Computer games and foreign-language vocabulary learning

Com put e r ga m e s a nd fore ignla ngua ge voc a bula ry le a rning Rolf Palmberg This article reports on the design and results of an experiment the aim of which was to test the effects of playing with computer games on the learning of English vocabulary by young, elementary-level, Swedishspeaking children. The article also discusses the role of computergames as teaching aids in the foreign-language classroom. Introduction and aim The experiment Subjects Of all the various criteria used when selecting vocabulary for presentation to learners of second or foreign languages, the question of language needs is of special importance when it comes to beginners. Ashton-Warner, for example, while teaching English as a second language to 5-year-old Maori children, soon found that the textbooks she was supposed to use were completely wrong for most of them in terms of culture and motivation. She then started to ask the children to make lists of words that they wanted to be able to read and write. Using such word lists as their first working material, the children rapidly built up a large vocabulary which was relevant to them (Ashton-Warner 1980). Applying the principle of language needs to vocabulary selection means, as described in Nation ( 1983: 16), that the vocabulary selected must satisfy all important everyday needs of communication, or, more specifically, the learners’ personal needs, social needs, thinking needs, labelling needs, and classroom and teaching needs. However, in classroom situations where the language taught has the status of a foreign (as opposed to second) language, the individual interests of the learners, rather than communicative needs proper, often determine what words they actually learn. Thus, in a study requiring Swedish-speaking beginners of English to write down as many English words as they could think of that began with given letters, it was found that the words that were immediately accessible to the learners were, in general, words that had been repeated regularly in class or introduced recently in the class textbook. Any variability in their on the other hand, could largely be vocabulary-production patterns, explained by the individual interests of the learners, especially rock music, but also computers and computer adventure games (Palmberg 1987c). Using computer games and learner interests as a starting-point, reports on an experiment carried out with the aim of studying children’s interaction with computers on their learning vocabulary. this article the effects of of foreign Two Swedish-speaking boys in Finland, aged 9 and 11, participated in the experiment. When the experiment started in December 1986, they had ELT Journal Volume 42/4 October 1988 © Oxford University Press 1988 247 both been exposed to English in the form of television programmes and popular music. In addition to this, the 11-year old boy had had three and a half months of English studies at school (two 45-minute lessons per week). The vocabulary that he had been taught at school comprised some 180 words, including the numbers l-20, the personal and demonstrative pronouns, and the most common interrogative pronouns. He had also been taught the regular plural forms of nouns, the definite and indefinite articles, and the most common auxiliaries. The computer game 248 In order to conform as much as possible to the learners’ young age and general interests, a computer adventure game, Pirate Cove (Schulak 1986a), was selected for the experiment. In Pirate Cove, the player is supposed to find a treasure chest hidden on one of ten islands shown on a map. The locations of the islands change on the map for each game, and along the way, the player is offered several options. She or he may go on any island, search native huts, swamps or sunken ships, or may meet with unexpected difficulties, such as severe storms or the Sea Monster. The program is basically a text-adventure game, and in order to take full advantage of its options, it is essential for the player to master the vocabulary used in the program. The vocabulary of Pirate Cove (from now on referred to as ‘target vocabulary’) consists of 118 different words, listed below in their dictionary forms: back can east fast galleon damage a/n captain eight Barbary again defeat gem chest ahead enough fight go do bay beach cliff alligator exotic dock find beautiful coast along doubloon Bluebeard comb and compass another but continue any at cove away current cutlass harbour have high hook hut if inlet interest inventory island it jewel jewelry journey jungle key lagoon lair land list location of old on open other over parrot piece pirate press read return sail sand sea search see ship sift silver skeleton snake some south spice start storm stream swamp Rolf Palmberg the through to treasure map may mermaid monster must native navigation north not nothing water YOU your way west will win without wood wreck Of these, eight, list, and old were the only content words that had been taught to the 11-year old at school, whereas 66 of the words can be found in the vocabulary list of the 1100 English words which, according to the recommendations made by the Finnish National Board of General Education, constitute the minimal basic vocabulary to be actively known by pupils leaving comprehensive school at the age of 16 (Engelska 1978). (From now on, the 1100 words will be referred to as FNBGE words.) Procedure Results and discussion The experiment was carried out in three phases. In Phase One, the testees were introduced to Pirate Cove and invited to play the game for threequarters of an hour. The experimenter acted as an interpreter throughout the