The Critical Period Hypothesis in Second Language Acquisition: A Statistical Critique and a Reanalysis

Abstract

In second language acquisition research, the critical period hypothesis (cph) holds that the function between learners' age and their susceptibility to second language input is non-linear. This paper revisits the indistinctness found in the literature with regard to this hypothesis's scope and predictions. Even when its scope is clearly delineated and its predictions are spelt out, however, empirical studies–with few exceptions–use analytical (statistical) tools that are irrelevant with respect to the predictions made. This paper discusses statistical fallacies common in cph research and illustrates an alternative analytical method (piecewise regression) by means of a reanalysis of two datasets from a 2010 paper purporting to have found cross-linguistic evidence in favour of the cph. This reanalysis reveals that the specific age patterns predicted by the cph are not cross-linguistically robust. Applying the principle of parsimony, it is concluded that age patterns in second language acquisition are not governed by a critical period. To conclude, this paper highlights the role of confirmation bias in the scientific enterprise and appeals to second language acquisition researchers to reanalyse their old datasets using the methods discussed in this paper. The data and R commands that were used for the reanalysis are provided as supplementary materials.

More Is More in Language Learning: Reconsidering the Less‐Is‐More Hypothesis

“…For example, an influential study involving native Korean and Chinese learners of English demonstrated that accuracy in judging the grammaticality of a wide range of English constructions declined steadily as a function of learners' age of first L2 exposure up to puberty, even when other experiential and attitudinal variables such as overall years of L2 exposure, amount of classroom instruction, identification with social values of the host country, degree of self-consciousness, and motivation to learn the L2 and to speak it well were taken into account (Johnson & Newport, 1989). Although these and similar studies have been shown to be severely underpowered (Hartshorne, Tenenbaum, & Pinker, 2018;Vanhove, 2013), recent evidence from an extraordinarily large sample exceeding half a million L1 and L2 speakers of English confirmed that ultimate grammatical proficiency declines with age of first L2 exposure, even if the estimated offset of the critical period between 17 and 18 years appears to be much later than previously estimated (Hartshorne et al, 2018).…”

Making the most of an early start to L2 instruction

“…As for the idea of thresholds: (1) there is no empirical evidence in favour of discontinuous bilingual development (see Vanhove, 2013Vanhove, , 2015; (2) unless thresholds are defined independently, they do not yield predictions that can be proven wrong (Berthele, 2017) -and so the probability of producing false positive results is increased (Vanhove, 2014); and (3) the threshold cannot be defined in absolute terms -i.e. it is impossible to identify a common threshold for all learners.…”

Age and Foreign Language Learning in School

“…These test results also showed a success rate that was inversely proportional to age of onset; the students who were last in starting to learn English outperformed the students who had been able to start classes at an earlier age. We encourage interested readers to consult Vanhove's (2013) An advantage for later starters also appears in the final report of a study by 10 Myles et al (2012). However, the information in this report is not sufficiently complete to gain a full picture of the study.…”

“…Yet because these studies are often cited in the literature on the age factor, we very briefly present two of the main study methods. We encourage readers who wish to learn more about the topic to consult the articles by Bloch et al (2009), Kim, Relkin, Lee, andHirsch (1997), Perani et al (1996Perani et al ( , 2003, Steinhauer (2014), Wattendorf and Festman (2008), Wattendorf et al (2012), andWeber-Fox andNeville (1996).…”