Aspects of memory in secondary school language learners
International Studies Bulletin 4, 2, 161-172, 1979
Presented at the colloque international sur l'acquisition d'une langue étrangère: perspectives de recherche, Université de Paris-Vincennes, April, 1979
During the nineteen-sixties memory became an important area of concern in psychology. To the psychologist memory included any storage or processing of information by the mind. For mental processes take time and therefore they all rely on memory in one way or another, whether it is fleeting memory for someone's face as it flashes past on a train, slightly longer memory for the sentence you have just read or ability to remember what Pit Corder said in 1967. Obviously there are many kinds of memory and many competing hypotheses about how these are organised. One useful way of looking at memory is that proposed by Craik (Craik, 1973) called 'levels of processing'. The idea behind this is that the mind can choose what to do with the information it receives; in what ways it should be processed. It may decide that the information is worth keeping only for a short time and that it should be processed only in terms of sounds; this superficial level of processing corresponds to short term memory. Or the mind may decide to store the information indefinitely and to squeeze out all its semantic meaning; this corresponds to long-term memory. Within this model of memory the deeper the level of processing the more language is involved. A short-term memory task involves language only minimally; a long-term memory task uses language extensively. One of the interesting questions has been: if this is true of the first language, how does it apply to a second language? Do we have to find new memory processes at each level of processing for the new language or can we use the ones that we have developed using the medium of our first language?
In earlier work with adult foreign learners of English (Cook, 1977), the conclusion was reached that the more superficial levels of memory processing could be transferred to a second language more readily than the deeper levels; short term memory was similar in a second language while long-term memory was rather different. The experiments to be described here extend this line of research to children learning French in English schools. The reasons for switching were threefold. One is the belief that English gets more than its fair share of attention; with a few honourable exceptions the bulk of research into second language learners to date has concerned English and this leads to the danger that it describes not universal characteristics of second language learning but simply the peculiarities of learning English. A second reason was the apparent lack of research into foreign language learning in English schools; one can speculate that this may be one of the causes for the alleged crisis in foreign language teaching in Britain. The third reason was simply that children are not adults. The adult learner comes to second language learning with memory processes already at their most mature; children, however, are still developing them. It is then interesting to see how children's memory processes develop in their second language and how this may be related to their first language.
The first experiment looked at the learning of French in an East London comprehensive school. The aim was to establish the children's capacity for words and for digits in both English and French. The children were tested in their normal French classes. First I read them strings of English numbers, starting short "7159" and getting up to 9 digits. After each string the children had to write down as much as they could remember. Then I did the same with list of monosyllabic English words starting with "girl street roof" and getting up to 8 words. Then their French teacher took over and read them first strings of French digits and then strings of French words. The French words had been chosen to be within the vocabulary of the lowest class tested, at least in the opinion of their teachers. They were in fact "bain, beurre, femme, lait, main, oeuf, pont, sucre, verre". The method of scoring the results was to establish the child's maximum span for each list. The results are given in Table A for classes 2 to It, a total of 117 children.
What do these results show?
1) They seem to show a slight increase in digit span for French digits of 0.7 between class 2 and class 4. They also show some increase in the span for French words; 0.5 by class 3 but for some reason a fall, probably not significant, in class 4.
2) They remember less words than they do digits, both in English and in French. Adult natives have better spans for digits than for words and this seems to be true for the foreign language as well.
3) What about the relationship to memory processes in their first language? First of all this shows that they are transferring a substantial amount of their capacity to their new language; they remember about 2.6 digits less in French than in English and about 1.4 words less. Again similar to results with adults. Secondly, however, there is an intriguing point that the difference between their capacity for digits in English and French comes out as almost the same for class 4 as for class 2. In other words their memory in English has expanded at the same rate as their memory in French, namely 0.7 digits. In memory for words the difference between English and French does narrow slightly from 1.6 to 1.3 but of course the difference between classes was not here a straightforward increase as class 3 for some reason were better than class 4 at word span in both English and French.
