Despite the widespread use and effectiveness of the Modern Language Aptitude Test (MLAT) composite score in predicting individual differences in L2 achievement and proficiency, there has been little examination of MLAT subtests, although they have potential for illuminating components of L2 aptitude and the mechanism of prediction. Here we use regression commonality analysis to decompose the predictive variance from the MLAT into unique components for each subtest alone and for each possible combination of subtests (duos, trios, etc.) that may have shared variance. The results, from a longitudinal study of 307 U.S. secondary students during 2 years of Spanish learning, provide strong evidence for the role of literacy-related skills in all subtests and in predicting all L2 outcomes. These and other results support a view of L1 literacy and language skills leading to metalinguistic development, which in turn leads to stronger L2 aptitude and achievement.

尽管现代语言能力测试 (MLAT) 综合分数在预测 L2 成绩和熟练程度的个体差异方面得到了广泛使用和有效，但对 MLAT 子测试的研究却很少，尽管它们有可能阐明 L2 能力的组成部分以及 L2 能力的机制。预言。在这里，我们使用回归共性分析将 MLAT 的预测方差分解为单独的每个子测试和可能具有共享方差的子测试的每个可能组合（二重奏、三重奏等）的独特分量。对 307 名美国中学生为期 2 年的西班牙语学习进行的纵向研究得出的结果，为读写能力相关技能在所有分项测试和预测所有 L2 结果中的作用提供了有力的证据。这些结果和其他结果支持了这样一种观点：第一语言读写能力和语言技能会导致元语言发展，进而导致更强的第二语言能力和成就。