Critical Period Effects in Second Language Acquisition: The Effect of Written versus Auditory Materials on the Assessment of Grammatical Competence*

Abstract

Current research has supported the existence of a critical period for the acquisition of the grammar of a second language. In one study, native speakers of Chinese and Korean, who had arrived in the U.S. at varying ages, were tested on their knowledge of English grammar using an auditory grammaticality judgment task (Johnson & Newport, 1989). Subjects were tested only after having lived in the U.S. for many years as students or academicians. The present study asks whether the poorer performance exhibited by the older arrivals was due to their difficulty with English grammar or merely due to extragrammatical properties associated with an auditory task. The same subjects who participated in the earlier study were tested a year later using an untimed written version of the same test. A strong negative correlation was still observed between age of arrival and performance on the written test: r=‐.54 for subjects of all ages of arrival; r=‐.73 for subjects who arrived to the U.S. before adulthood. Performance was higher on the written version than on the auditory version, but only for the older arrivals. The possibility that earlier formal classroom training boosted the adult arrivals’ performance on the written test is discussed.