Play it again Sam: on Effectiveness of Repetition.

Ewa Wasik

Ewa Wasik lectures at the Catholic University of Lublin, Poland.
E-mail: ewasik@lublin.mm.pl

Menu

Introduction
Preconceptions and misconceptions
The Dicto-Comp.
The Study
Conclusion

Introduction

This paper attempts to examine the role and effectiveness of repetition in vocabulary learning as applied in a particular type of controlled practice task called a dicto-comp. The notion of repetition, as used in this paper, is manifold and entails at least four different aspects of the phenomenon. Firstly, it may be understood as mere rote rehearsalof lexical items i.e. repeating them aloud. Secondly, it may be seen as equivalent to multiple exposureto words in comprehensible context (e.g. as in reading, listening passages) as well as repeated retrieval/recall of words from memory. Finally, repetition may refer to repeating the whole practice task itself as a means of language development. The goal of this paper is to investigate the factors that safeguard successful 'repetition' (as defined above) and the influence of repetition, if any, on the storage of lexical data in long-term memory. The data used in this paper comes from a study carried out on 6 Polish advanced learners of English and consists of written passages produced by the subjects in the course of a dicto-comp.

Preconceptions and misconceptions

Let me first clear up some of the misconceptions and review the existing preconceptions connected with the term of repetition.

1.1 Repetition vs frequency

Repetition, in extra-ELT terms, may be defined as "doing or experiencing something again or several times". In this sense repetition accommodates the notion of frequency. It's a well known fact about not as much learning as human behaviour in general that it takes more than just one encounter with a an object or a person item to memorize it. The more often you come across a an object or a person the better chance it stands of sinking in. That is why most of us will occasionally find ourselves remembering lexical chunks stripped of meaning and context such as ridiculous fragments of TV commercials or pop songs. Obviously, the reasoning of copy- and song- writers is based on the assumption that multiple exposure can improve or even secure memorization. Does the same apply to learning processes? To a great extent, yes. In order to learn a lexical item we have to notice it first i.e. acknowledge that it exists, and the more often we bump into it the more likely it is to be noticed. However, mere "(…)awareness, even for a sustained period of time does not necessarily lead to memory (…)", Stevick (1976:7). In order to be learned the item has to be memorized in all its complexity [form and meaning]. The role of memorizing is particularly salient in learning lexis as, by and large, learning vocabulary boils down to committing lists of lexical items to memory - "(…) learning (vocabulary) is remembering (…)Thornbury (2002:23). The task of the language instructors is, then, to come up with devices [i.e. design such tasks] that will, in the first place secure multiple encounters with lexical material, and then, convert the 'awareness' into 'knowledge'- understood here as 'remembering the item and being ready to recognize or use it on cue'. Therefore, the big question I want to address in this paper is if repetition can be used to teach vocabulary effectively, and if so, how to use it so as to make the learned material linger in the learner's memory.

1.2 Rote rehearsal

When we think of repetition one of the first images to cross our minds is that of a bored class brainlessly repeating aloud after the T. Drills have a history as long as language teaching itself. They happened to play the key role in teaching practises of the first part of our century, when the better part of class time was devoted to never-ending rote rehearsal. The problem with rote rehearsal is, however, that is incomplete in that it focuses only on the form of lexical items and ignores the meaning. Vocabulary learning is not brainless parroting but a generative process which should involve emphasis on both form and meaning. Rote rehearsal lacks cognitive depth (Schmitt, 1997 in Kudo (p.7)) as it involves parrot-fashioned voicing of a string of sounds disregarding its other aspects such as the meaning, collocation, grammar etc.. Naturally, the merits of rote rehearsal for teaching psychomotor skills cannot be overlooked but in order to be memorized as a whole, a word has to leave a "mental trace" i.e. involve some cognitive activity (Henriksen, 1995). Further evidence to support the claim that rote rehearsal is not particularly effective for vocabulary learning comes from word-learning experiments in which subjects were asked to say words aloud when learning them (Tell and Ferguson, 1974:349 in Stevick, 1976:31). The results showed that echoing or voicing the target item disrupted the whole memorizing process. On the whole, subjects who were asked to vocalize the items performed worse in terms of words recall that those who read their words silently or heard them as read by other speakers. Parrot-fashioned repetition of language forms did not do any good, on the contrary, it confused the learners and disturbed memorization. For the above reasons, repetition understood as rote rehearsal will be ruled out from this discussion and study.

