Chapter 12: Adult Language Learning

12.4 Cognitive processes in language learning

So far we’ve focused a lot on the social side of adult language learning. This section considers the psychological side: what cognitive processes occur when you’re learning a later language? Just like in L1 learning, the overall goal is to construct a mental grammar for the language. What has to happen for the mind to arrive at a useable grammar?

One crucial ingredient seems so obvious that we might even overlook it: a language learner needs input: they need to see the language signed or written, or hear it spoken. The more input, the better, but input isn’t just about quantity – the learner also needs the input to be used communicatively, so that the mind links the forms of the language (the signs, sounds, words and phrases) with a meaningful message.

There’s quite a lot of debate in the literature as to whether the patterns of a later language are learned from overt instruction, or unconsciously from the language input. Researchers in this subfield often contrast declarative knowledge with procedural knowledge (Levelt, 1989; Paradis, 2009; Ullman, 2005). Declarative knowledge consists of things you know explicitly, such as your postal code, how to do laundry, or how to get to your Linguistics classroom on campus. On the other hand, procedural knowledge is unconscious – it’s the kind of knowledge that allows you to recognize your sibling even if you haven’t seen them in months, or to shift your weight so you don’t lose your balance when you’re climbing stairs. (In the Exercise section at the end of this chapter, you’ll think about which parts of your grammar are declarative knowledge and which are procedural.)

Regardless of whether your mind is learning the grammar implicitly or explicitly, the input is only useful if your mind can process it. When you first start learning a language as an adult, you can probably only pay attention to a small part of the input. Here are a couple examples of languages that are probably unfamiliar to most of you:

  • a user of Auslan (Australian Sign Language) describing an event happening in Melbourne in January 2022 [YouTube video], and
  • a user of Kaurna (an Indigenous language of South Australia) introducing himself on the radio [Audio at First Languages Australia].

Unless you know some Auslan or some Kaurna, you probably can’t make much use of the input: you can’t recognize individual words or morphemes, and maybe you can’t even tell where words begin and end. What these two examples illustrate is that an adult learner’s intake is often considerably less than what’s available in the input. Your mind can only construct a grammar from its intake: the language input that it actually processes. And it can only process a limited amount at a time because your attention and working memory have a finite capacity.

Because processing constraints limit the intake a learner can obtain from the input, many adult learners’ early grammars consist disproportionately of content words, rather than function words or inflectional morphemes. When you’re trying to understand or express a certain idea, the semantic content of nouns and verbs gives you more crucial information than, say, conjunctions or tense morphemes!

Limitations on processing capacity also affect how adults produce the language they’re learning. When adult learners of signed languages are starting to sign, they tend to make systematic simplifications in their utterances. The most common articulation errors are in handshape, especially in signs produced with two hands, and all the more in signs where the two hands move independently (Marshall, 2020; Ortega & Morgan, 2015). Notice that these errors happen even though adults usually have a fully developed motor system: it’s not that the learners haven’t yet mastered the control of their articulators, it’s that they haven’t yet mastered the complex phonology of the language they’re learning.

For both comprehension and production, gaining proficiency in a later language involves moving from conscious, effortful processing to rapid, automatic processing. And that takes practice – lots and lots of practice. The only way to build up robust neural pathways in the brain for the new grammar is to use the language. At first, an adult learner’s grammar is likely to include a lot of memorized phrases. To develop a grammar that can be used productively, generatively, you’ll want to practice using the language in complex, naturalistic settings, not just the structured dialogues of a classroom (Götz, 2013).


When you’re learning a later language, there’s a period during which you have some degree of competence in producing and understanding, but your grammar is still quite different from that of an L1 user. This not-yet-fluent grammar is called interlanguage (Selinker, 1972), and some researchers have argued that interlanguages have a lot in common with each other, no matter what your L1 is, and no matter what later language you’re learning (Klein & Perdue, 1997). The idea is that the process of learning a later language is similar across many languages because the human mind is similar.

This interlanguage phase has some interesting consequences when it comes to speech accents. A team of researchers in Chicago worked with international students who had just arrived in the USA to begin graduate programs (Bent & Bradlow, 2003). The students had arrived from many different countries and spoke many different languages as their L1. The researchers asked these adult English learners to record a set of simple English sentences, and had some L1 English speakers do the same. Then they mixed the sentences with white noise, played them back to a variety of listeners, and asked the listeners to write down what they heard.

For each talker, the researchers calculated an intelligibility score. The word intelligibility sounds like it refers to some objective, acoustic measure of speech clarity, but in this case what it means is how many errors listeners made in writing down what a talker said. If listeners make few errors, you have a high intelligibility score, and if listeners make a lot of mistakes, your intelligibility score is low.

The obvious prediction is that that the L1 English speakers would have the highest intelligibility scores. And that was partly true: for L1 English listeners, the L1 English talkers were the most intelligible. But that was not the case for the later learners! When the listeners were adult learners of English with L1s as diverse as Bulgarian, Dutch, Greek, Hindi, Spanish, and many others, the late-learning English talkers were more intelligible than the L1 English talkers! Bent and Bradlow (2003) called this effect the “interlanguage speech intelligibility benefit” (2003, p. 1608). They argue that, because the effect was observed even when the talker and listener had unrelated L1s, the benefit is probably not due to similarities in the L1s, but due to shared experience in learning English as a later language – in other words, the talker and listener probably have some common properties in their interlanguage.

Check your understanding


Bent, T., & Bradlow, A. R. (2003). The interlanguage speech intelligibility benefit. The Journal of the Acoustical Society of America, 114(3), 1600–1610.

Expression Australia. (2022, January 4). Rainbow Project Midsumma Video.

Gambay: Australian First Languages Map. (n.d.). Retrieved July 5, 2022, from

Götz, S. (2013). Fluency in Native And Nonnative English Speech. John Benjamins Publishing Company.

Klein, W., & Perdue, C. (1997). The Basic Variety (or: Couldn’t natural languages be much simpler?). Second Language Research, 13(4), 301–347.

Levelt, W. J. M. (1989). Speaking: From intention to articulation. MIT Press.

Marshall, C., Bel, A., Gulamani, S., & Morgan, G. (2020). How are signed languages learned as second languages? Language and Linguistics Compass, e12403.

Onkwawenna Kentyohkwa. (n.d.). Retrieved August 18, 2022, from

Ortega, G., & Morgan, G. (2015). Phonological Development in Hearing Learners of a Sign Language: The Influence of Phonological Parameters, Sign Complexity, and Iconicity. Language Learning, 65(3), 660–688.

Paradis, M. (2009). Declarative and Procedural Determinants of Second Languages (Studies in Bilingualism Vol. 40). John Benjamins Publishing Company.

Selinker, L. (1972). Interlanguage. IRAL : International Review of Applied Linguistics in Language Teaching, 10(3), 209–231.

Tavakoli, P., & Wright, C. (2020). Second Language Speech Fluency: From Research to Practice (1st ed.). Cambridge University Press.

Ullman, M. T. (2005). A cognitive neuroscience perspective on second language acquisition: The declarative/procedural model. In C. Sanz (Ed.), Mind and Context in Adult Second Language Acquisition: Methods, Theory, and Practice (pp. 141–178). Georgetown University Press.


Icon for the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License

Essentials of Linguistics, 2nd edition Copyright © 2022 by Catherine Anderson; Bronwyn Bjorkman; Derek Denis; Julianne Doner; Margaret Grant; Nathan Sanders; and Ai Taniguchi is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License, except where otherwise noted.

Share This Book