The critical period hypothesis (CPH), proposed by Eric Lenneberg (1967), holds that there is a biologically determined period — roughly from birth to puberty — during which language acquisition proceeds naturally and effortlessly. After this period, language learning becomes more difficult and less likely to achieve native-like proficiency. The hypothesis draws on evidence from first language deprivation, second language acquisition, and the neurobiology of language development.
Evidence
The strongest evidence comes from cases of extreme language deprivation. Genie, discovered at age 13 after years of severe isolation, acquired some vocabulary but never mastered grammar, consistent with a critical period for syntactic development. Deaf children who receive cochlear implants or are exposed to sign language later in childhood show progressively poorer language outcomes with increasing age of exposure. Second language acquisition studies consistently show that age of acquisition predicts ultimate attainment, particularly for phonology and grammar, though the relationship may be gradual rather than showing a sharp critical period boundary.
Many researchers prefer "sensitive period" to "critical period," arguing that the decline in language learning ability is gradual rather than abrupt and that older learners can achieve high proficiency with sufficient input and motivation. The largest study to date (Hartshorne et al., 2018, with 670,000 participants) found a gradual decline in grammatical learning ability beginning around age 10 and leveling off around age 17, consistent with a sensitive rather than strictly critical period.
Neural Basis
The critical period may reflect declining neural plasticity. Synaptic pruning, myelination, and the stabilization of cortical circuits reduce the brain's flexibility for acquiring new language structures. Early language experience shapes the neural organization of language processing, with left-hemisphere language networks becoming progressively more committed to the native language. Late learners show different patterns of brain activation during language processing compared to early learners, even when behavioral proficiency is matched.