1.3 Repeated exposure and retrieval (recall).

Lexical items are complex entities. They are loaded with inherent, arbitrary information of different origin. To learn a word we have to master its grammar, semantics, phonology etc. It comes as no surprise then that it takes more than just a single encounter with the word to successfully learn in its entirety. In this sense repetition may improve the quality of language - including the quality and range of vocabulary. That is exactly why repeated exposure and retrieval of lexical items are so important. However, the introduction of these two elements as the axis of classroom tasks cannot be haphazard. There is a number of principles that should be observed, some of which stem from studies in human memory and in second language acquisition.

1.4 Factors that safeguard successful memorization through repetition

Nation (2001: 74-80) mentions three elements essential for successful memorization through repeated exposure and recall. They have to do with: time, number and type of repetition. First of all, spaced repetition is more effective than massed repetition. From the point of view of memory storage, drawing learners' attention to a given item repeatedly for short periods of time over a longer time is better than studying a given item for a longer, uninterrupted period of time (Baddaley 1990, Bloom and Shuell,1981, Dempster, 1987 in Nation, 2001:76) What is more, the intervals between consecutive repetitions should become progressively longer (Pimsleur, 1967 in Nation, 2001:77). The longer the time the given lexical item has 'resided' in the learner's memory, the better chance it stands to stay there for good and be recalled.
The second element has to do with the precise number of repetitions needed to secure long-term retention in memory. The number of times the learners are confronted with the word through either just seeing or hearing it or being forced to recall it may have some bearing on how effectively the word will be remembered. As evidenced by research in this area (Kachroo, 1962, Crothes and Suppes, 1967, Tinkham, 1990 in Nation, 2001:81) the exact number of repetitions needed varies in individual learners and ranges from 5 to 20. The way the new item is presented to the learners for the first time carries some weight as well. Presenting both form and meaning renders best results for the first exposure to an item, whereas delaying presentation on the subsequent occasions and allowing learners to think for a moment before recognizing the word proves to be more effective for recall. The last factor, that can affect verbal memory- as pointed out by Stevick (1976:25) is cognitive depth i.e. how deeply the lexical item is processed. A meaningless string of sounds is not as likely to stick in memory as an item that was given some thought. In order to enhance retention, in the long run, we have to dip into word's meaning and it's meaning associations.
The above conditions have some important implications for the L2 classroom. If repetition is to be used as a vocabulary storage booster we have to make sure that the tasks we design comply with the rules outlined above. To recap, an ideal task [from repetition point of view] should 1.entail multiple exposure to a lexical item as well as its multiple recall, 2. be duly administered in time [i.e. involve spaced rather that massed repetition] and 3. compel the learner to reflect on the meaning of the item. To cap it all, such a well-suited task itself could then be redone on several occasions so as to provide the learners with enough opportunity of meeting the lexical material over and over.

1.5 Task repetition

There has been a surge of interest in the last couple of years in the effects of task repetition supported by ample research in this area (Bygate ,1996, Bygate, Skehan and Swain, 2001, Lynch and Maclean, 2001). It has been proven that repeating a task results in improving learners' performance in terms of overall complexity, fluency and / or accuracy (Bygate, 2001: 42-43). However, the bulk of these works focused on the benefits of repeating the task for overall language performance rather than on one aspect of language output e.g. a particular skill, lexis, grammar. As vocabulary is of particular interest to me I have decided to concentrate on the lexical aspects of the output of the task.

The Dicto-Comp.

The study of this paper is based on one type of task that happens to feature most of the above (section 1.4) elements. It is called a 'dicto-comp' and goes back to the 70-ties. As the name itself suggests, the dicto-comp is a combination of two types of activities: a dictation and a composition. The dicto- element, in which learners listen to a given passage several times is followed by the comp- element, in which they are asked to reproduce the text as accurately as possible in writing. In the course of the text reconstruction learners may refer to a kind of outline or prompts as a kind of reference. According to its creators (Riley, 1989) the dicto-comp serves many functions. As it involves "proceeding from imitation to improvisation" (Riley, 1989:238) it improves performance and the quality of the produced language. According to its creators, it may also raise learners' awareness of discourse structure, promote practising the language in the meaningful context, improve listening and writing skills and teach aspects of vocabulary and /or grammar. Because the dicto-comp is also a listening exercise, the question of learning vocabulary through listening should be mentioned. Some research has been done in this area (for a detailed discussion of this issue see Nation, 2001: 116-121) and it has been concluded that "learning vocabulary from spoken input is an effective means of vocabulary expansion" (Nation, 2001:121) and that being exposed to lexical items through listening turns out to work for long-term retention.

