August 2021: What Does a Critical Period for Second Language Acquisition Mean? Reflections on Hartshorne et al.

Our article (Hernandez, Bodet, Gehm, & Shen, 2021) responds to the proposed critical period for second-language (L2) learning in Hartshorne, Tenenbaum, and Pinker (2018), henceforth referred to as HTP. When we refer to critical periods here, we mean to say a period of heightened learning and neural change related to a particular trait or skill. We must clearly outline this term because critical periods are a topic of intense debate concerning their definition, the approaches to researching them (Colombo, 1982; Nash, 1978), and their application in the many fields that produce research on them. Language, in particular, has been a field of prolific research concerning critical periods because both first-language (L1) learning (early in life) and L2 learning (which could be done early or late in life) provide clear opportunities to observe the mastery of a skill dependent on the age at which it was learned. As such, HTP’s paper may prove to be one of great impact in the fields of language learning and critical periods, considering its unique methodology and controversial conclusions.

In their work, HTP present evidence that a critical period of L2 syntax acquisition (or learning language rules in L2) extends into late adolescence, only declining around the age of 17. This contradicts previous studies’ findings that the window for learning a second language occurred in late childhood or early adolescence (Johnson & Newport, 1989), and HTP argue that the exceptionally large sample of individuals (680,333 subjects) they used to investigate the critical period serves to support their contradictory finding. By using such a massive sample compared to previous studies on the topic, HTP overcame an important hindrance: they were able to now separate the potential effects that the age of an individual, age of first exposure to L2, and duration of exposure to L2 could have on their results, allowing HTP to arrive at a model better able to determine the shape of any decline in L2 syntax learning abilities. Having accounted for these confounds, HTP’s analyses place the beginning of a decline in this ability at the age of 10 to 12, marking the end of the proposed critical period for L2 learning at 17. 

In our response to this finding (Hernandez et al., 2021), we consider three different questions:

1. Why does HTP’s critical period close so much later than those generally used in the language acquisition literature?
2. What role might the decline of other broader cognitive skills play in the decline of L2 learning abilities?
3. What models may best explain the late close of HTP’s critical periods in terms of said other skills? 

As Question 2 may imply, we suggest that the critical period seen by HTP may reflect developmental changes in a behavioral skill encompassing more than just L2 syntax learning. That is, we propose that the window closing on syntax acquisition in L2 (or, for that matter, any other subskill of language acquisition) falls under the closing of a larger biological window of opportunity of some kind. This window does not involve just the learning of grammar; rather, it likely involves a number of different skills that either go along with or form the base on which syntax learning rests. Because we propose that the culprit for closing HTP’s L2 syntax learning critical period is a larger biological window, our response brings biology into HTP’s behavioral results.  

To guide our incorporation of biology in explaining HTP’s findings, we refer to Johnson’s (2011) framework distinguishing between neurobiological and behavioral views of the critical period. Johnson states that it may be unreliable to reduce behavior solely to brain or biological processes, though modeling the underlying hardware to a behavior (in this case, the underlying hardware to L2 syntax learning) could provide insight on the behavior’s origins and, simultaneously, its growth. In other words, Johnson says one can look at the purpose of a structure (behavior) by looking at the scaffolding (neural hardware) and that as the scaffolding grows, so must the structure. Following Johnson’s advice, our paper uses two neurodevelopmental frameworks (which predict the development of a behavior alongside its supporting hardware) to answer Question 3 and better understand HTP’s results: Interactive Specialization (IS) and Neuroemergentism (NM).

IS predicts that behavioral development and physical brain development are reciprocal. Regions specialize over time based on their neighbors’ specialties, and a neighborhood specializes with its regions. As these regions and networks specialize, associated skills and behaviors follow suit. The yeast factory’s neighbor becomes a baker, the baker’s neighbor becomes a grocer, the yeast factory sells extra yeast to the grocer, the neighborhood becomes a food market district. Applied to our critical questions, these might offer some solutions. In response to Question 1, the later-closing critical period could be attributed to a longer period of brain development than previously thought—maybe the process of regions hyper-specializing for this skill (or related skills) continues long after the heightened plasticity of infancy typically associated with L2 learning ability. IS would also predict that researchers may also be able to answer Question 2 by observing the development of specialized regions involved in the skill’s network through early childhood and adolescence. Through this method, it would be possible to see what regions were pruned out of the network of interest (and, through their specializations, what skills) and how their pruning could lead to the changes in consequent abilities in L2 learning. 

The second framework, NM (Hernandez et al., 2019), accounts for the specialization of regions and networks in the brain supporting evolutionarily newer skills (like language) by considering individual development alongside evolutionary predispositions (like the primary auditory cortex to processing auditory information). These predisposed regions can be recruited during an individual’s developmental periods (i.e., HTP’s critical period) to perform new functions in a broader ability (like leveraging skills from the auditory systems in language comprehension). NM also states that no developmental change is isolated. For instance, as a giraffe’s neck grows, it does so with the same neck bones as a human, but it does not do so alone—the cardiovascular system must find a way to pump blood higher, and the back legs must shorten for balance. Change to any one region or skill will snowball across all related skills, no matter how distal.

NM would address our questions concerning HTP’s proposed critical period in a more general way than IS. First, NM would predict that early-life injury to any system involved in syntax learning would have an observable effect on its critical period development (because the regions that would be leveraged into that skill would not develop along the same tract). Armed with that information and neuroimaging techniques for comparing the injured subjects’ data to neurotypical individuals, this could answer Question 2 and elucidate what other cognitive processes are involved in L2 syntax learning through the effect of their injury on that skill (again, no matter how tangentially connected). Studies on the evolutionary bases of those involved systems may also elucidate how they were adapted to L2 syntax learning (and answer Question 2 from a different angle). Again, these predictions are far more abstract, but this matches the more general framework of NM. 

HTP’s publication involving a large number of participants led to the identification of a critical period for syntax with the window for the acquisition closing at 17.4 years of age. In this piece, we describe two frameworks, IS and NM, that could be leveraged in further investigations of HTP’s evidenced critical period and the mechanisms underlying the critical period’s closure. By offering this view, we hope to stimulate future studies that seek to more closely link the closing of this window with neurobiological indices and accompanying psychological processes. 

References

Colombo, J. (1982). The critical period concept: Research, methodology, and theoretical issues. Psychological Bulletin, 91(2), 260–275. https://doi.org/10.1037//0033-2909.91.2.260

Dehaene, S., & Cohen, L. (2007). Cultural Recycling of Cortical Maps. Neuron, 56(2), 384–398. https://doi.org/10.1016/j.neuron.2007.10.004

Hernandez, A. E., Claussenius-Kalman, H. L., Ronderos, J., Castilla-Earls, A. P., Sun, L., Weiss, S. D., & Young, D. R. (2019). Neuroemergentism: A framework for studying cognition and the brain. Journal of Neurolinguistics, 49, 214–223. https://doi.org/10.1016/j.jneuroling.2017.12.010

Johnson, J. S., & Newport, E. L. (1989). Critical period effects in second language learning: The influence of maturational state on the acquisition of English as a second language. Cognitive Psychology, 21(1), 60–99. https://doi.org/10.1016/0010-0285(89)90003-0

Johnson, M. H. (2011). Interactive Specialization: A domain-general framework for human functional brain development? Developmental Cognitive Neuroscience, 1(1), 7–21. https://doi.org/10.1016/j.dcn.2010.07.003

Nash, J. (1978). Developmental psychology: A psychobiological approach. Prentice Hall.