session, providing each new word appearing on the screen with an appropriate Swedish equivalent. When the words reappeared, he checked whether the testees had understood their meanings correctly, making any necessary corrections. Phase Two was conducted one month later, without prior notice. The time allotted was the same, three-quarters of an hour. The programme had not been available to the testees between the two sessions. During this session, the testees were asked by the experimenter to provide any words or text displayed on the screen with Swedish translations or explanations. As in Phase One, the large majority of the target vocabulary appeared up to 10 or 15 times on the screen in various contexts, irrespective of the game options chosen by the testees. Furthermore, the testees were encouraged to discuss freely with one another, and to ask for help whenever needed. Phase Three took place one month after Phase Two. Like Phase Two, it was conducted without prior notice, and without the testees being allowed any practice with the computer programme between the two sessions. For this session, however, instead of working with the computer, the testees were given a list of 50 words selected from the target vocabulary. They were asked to give each word on the list a Swedish translation, or, that failing, to describe any associations triggered off by the words. To make the task as easy and unambiguous as possible, only concrete nouns were selected. The following words were included in the test list: beach, coast, harbour, island, key, map, piece, sand, sea, ship, silver, storm, water, way, wood, alligator, bay, bluebeard, captain, chest, cliff, compass, cove, cutlass, dock, doubloon, galleon, gem, hook, hut, inlet, interest, inventory, jewel, journey, jungle, lagoon, lair, location, mermaid, monster, navigation, parrot, pirate, skeleton, snake, spice, stream, swamp, and treasure. Of these, only the first 15 were FNBGE words. Judging from their performance during Phase One, the testees quickly learned how to read the map and how to navigate on the sea. At the end of the session they also knew how to search specific areas and were able to list what items they had gathered on their journey. Furthermore, they could provide the large majority of the target vocabulary with acceptable Swedish renderings. In Phase Two, it was evident from the very beginning of the session that the testees had a perfectly adequate working knowledge of the required procedure for playing the game. They were fully in command of how the game options were related to one another, and which actions contributed most efficiently to which sub-goals. As for their vocabulary knowledge, they seemed to have forgotten the dictionary meaning of about half the target vocabulary when asked for Swedish translations; words that they did remember correctly were often words that share formal similarities with Computergames and vocabulary learning 249 their Swedish equivalents (e.g. alligator, compass and skeleton), but also others (e.g. fight and spice). In a recent paper, Bialystok and Sharwood Smith illustrate by means of a library analogy the difference between the concepts ofknowledge and control. According to them, knowing what books a specific library contains plus access to the library classification system (i.e. knowledge) must be distinguished from retrieving desired information from the books at a given time. Such a retrieval process (i.e. control) involves the procedure for retrieving information as well as the speed or efficiency with which the library user can put the procedure into operation (Bialystok and Sharwood Smith 1985:105). Thus, in the terms of Bialystok and Sharwood Smith, it was obvious from the performance of the testees that they had gained knowledge of the computer game during Phase One, and were now working on gaining control. Most of the words that they failed to provide with Swedish translations they could nevertheless describe as being either names of islands, specific instructions or information, or valuable or worthless possessions. However, having been reminded of the Swedish equivalents of such words, the testees easily reactivated them in their mental lexicons and had little difficulty in transferring their existing pragmatic knowledge into receptive vocabulary knowledge if and when the words reappeared during the session. Having received the word list at the beginning of Phase Three, the testees immediately recognized where the vocabulary derived from. In the discussion following, they also managed to assign correct translation equivalents to 35 of the 50 words in the list. Another ten words were attempted, whereas five words were left totally although less successfully, unattempted. As in Phase Two, the testees easily recognized all but one of the words that share formal similarities with their Swedish equivalents, i.e. sand (interpreted as ‘strand’, ‘beach’). E xamples of correctly recognized words with no similarity to their Swedish counterparts were cutlass, hook, island, key, map, mermaid, snake, spice, swamp and wood. Partly correct associations were assigned to three of the words, i.e. gem, lagoon and treasure, in so far as the testees recognized them as being ‘some kind of jewelry, probably a necklace’ (for gem) and ‘parts of names of islands’ (for lagoon and treasure), respectively. Less successfully rendered words were beach (interpreted as ‘nothing’), lair (‘segla’, ‘to sail’), parrot ‘slåss’, ‘to fight’), inlet (‘inget’, ‘piece of jewellery, jewel’), way (‘skatt’, ‘treasure’), and sea (‘smycke’, by (interpreted as ‘segla’, ‘to sail’, by one testee and as ‘skatt’, ‘treasure’, the other). The five totally unattempted words, finally, were