The Study

Participants
The task was carried out on 6 students enrolled in the FCE preparatory course in a private language school. Their level of proficiency was established on the basis of three tests: 1) a school placement test that was administered at the beginning of the course, 2) grammar-lexis test and 3). listening tests [both standard Cambridge Exam Practice test] that concluded the first semester. In the selection procedure participants with either very low or very high average scores were eliminated and 6 learners were picked up to take part in the study. The subjects were not familiar with the task prior to the experiment.

Data collection
The task involved listening to a passage read by the teacher three times. A short passage on eating habits was selected (FCE Nelson, 1996:47). Learners were instructed to relax and listen trying to concentrate on the meaning. No note-taking was allowed. After the first reading they were provided with an outline of the passage that served as a kind of a map or reference for later reconstruction. The second reading was followed by a short vocabulary presentation. At this stage participants were allowed to ask questions to dispel any doubts they had in relation with the lexical content of the passage. Their attention was drawn to a number of possibly problematic items .Then, the passage was read for the 3^rd time and thereafter subjects were required to reproduce the passage in writing as closely as possible. The task was performed under no fixed time limits. The whole procedure was repeated after a period of 5 days.

Questions
The goal of the study was to answer the following questions: 1) How does the learners' first and second performance differ in terms of lexical content? 2) What are the differences e.g. what aspects of lexis will be the most / least affected? 3) How will the rate of learners 'lexical innovation" i.e. coming up with words not found in the original passage be altered ?

Results
The data obtained in this study comprises passages reproduced by the learners on two separate occasions. The following criteria were chosen to measure the subjects' lexical performance (adopted from (Bygate, 1996), Thornbury 2002):
1. lexical repertoire i.e. range and amount of lexical items used - the extent to which learners are able to reproduce the passage using as many of the original lexical items as possible, as indicated by two numbers:
a) lexical density: the total of lexical items used relative to the total of items available for usage *
b) lexical sophistication: variety of different lexical items typical of advanced vocabulary [the following measures were chosen to indicate sophistication: the total of collocations, phrasal verbs, idioms and chunks, cohesive devices, other**] as indicated by total of items used relative to the total of items available for usage (Table 2)
*lexical items used = content words used [articles, pronouns, conjunctions, auxiliaries, verb 'be' in a predicative function were left out of count]
** infrequent words e.g. to combine, to face, sensible etc.

***When looking at the selected criteria an objection might be raised against opting for two separate criteria for lexical innovation and imprecision as these two may be seen as overlapping in that when the subjects were not able to use the original item they replaced it with a new one. However, on the account of the fact that not all instances of lexical innovation are examples of imprecision these two measures have been investigated separately.

The assumption was that task repetition, as well as the repetition elements inherent to the task itself will boost subjects' lexical performance. This could be indicated by:
- possible enhancement in lexical repertoire i.e. greater lexical density and sophistication at Time 2,
- possible increase in accuracy at Time 2
- lower lexical innovation i.e. greater precision of expression
The data obtained is illustrated in the Tables 1, 2 below.

Table 1.

Results
The first prediction concerned the lexical repertoire i.e. sheer number and variety of the items used. As shown in Table 1, a), b) mean growth of 8% of items per capita was observed in both lexical density and sophistication. This means that on the second occasion subjects expressed themselves by means of more words and, on top of that, they opted for the more refined items. On the other hand, it has to be noted that the growth rate - 8% for both measures, is not particularly impressive.
Another set of features concerned the learners' ability to use lexis accurately. Contrary to my expectations, repeating the task did not have any impact on the number of errors in the lexical items produced. This number remained relatively low - 13% on both occasions. However, the amount of the imprecisely used items increased by 6 items per capita which could point to a decrease in accuracy. Two arguments may be put forward against this claim. Firstly, the rate of lexical imprecision turned out to be particularly susceptible to individual variations as compared to other rates (Table 1, 3, e)). Secondly, the higher proportion of imprecisely used words may be due to the increase in lexical density and sophistication. In short, it may well be that the more words on the second occasion the learners used the more imprecisely they did it. The last prediction concerned the number of 'new' items used. A general tendency to use fewer 'new' items was observed but the exact drop varied from 1 to 18 items per learner. An increase in lexical innovation was noted in case of one learner.

Let me now address the second question concerning the types of lexis that were the most or the least affected. The table below illustrates the number of different types of 'sophisticated' items used on both occasions.

