The critical period hypothesis has been the main focus of research on age and L2A for decades. This hypothesis originated with Lenneberg. In his 1967 study, he proposed that for language to develop fully, it must be acquired before the start of puberty. His hypothesis was solely regarding first language (L1) acquisition. Johnson and Newport (1989) were among the first to question if and how this hypothesis applies to second language acquisition. They conducted a test with a group of 46 Chinese and Korean immigrants who had come to the United States at a variety of ages, ranging from 3 to 39, and learned English as a second language. To measure their ultimate attainment in English grammar, the experiment asked each individual to decide…show more content… For example, some research yielded results in which native and non-native speakers achieved identical scores on a variety of language tasks if the non-native speakers’ AoA was 6 or below. It thus seems that there is a sensitive period from birth to somewhere between age 4 and 7 for phonology and some lexical and collocational skills (Long, 2005). Based on a study by DeKeyser (2000) that replicated Johnson and Newport (1989), there might be a separate sensitive period for morphosyntax until sometime in the mid-teens, around the age of 16. DeKeyser attributes the differences in his results to the fact that Johnson and Newport only tested a small number of subjects between the ages of 12 and 16.
DeKeyser used the findings from his study to reconceptualize the Critical Period Hypothesis. He found that learning ability does not decline immediately at puberty, as previously supposed, but that there is a gradual decline from age 6 or 7 to age 16 or 17. In a study of 57 Hungarian immigrants to the US (DeKeyser, 2000), he found that most of the subjects with an AoA over 16 did not score anywhere near the younger learners on a grammaticality judgment test. The few adult learners who did score highly all had strong verbal analytic ability. Instead of interpreting this evidence as a counterargument for the CPH,