础。MAXCOV分析结果显示,有明确证据表明婴儿反应性行为测量指标存在潜在的不连续性。潜在类别(或分类群)的基础比率估计为10%。与不在该分类群中的婴儿相比,假定的高反应性分类群中的婴儿在4.5岁时的行为抑制测量指标上得分更高。这些结果通过支持婴儿反应性的分类性,为我们理论模型中的一个核心原则提供了客观的实证支持。