Table 2

A clear pattern emerges from the data in the table above - at Time 2 the total of different types of sophisticated items used by the subjects has gone up for all the categories selected except for frequency expressions. The only plausible explanation that can be offered is that learners are taught this particular group of words relatively early on in the course of their language education and, perhaps, as a result, they tend to take them for granted. This leads to their paying more attention to the more 'advanced' words and overlooking the frequency expressions. The most notable mean growth rate was noted for phrasal verbs (27%). However, this number should be approached with caution as it does not reflect the performance of all the subjects (students 5 and 6 failed to use any phrasal verbs on both occurrences of the task).
The last batch of data is connected with the imprecision of expression entertained by learners when reconstructing the text. Table 3 below features a selection of the most interesting and common expressions used.

Table 3

When faced with an item they could not reconstruct accurately the learners dealt with the situation by either supplying a synonymous item, usually a less sophisticated and more frequent one e.g. 'choose' for 'go for', or by paraphrasing it in a more complex way e.g. 'flavours and cuisines' instead of 'different sorts of cooking' The former way of bridging the lexical gap was by far more common than the latter. The number of lexical imprecision increased slightly at Time 2.

Conclusion

The goal of this study was to look into the influence of repetition [repeated exposure, repeated retrieval and finally repeating the whole procedure] on the learners lexical performance i.e. measure the vocabulary development that has taken place as a result of repeating the task. Two conclusions can be drawn from the results. The subjects' lexical performance in terms of both number and the range of items used improved at Time 2 whereas the initial high accuracy rate was left intact. As far as different types of sophisticated lexical items are concerned, the subjects performed better in most of the selected categories

with the exception of frequency expressions. Finally, in the second reconstruction of the total of new items used was higher. These findings are consistent with the results of earlier research in this area (Bygate ,1996, Bygate, Skehan and Swain, 2001, Lynch and Maclean, 2001).

When discussing the results of the study, the question of subjects' response and reaction to the dicto-comp should be brought up. Though slightly taken aback when asked to repeat the task, after the task was concluded, the learners unanimously expressed a wish to do it again in future. In their opinion, the dicto-comp helped them to remember "bigger fragments of language". Another comment was connected with the fear of making the same mistakes on both occasions. This could be easily prevented by correcting the passages after Time 1 and drawing subjects attention to possible problems.
Despite all the benefits for the development of learners' lexicon, the role of repetition cannot be overrated and should be approached as one of many vocabulary learning techniques rather than 'a miracle cure' to vocabulary learning problems.
As regards classroom application of repetition, teachers might be wary of implementing for fear of boring their students - all in all doing the same thing again and again may be dull. However, we cannot forget that repetition constitutes a vital part of our lives - and for this reason it could be transferred to language classrooms.

References

Bygate, M., Skehan, P. and Swain, M. (2001) Researching Pedagogic Tasks Second Language Learning, Teaching and Testing. Longman.
Bygate, M. (2001) Effects of task repetition on the structure and control of oral language. In Bygate, M., Skehan, P. and Swain, M. (eds) Researching Pedagogic Tasks Second Language Learning, Teaching and Testing. Edinburgh: Longman
Bygate, M. (1996) Effects of task repetition: appraising the developing language of learners. In Willis, D. & J.Willis (eds.)
Challenge and Change in Language Teaching. London: Heinemann.
Swain, M. and S. Lapkin, (2001) Focus on form through collaborative dialogue: Exploring task effects. In Bygate, M.,
Skehan, P. and Swain, M. (eds) Researching Pedagogic Tasks Second Language Learning, Teaching and Testing.
Edinburgh: Longman.
Lynch, T. and J. Maclean, (2001) ' A case of exercising': Effects of immediate task repetition on learners performance. In Bygate, M., Skehan, P. and Swain, M. (eds) Researching Pedagogic Tasks Second Language Learning, Teaching and Testing. Edinburgh: Longman.Nation, I. S. P. (2001) Learning Vocabulary in Another Language. Cambridge University Press.
Stevick, E. W., (1976) Memory, Meaning and Method. Newbury House Publishers.
Keh, C., L. (1989) How I Use the Dicto-comp. In English Teaching Forum 27, pp.238-240.
Riley, P. M. (1972) The dicto-comp. In English Teaching Forum, 10, 1, pp.21-23.
Henriksen B. (1995) What does it mean to know a word? In Sprogforum, 3, pp.12-18.
Thornbury, S. (2002) How to teach vocabulary? Longman.

Please check the English for Teachers course at Pilgrims website.
Please check the What's New in Language Teaching course at Pilgrims website.