Things to know about phonological memory

This is an adapted extract from our book Memory: What Every Language Teacher Should Know (2021). It's about the the concept of phonological memory.

In the multi-store model of working memory shown above, based on the Alan Baddeley model, (figure from Smith and Conti, 2021), the phonological loop is the most important part as far as language learning is concerned. How does  it work?  Think for a moment about how you hold a word or phrase in your head as you make sense of it, or prepare to say it; or how you say words in your head as you read from a book; or how you try to make sense of some spoken language. These processes are carried out by the phonological loop, what psychologists sometimes call verbal or phonological working memory.

The phonological loop has been described as the mind’s ear, processing and storing sounds. You could think of it as our language learning device, activated in a particular part of the brain (the temporal lobe of the left hemisphere - above and around your left ear). In the common Baddeley model of working memory, the phonological loop consists of two components:

·        The phonological store, which has a limited capacity and holds acoustic information briefly.

·        The articulatory rehearsal process, which allows us to mentally rehearse language to prevent it from decaying, e.g. when you repeat a sentence in your head a few times.

The phonological loop interacts with long-term memory and plays a vital role in the long term recall of the phonological form (sound) of new words and phrases. As new phonological forms are held in the phonological store during rehearsal, so more permanent memory representations are constructed. This is one reason why it’s so important to allow students to hear and repeat new language as often as possible. They need to have a phonological representation of words, not just know what they mean or look like. This makes it easier for them to recognise words and chunks in the continuous stream of speech.

We know some other significant things about the phonological loop. Its capacity or span  is as few as three to five items of language at the same time: but recall that these can be sounds, words, phrases or even whole sentences. This capacity varies to a limited degree from person to person, has a genetic basis and declines as we age after our teens.

The limited capacity of the phonological loop means that a novice second language learner can hold fewer words in working memory than in their first language as they pronounce the words more slowly. The more rapidly a second language speaker can utter a word or phrase, the less space it takes in working memory. So the more you know, and the more fluently you can speak, the easier the job becomes for working memory. This means that we need to carefully control the amount and difficulty level of language students hear and read.

So the phonological store holds sounds very briefly (about two seconds, before for one sound is erased by the next is erased by the next one). This means, for instance, that any distracting sound, including a correction by a teacher or a giggle by a friend sitting next to you may cause instant forgetting.

Furthermore, research shows that learners can recall pairs of words which sound the same less well than those which sound different. So for example, a students might find it easier to remember the pair of French words violon and trompette than the pair fraise (strawberry) and frais (fresh). This is called the phonological similarity effect. Research suggests that the negative interference caused by this effect on memory is stronger than the one caused by the orthographic similarity effect (when two words have similar spellings). Often, though, the two effects combine to make it harder to learn words which sound similar.

This phenomenon has important implications for learning and helps us understand why learning verb conjugations in highly inflected languages (e.g. French and Spanish) can be so challenging, especially for students with a less efficient working memory. In the case of French verbs, the phonological similarity effect is particularly strong, as many verb forms are pronounced exactly the same (e.g. je mange, tu manges il mange, elles mangent). Not surprisingly, many learners of French find it hard to map the phonological form these verbs to the correct spelling.

To deal with this, we can do a range of phonics (sound to spelling) activities, focused on verb forms (e.g. partial dictations in which the verb endings are gapped, crossing out silent letters, correcting faulty transcripts, and so on). Input-enhancement techniques can also be used, for instance using a specific colour for each problematic verb form and its associated pronoun (e.g. in French je mange, tu manges, il/elle mange, ils/elles mangent). These provide students with an extra retrieval cue to help them remember.

Next, a word about quiet reading... As you read silently to yourself you cannot resist saying the words in your head. Try it now! When we see written language, we simultaneously hear it. By combining what we see with its mental, phonological activation we engage both visual and phonological processes to reinforce memory. This is an example dual coding. Quiet reading, reading while listening, or reading aloud, all help to build memory and therefore long-term proficiency. But the quiet reading will be far more useful if students already have a firm grasp of phonics. To put it rather snappily, our visual field says "shapes!," while our phonological loop says "sounds!" 

The phonological loop is also the area which “knows” the acceptable sounds which can appear together and at the starts and ends of words when we decode incoming speech (known as phonotactic rules). For example, in Spanish the combination /str/ never occurs at the start of a word. This means that a common mistake made by many Spanish native speakers is to add an extra ‘e’ before words beginning with /str/.

When we acquire our first language(s) we learn the acceptable sounds and sound combinations within the first few months of life. Not surprising perhaps that when phonotactic rules (many teachers will think of grammar rules here) for a first and second language are the same, words are more quickly detected by the phonological loop.

In other words, words which are easier to pronounce are learned more quickly. This means that cognates (words which mean, sound and look the same across two languages) are more easily remembered, when spoken, than other words, as the phonological loop recognises their similarity with the first language phonology.

Since words which are easier to pronounced are learned more quickly, internalising the phonotactics of a language is important. It therefore helps to explicitly teach the phonotactic rules of the target language and give plenty of opportunities to use connected speech right from the start.

This helps along the process of chunking words together, speeding up both acquisition and spoken fluency. Why? Fluent speech is about sequencing meaningful chunks (a process we called chaining). With this in mind, the common practice of teaching words in isolation is not the best way to develop fluency. Better to do individual, choral and interactive reading aloud activities, structured question-answer sequences, structured drills or any task requiring the repeated use of common chunks of language. Providing large amounts of comprehensible listening and reading input, flooded with the same chunks, also contributes to memory building, speeding up language processing and developing oral fluency.

Reference

Smith, S.P. & Conti, G (2021). Memory: What Every Language Teacher Should Know. Independently published. Available from Amazon.