Why reading aloud leads to better recall

The production effect states that when we read aloud, our memory of the information is stronger than if we read silently to ourselves. Yet this behaviour is often viewed with disdain, especially in older students.

When we first learn to read we read out loud, perhaps to a teacher or a parent. Once we become more competent we develop the ability to read silently to ourselves and, as we approach adulthood, we rarely read out loud at all. This isn’t the case for everyone, of course. Teachers read out loud to their students and parents and carers read to their children, but more generally, the words remain firmly contained within our own minds.

Back to the ’80s

My own schooling was very much like this to some degree or another. Silent reading was the norm and when exams approached we were expected to revise in silence. We would resort to rereading, note taking and engage in copious amounts of underlining, all strategies that have been found to be highly ineffective. Silence gave us little space to quiz each other or engage in verbal interrogation of the content. Reading out loud was an activity that nobody really considered to be useful, indeed, reading out loud or even moving our mouths in the absence of external vocalisation was thought to be a sign of low intelligence or regression to an earlier stage of development. 

Hello. My name is marc

There is a strategy that some people advocate to help them remember the names of people they have recently met. When a person tells you their name, you repeat it back to them, so if I were to say, ‘Hello, I’m Marc’ you would reply with ‘Marc, pleased to meet you’ or, ‘Pleased to meet you Marc, I’m [insert name here].’ Vocalising the name acts as an anchor and makes it more likely that we are able to recall the name when we next meet. We could, of course, simply make a mental note but, curiously, this doesn’t appear to work quite as well.

The production effect

This phenomenon has been called the production effect by psychologist and has been studied since as far back as the early 1970s. The basic premise is that reading words aloud results is substantially better memory than reading them silently. The hypothesis makes sense, especially if we relate to established models of memory. Put simply, by increasing the number of possible access routes to specific information, the more likely we are to recall it. As we have seen, the working memory model proposes a specialised component that processes sound-based information (the phonological loop). The phonological loop itself is divided into a phonological store capable of holding speech-based information and an articulatory control process based on inner speech. When we say a word out loud we also hear it externally, so it should involve an additional stage of processing than simply repeating it silently to ourselves. However, we can still see a similar (albeit weaker) effect when we mouth the words silently.

Studies investigating the production effect have tended to assess the ability of volunteers to recall word lists under different conditions. Such studies generally have participants learn words by, for example, reading them aloud, writing them down, typing them or spelling them and, while many of these methods produce better recall than rehearsing the words silently, reading them aloud produces consistently higher results (see MacLeod, 2010). Quinlan and Taylor (2013) also found that singing may enhance memory more than speaking, while Jamieson and Spear (2014) discovered that imagining typing a word can increase recall, although actually typing the word is better.

The production effect also lasts, so in this respect it fulfils the relatively permanent criteria of our definition of learning. One study for example, found that volunteers asked to learn textbook passages by reading them aloud performed better on a fill in the blanks task a day later compared to passages read silently (Ozubko et al., 2012 PDF). Knutson and Le Bigot (2014 PDF) found that people are more likely to re-use a reference that they had spoken earlier compared to a reference spoken by someone else.

Encoding

The production effect has obvious links to some of the other learning strategies discussed in this book. First, the generation (or testing) effect implies that retrieval practice enhances recall and learning. However, it’s thought that the production effect enhances memory at the encoding stage (the point at which information is converted into a code that can be stored in long-term memory, while the generation effect appears to strengthen the memory trace at the point of recall. Similarly, the production effect is related to dual-coding, as it creates extra pathways by which the information can be accessed. The difference is that while methods of dual coding tend to combine written and pictorial (phonological and visuo-spatial information), the production effect pairs two types of sound-based code.

 If you enjoyed reading this, you might like to visit my TES page

Or buy me a coffee