Bilingualism and Cognition: Some Recent Findings

Abstract

The purposes of this cross-sectional study were (1) to assess and describe the English/Spanish relative linguistic proficiency (RLP) of four Hispanic background populations in grades one and three, and (2) on the basis of this description, to assess the performance of children at increasingly higher levels of RLP on three measures of cognitive functioning. Following a rationale emphasizing cognitive-development and learning theory approaches, the primary prediction was that there would be significant differences in performance by children at increasingly higher levels of RLP. In accordance with an interpretation of bilingualism based on the metaset (De Avila & Duncan, 1979) it was predicted that the bilingual children’s performance on all measures would be superior to that of the monolingual children. The subjects were 204 school children in grades one and three, selected from urban and rural Mexican-American, and urban Puerto Rican-American and Cuban-American communities. The Language Assessment Scales (LAS) were used to identify five linguistic comparison groups: Proficient Bilingual, Partial Bilingual, Monolingual, Limited Bilingual, and Late Language Learners. One of the most important findings was that there was a positive and significant relationship between degree of relative linguistic proficiency and cognitive-perceptual performance of the children. The Proficient Bilingual children consistently and significantly outperformed the four other comparison groups on two of the. three dependent measures. The children identified as Late Language Learners consistently produced the lowest mean scores on all dependent measures. For all groups, level of RLP accounted for significant differences on measures of field dependence and intellectual development. There were no consistent differences between the Limited Bilinguals (children identified as having deficiencies in both languages) and Monolinguals. This finding supports a reversal of the usual view of limited-English speaking children as being intellectually inferior to their monolingual peers. Furthermore, this no-difference finding is interpreted as support for the proposal that a “threshold level” of bilingualism is required before the relationship between bilingualism and metacognition becomes apparent. Results of this study would seem to indicate that the minimum threshold level is the “nearnative fluency” in two languages which the least articulate of the Proficient Bilinguals have attained. Finally, implications for education and further research are discussed.