Variations in Native Speaker Competence: Implications for First-language Teaching

Experiments were carried out to determine if variations in adult native speaker performance arise from variations in working memory capacity (Just & Carpenter, 1992) or variations in procedural grammatical competence (Ericsson & Kintsch, 1995). Three groups of 18-year-old native speakers of English were tested on comprehension and recall of English Complex NP constructions. Two groups were ‘low academic ability’ (LAA) students and one group were ‘high academic ability’ (HAA) students. Pre-test HAA levels of recall and comprehension were much higher than those of the LAA groups. One LAA group underwent training to improve sentence recall and then carried out a post-test involving recall and comprehension. Levels of recall, but not comprehension, rose to HAA levels. The second LAA group underwent training to improve comprehension and then carried out a post-test involving recall and comprehension. Levels of both recall and comprehension rose to HAA levels. Contrary to the working memory hypothesis, the results support the view that variations in adult native speaker performance arise from variations in procedural grammatical competence. Some implications of this conclusion for the teaching of first-language grammar during the school years are outlined.