The Critical Period for Language Acquisition: Evidence from Second Language Learning

English speakers of several age groups acquiring Dutch naturalistically in the Netherlands

English-speaking children, adolescents and adults newly arrived in the Netherlands, tested repeatedly over their first year of Dutch exposure

Repeated-measures naturalistic study across age bands (roughly 3-5, 6-7, 8-10, 12-15 and adult), with pronunciation, morphology, vocabulary, sentence repetition, translation and story comprehension measured at intervals over the first year. Observational: age is not assignable and the age groups differ in schooling, social contact and literacy.

The classic finding, and the one most often reversed in popular retellings: the 12-15-year-olds and the adults made the FASTEST initial progress in Dutch, and the 3-5-year-olds the slowest, over the first months of exposure. Younger learners closed part of the gap by the end of the first year, particularly on pronunciation, but the older learners were not behind at any point in that year. Rate of early acquisition therefore favours older learners; it is only ultimate attainment, measured years or decades later, that favours early starters.

Medium-to-high; groups differ by age in literacy, metalinguistic knowledge, schooling and social context as well as in maturational state, and none of this is randomised.

The "older learners are initially faster" result is one of the most reliably reproduced findings in second-language research and is reproduced in instructed settings by the Barcelona Age Factor project. It is not in dispute; what is disputed is what it implies for ultimate attainment, which it does not measure.