Thus this experiment tells us something about a particular aspect of memory in the child learner of French. However, it only dealt with a fairly superficial level of memory in which language was barely involved. The second experiment dealt with short-term memory for sentences. The subjects that were tested were again children learning French in an English comprehensive school, this time 105 in number and ranging from class 1 to class 5. The same experimental technique was used in which their normal class teacher read them sentences in French which increased from four words to 16 words in length and varied in grammatical structure at each length. The test sentences are given in an appendix.
The results are shown in Fig. A. This time they are expressed in terms of the percentages of correct answers at each length of sentence. What do these results show?
1) First the differences between the five classes. We find that class 1 could manage a sentence with four words fairly well but could not cope at all with longer sentences. Classes 2 and 3 managed very well with four-word sentences, not so well with eight-word sentences and faded out altogether after that. Classes 4 and 5 did well with four-word sentences, not so well with eight-word sentences, badly with twelve-word sentences and very poorly with sixteen-word sentences. These results seem to show that there is improvement in capacity for sentences between class 1 and class 5 but that this improvement is considerably less than one might have expected. After five years of French the children were only able to cope with about half the eight-word sentences and were still hardly able to perform it all with longer sentences. Indeed the improvement in capacity seems to come to a halt in the 4th class.
2) If we compare this to native adults what do we find? Unfortunately the only experiment to measure memory for syntax in this way is an unpublished one of my own which deals with adults speaking English as a native language so they are not really comparable. Nevertheless, to give some idea, the results for native adults are superimposed in a dotted line on Fig. A. If we adopt a level of 50 per cent as acceptable performance then we find that the oldest children achieve this with eight-word sentences, the adults with fourteen-word sentences. In short there is a difference of 6 words. This shows that there is some contribution from the syntax of the sentence to their capacity to remember; they do remember sentences better than isolated words. However, their capacity is way below what might be expected from them in their native language.
Let us now try to put this kind of experiment in a wider context. To do this we need to take a slightly broader view of second language learning in a classroom setting. One way of doing this is to imagine oneself in the position of a learner attending his or her first lesson in a foreign language. When they enter a classroom door, learners bring something in with them - their minds. They already have minds that are furnished in all sorts of ways with their experience of the world, their personalities, their processes of thought and so on. So one important aspect of second language learning is the learner's mind, his mental make-up Secondly once they are through the door of the classroom they find themselves taking part in a particular situation; they have to find ways of coping with the demands that this new situation makes of them, they will try to satisfy these demands either by drawing on the resources that they bring with them or by devising new techniques appropriate to the new situation. Let us call these the situational strategies that the learner uses in the language learning situation. Finally in a classroom situation what happens is not determined solely by the learner, instead it is largely under the control of a teacher. Let us call what the teacher does teaching techniques, the ways that he overtly tries to control the situation and covertly gets the student to learn. Let us now look at each of these three components more closely.
First mental make-up. In itself this will include a variety of aspects. One of them for instance is the communicative competence that the learner possesses in his first language, not just the grammatical and phonological competence that interference theory tends to have concentrated on but all the knowledge that is necessary for the native speaker to take part in a conversation including knowledge of functions and intentions, ways of turn-taking, discourse rules, style-switching, and so on. The learner does not have just a linguistic competence in his native language which may interact with his learning of a second language; instead he has an LI communicative competence which can relate itself in all sorts of ways to his second language learning.
The second aspect of the learner's make-up that determines his situational strategies and indeed his success in second language learning is his motivation and attitude; the reasons why he wants to learn the foreign language and his attitudes to the target culture based upon the social relationship between his own culture and the target culture are all as we know crucial elements in second language learning, A third aspect which has been almost unexplored but which impresses me more and more is the students' expectations of what happens in the classroom. The student's conscious ideas about language teaching and teaching in general are highly relevant to his classroom performance. If he expects the teacher to use translation and grammatical explanation he will be disappointed by a strictly audiolingual method so much so that he may devise all sorts of ways round this. I remember discovering students in an audiovisual class of mine who were making notes in their own invented spelling of all the new words they were encountering. Their expectations that the classroom would be based on the written language not being fulfilled, they were trying their own way of making the class live up to what they wanted. This is not to say of course that the student is necessarily right in his expectations of the classroom but simply that right or wrong they are part of the mental make-up of the learner and influence his situational strategies. As well as these two there are obviously many other aspects of the learner's mind that are important to second language learning.
