Learning from the unexpected

 This is a short extract from our book Memory: What Every Language Teacher Should Know (Smith and Conti, 2021). The chapter this is from was initially inspired by some reading I had done from a book by French cognitive neuroscientist Stanislas Dehaene. One of his main ‘pillars of learning’ is the power of the unexpected to trigger new learning. When we encounter something we don’t anticipate our mind is alerted and prompted to pay extra attention. With that extra attention comes learning.

Here is the extract.

Are you the type of language teacher who believes students need to be accurate from the start or that it’s fine to make mistakes and that, indeed, we learn and remember more by doing so? Do you like to correct students’ speech and writing? Do you think it’s useful to show examples of faulty language in the name of building memory? To what extent can producing and being exposed to errors increase memory? This is the subject of this chapter.

In general terms, psychologists believe that feedback is a vital part of the learning process. Educationalists John Hattie and Helen Timperley define feedback as “information provided by an agent… regarding aspects of one’s performance or understanding” (Hattie and Timperley, 2007). Winne and Butler (1994) note that feedback is “information with which a learner can confirm, add to, overwrite, tune, or restructure information in memory”. 

For language learners feedback and error correction come mainly from the teacher and to a lesser extent peers. In classroom work it’s an ongoing process of confirmation, recasting of responses and occasional explicit correction. On paper, it’s a range of corrective practices which we’ll look at below. For the best practitioners it is a finely-tuned, subtle process of formative asssessment, or what has been called responsive teaching (Fletcher-Wood, 2018) – the bread and butter of effective teacher practice.

Instinctively language teachers think feedback is vital and they devote countless hours of time correcting students’ written work. But when it comes to error correction, the second language acquisition literature has provided mixed messages over the years (our best bet being that correcting spoken and written errors may be useful if done at the right time, with the right students in the right way). Cognitive science and second language acquisition research provide some other useful insights about whether receiving feedback, and making, hearing or seeing mistakes, can enhance memory. 

Let’s turn to cognitive science first. Neuroscientist Stanislas Dehaene has suggested what he calls four “pillars of learning” based on research into cognitive psychology and neuroscience. He argues that each of these pillars is vital for building long-term memory. He calls them attention, active engagement, feedback and consolidation.

To put these four pillars in everyday terms, imagine you are learning to do a dance. You pay attention to the teacher, try out the steps, get some wrong and correct them, then practise over and over again. It’s the third pillar, responding to error, which we are considering in this chapter. 

Firstly, Dehaene explains that we learn not so much by making associations between information, but by noticing things which seem wrong. Dehaene puts it, surprise is the driving force of learning. Researchers Robert Rescorla and Allan Wagner hypothesised that the brain learns only if it perceives a gap between what is predicts and what it receives. This goes back to Jean Piaget’s idea that cognitive development happens when there is a mismatch between a child’s existing schemas and feedback from the environment. For learning, cognitive conflict (an error signal) is needed or, as Rescorla and Wagner (1972) put it, “organisms only learn when events violate their expectations”. As a second language speaker you will be aware of those moments when you hear something unexpected or which does not match with what you thought. It’s often at those times that you notice something interesting and often remember it for later. Put another way, you can see the brain as a type of machine that generates predictions, receives information, then compares the two. If an error signal is created which generates a better prediction for the future. The following sequence is continually repeated: prediction – feedback – corrections – new prediction. You can see this as a sort of statistical process when learning is only triggered if there is an error signal.

According to Dehaene, this doesn’t mean we have to make mistakes to learn, but it’s about a perception of a discrepancy between what we expect and what we get. Interestingly, however, in some theories of second language acquisition, it is argued that perception of errors we make ourselves is important for developing proficiency. The so-called Interaction Hypothesis (Long, 1996) claims that we need to test out our utterances with other speakers to get feedback and notice (Schmidt, 1990) when we make mistakes in order that we can improve. Interactions with teachers and other students can result in negative feedback. That is, if a student says something that their interlocutor doesn’t understand, after negotiation the interlocutor may model the correct language form. Think of it like this: when a student makes a mistake they are trying out a hypothesis about the language. Corrective feedback tells them if it was right.

You may have noticed that this theory ties in rather neatly with what we said about curiosity and the testing effect in previous chapters. The simple act of being curious about an answer and trying to retrieve it, even if you make a mistake, leads to learning and better memory.