harbour, hut, journey, location, and navigation. As demonstrated elsewhere (Palmberg 1987a, 1987b), Swedish-speaking learners faced with the task of inferring the meaning of text passages in English are, irrespective of their level of English proficiency, much helped by the formal similarities that exist in the two languages. Not surprisingly, therefore, the testees in this experiment had little difficulty with such words in the word list. They also managed fairly well in their attempts to assign correct meaning to the remaining words in the list, which is worth noting for at least two reasons: firstly, considering the young age of the testees and their very elementary (or, rather, virtually non-existent) knowledge of English, and secondly, considering that recognizing and understanding isolated words is generally thought to be a much more difficult task than inferring the meaning of words contained in reading passages. Or, to quote 250 Rolf Palmberg Crow: ‘any decontextualized activity is always beyond the bounds of primarily receptive vocabulary work’ (1986:246). The testees, finally, seemed to be helped a lot by having correctly assumed that all words in the list belonged to one specific, topically organized group of words, that of pirates and treasure hunting. Furthermore, while working on the word list, they explicitly restricted all suggestions for possible word meanings to ones which, although plausible from a topic point-of-view, could not be correct ‘because they had not appeared in the computer game’. The only exception to this assumed criterion was the introduction of ‘necklace’ as the possible meaning of gem. Summery and conclusion With the aid of Pirate Cove, a computer game intended for native, Englishspeaking children, the testees learned in only two 45-minute lessons to recognize and interpret correctly the major part of the vocabulary used in the game, and, while playing, they demonstrated ample control of the program pragmatics. One month later they were still able to demonstrate a fairly good receptive knowledge of a sample vocabulary on pirates and treasure hunting selected from Pirate Cove and presented to them out of context. These findings indicate that computer games, and especially motivating text-adventure games in a foreign language, constitute a good example of material that satisfies the criterion of language needs relevant to young learners of that language, and, at the same time, promotes vocabulary learning. The foreign-language teacher could therefore make use of carefully selected computer games for example in order to introduce new areas of vocabulary to his or her pupils, irrespective of their previous experience of the foreign language. As for the principles of vocabulary selection, it is obvious that the vocabulary of the computer game(s) used must be appropriate as far as the teaching objectives are concerned. In cases where the program listings are available to the teacher (for the listings of Pirate Cove, see Schulak 1986b), vocabulary selection will constitute no problem, since he or she could easily increase or modify the vocabulary contained in the game(s) so as to conform to any teaching objectives, whether general or specific. Computer games such as Pirate Cove can be used as teaching aids in a number of different ways. They can be used primarily on a self-access basis, thus relieving pressure on classroom time and allowing pupils to work at their own pace. They can also be used in class, with the pupils working in groups of two or three on each computer set, and with the teacher supervising and providing help when needed. In such groups, every pupil could be assigned an individual task (in addition to that of the group), for example making lists of objects found in specific areas, or statistics on specific events in the game. Ready-made worksheets are very useful for this purpose, and the information thus collected could later be rewritten by the pupils (as homework or in class) into reports to be read out and discussed in class. receptive vocabulary activated in the pupils during the Any computer session can in this way be practised and reactivated in class and • gradually become part of their productive vocabulary. Received August 1987 Computergames and vocabulary learning 251 Palmberg, R. 1987c. ‘Patterns Ashton-Warner, S. 1980. Teacher. Chicago: Virago. Bialystok, E. and M. Sharwood Smith. 1985. ‘Interlanguage is not a state of mind: An evaluation of the construct for second-language acquisition.’ Applied Linguistics 6/2:101-117. Crow, J. 1986. ‘Receptive vocabulary acquisition for reading comprehension.’ Modern Language Journal 70/3:242-50. Engelska. Förslag till basmål och basstoff i grundskolan. Helsingfors 1978: Skolstyrelsen. Nation, I. S. I’. 1983. Teaching and Learning Vocabulary. Wellington: Victoria University. Palmberg, R. 1987a. ‘On lexical inferencing and the young foreign-language learner.’ System 15/1:69-76. Palmberg, R. 1987b. ‘Language distance and beginning to read a foreign language.’ Submitted to Journal of Pragmatics. 252 Rolf Palmberg of vocabulary development in foreign-language learners.’ Studies in Second Language Acquisition 9:202-21. Schulak, B. 1986a. ‘Pirate Cove.’ Compute! Gazette 4/8:55-6. Schulak, B. 1986b. ‘Pirate Cove. Programme Listing.’ Compute! Gazette 4/8:98-9. T ha a ut hor Rolf Palmberg is Senior Lecturer in EFL Methodology at Åbo Akademi (Department of Teacher Education) in Finland, where he is responsible for teacher training at various levels. He has published reports and articles in the fields of applied linguistics and EFL methodology, and is co-author of a book containing communicative activities for learners of Swedish as a second language.