To come back to our experiments, one aspect that is often overlooked is the memory processes that the learner uses for thinking and for processing language. An adult for instance can process incoming information at different levels of memory, each with certain capacities and certain ways of dealing with the Information. We take it for granted that the adult learner has an already well-developed cognitive system when he starts a second language. The implications of this for second language learning are fascinating; as we have seen one question is how these cognitive systems can be transferred to a second language; does the learner use the same processes with the same capacity as in his L1? Another question is about the causal relationship between cognition and language learning. People dealing with first language acquisition often assume today that, despite Chomsky, the bulk of language acquisition can be explained in terms of social interaction and cognitive development (Cook 1979). But where does that leave second language learning? So working with child learners has an extra interest since we can no longer assume that they bring with fully adult memory processes. The particular stage of cognitive development of the child may interact in all sorts of ways with learning. Developmental work with children learning a second language can see how a particular stage of mental development correlates with particular aspects of language learning. This was part of the rationale for the first experiment, however crude the actual method lying these processes actually was. A child of 12, regardless of his level of ability in the new language, may show certain characteristics in the second language because of the stage of mental development he has achieved. Inasmuch as language is the product of level and capacity then it must have some interaction with second language learning.
The second major component is situational strategies; how does the learner cope with the learning situation both consciously and unconsciously? This can be divided into at least three aspects, although the divisions are by no means watertight. First comes communication strategies, in other words the kind of area that has started to be explored in second language learning by Evelyn Hatch and her associates (Hatch, 1978). These are the foreigner's attempts to hold his own in conversation; he has to be able to get agreement on a topic of conversation; he has to be able to check what the other person says, he has to be able to show his reactions to what has been said; he has to know how to take turns in the conversation. Obviously because of the structured type of communication in the classroom the ones that can be adopted by the learner are going to be circumscribed and will be more the choice of the teacher than his own. The second division is learning strategies, both the kind of conscious strategy revealed in the studies of the Good Language Learner and the less conscious strategies of transfer, simplification and so on familiar from the literature. Last but not least are speaking and listening strategies. By these I mean the actual psychological processes involved in the production and understanding of speech. To be able to understand a sentence in the new language the learner has to have a strategy for comprehending it; to produce an utterance again the learner needs to have ways of production such as the use of undigested formulaic utterances or the avoidance of structures or lexical items for one reason or another.
One factor determining the situation strategy that the learner employs is his mental make-up. Hence among other things he will be transferring various aspects of his communicative competence to the task of communicating in the new language; these second language communication strategies do not come out of thin air but are adaptations and improvisations from the learner's own experience and from his expectation of what it is like to use a second language. Similarly the learning strategies that he employs are based on his attitudes towards learning and his preconceptions about the nature of language. If he thinks that language is chiefly words he wll think of language learning as a question of learning words and hence will have a strategy of looking for words in the classroom. The use of formulas and avoidance may also be connected with various aspects of his mental make-up. The point about the present experiments is that they show some of the constraints that are in operation in using speaking and listening strategies in the second language. They reveal some of the limitations on processing at different ages and levels in the learning of the foreign language. It shows that for instance in class 2 the sentence length that will be taken in by pupils is still only four words and that very few of the pupils will take in a sentence that is eight words long. Even by class 5 pupils are only likely to be able to cope with eight word sentences about half the time. Of course this is making a leap from the present experiment with only a few sentences to their general ability and obviously a broader experiment would really be needed to establish this better. But there seems no reason to doubt that the capacity for words in sentences is roughly of the dimensions that were found.
This increase in syntactic capacity may be one of the important developmental changes in learning a second language, or for that matter a first. The reason for the developmental order that some people have found for syntax in second language acquisition or indeed for the order of difficulty of particular morphemes may be caused by some underlying explanation such as the gradual expansion of the capacity to handle syntax. It is vital to know some of the limits that particular learners have in their processing of a second language and how this may be related to their first language. The results above for instance suggested, if not proved, that the capacity for digits expands by the same amount in the second language as it does in the first. Hence an important thing to know about the learner may be his span in his own language, another way perhaps of putting the point that many teachers have made to me, namely that one's ability to speak a foreign language is related to one's ability to speak one's first language. But equally it may be vital to know the learner's capacity for various types of material in the second language as this develops.
