Critical Period Claim Revisited: Reanalysis of Hartshorne, Tenenbaum, and Pinker (2018) Suggests Steady Decline and Learner‐Type Differences

A reanalysis of data drawn by Hartshorne, Tenenbaum, and Pinker (2018) from two‐thirds of a million English speakers showed that their overall conclusion of one sharply defined critical age at 17.4 for all language learners is based on artificial results. We show that instead of a discontinuous exponential learning with sigmoidal decay (ELSD) model, a continuous ELSD model had a better fit when applied separately to monolinguals, bilinguals, and early immersion learners. Only for nonimmersion learners and later immersion learners did a discontinuous ELSD model have a better fit, with a critical age of 18.6 and 19.0 years of age, respectively. These age effects can be interpreted as schooling effects. We suggest that personal and societal factors, including differences in living circumstances and socialization, may bring about age‐specific discontinuity patterns in language learning and in language learning rate. The implication is that they are not the outcome of cognition‐driven developmental factors leading to one or more critical periods.