While the first two parts of this framework for looking at second language learning apply equally inside and outside the classroom, the third is aimed specifically at the learner. In a classroom situation the learner is not free to use his situational strategies as he wants; he has to do so under the direction of a teacher. Obviously one is not going to be naive enough to think that just because the teacher says implicitly or explicitly "We're all going to learn language in this way" that the pupils automatically follow his lead. But on the other hand the teacher's choice of techniques curtails the pupil's choice of strategies. It seems to me fairly obvious that the teaching technique that will be most successful is that which goes along best with the pupil's own strategies which in turn arise out of his mental make-up. There is no point in devising remarkable teaching techniques if they do not also fit in with learning strategies. Take grammatical explanation for instance. This implies that the learner is seeking a conscious awareness of the language system, indeed a plausible strategy for the good language learner, even if we can have doubts about its effectiveness. This in turn implies that the learner has certain preconceptions that language learning is a content subject like history and that language is a rule-governed system of a certain kind. Seen in these terms one can ask straightforward factual questions about whether these implications are true: do many learners believe in learning a grammatical system consciously? What beliefs about language or language learning go with this? Even if we think that the learner's strategy and his beliefs are misguided they are still part of the learning process.
Once the teacher knows they exist he can either go some way towards exploiting them, if he thinks they are beneficial, or try to counteract them if he thinks they are harmful.
What do our two experiments provide for the teacher? The most simple application is the light they show on capacity. The teacher can safely use techniques with let us say class 4 that rely on sentences with four words; if he uses sentences with eight he stands a fifty-fifty chance of them being able to cope; with longer sentences few of them will be able to manage. Thus any technique he uses has to be looked at in the light of the capacity that the students are bringing to it. One obvious example is length of sentences whether used for repetition in dialogues or for production in structure drills or what have you. The teacher should at least be aware of the limits of his pupils' capacity. This is not to say that he should stay always within them but that he should be aware when he is going beyond their limits. Of course teachers already do this in a subjective way. What one hopes eventually to provide is a more precise idea of how these capacity limits develop in learners.
So, to sum up, these results illustrate only one small aspect of the many-faceted process of learning a second language. Feyerabend (1975) suggests that science needs to pursue multiple approaches rather than concentrating on a single research paradigm. It is dangerous to try to explain everything in L2 learning in terms of a single concept such as interlanguage or interference or in terms of such binary oppositions as competence and performance, acquisition and learning, implicit and explicit. To account adequately for second language learning we need evidence from many sources, not just the syntactic mistakes found in the sentences that learners produce. Memory is one aspect of this and experiments such as the ones that have been described go some small way towards giving us evidence that will contribute to our knowledge of the processes that underlie second language learning.
I am grateful to the staff of Caterham High School, Redbridge, and of St. Benedict’s School, Colchester, for allowing me to carry out these experiments and to their pupils for their cooperation.
Cook, V.J. (1977), ‘Cognitive processes in second language learning’. IRAL, 15/1, 1-20 online version
Cook, V.J. (1979) Young Children and Language. Edward Arnold.
Craik, F.I.M (1973), ‘A 'levels of analysis' view of memory’. In P. Pliner, L. Kramer and T. Alloway (eds.) Communication and Affect. Academic Press.
Feyerabend, P. (1975), Against Method. Verso Editions.
Hatch, E. (ed) (1978), Second Language Acquisition. Newbury House.
Sentences used in Experiment 2.
1. Le train est long.
2. Il voit Jean peu.
3. La grande clé est dans mon sac rouge.
4. Jacques tient le chien blanc par la tête.
5. Georges voit Anne dans la rue le soir.
6. Le gros chat noir et blanc mange le rat jaune et brun.
7. Mon beau frère Charles fume sa pipe rouge dans une chaise bleue.
8. Sa mère vient voir ma femme tous les jours à sept heures.
9. La jeune fille blonde de ma tante Anne boit un verre de vin sec très frais.
10. La belle soeur de mon vieux père met sa robe de laine verte sur la table.
11. Le vieux maire offre du vin aux gens de la ville à six heures du soir.