Can Interactive Multimedia Be Used To Improve The Mnemonic Keyword Method Of Second Language Vocabulary Acquisition? By Matthew Francis, May 2005 Copyright AppleCraft

Presented as Part of the Requirements for an Award Within the Undergraduate Modular Scheme at the University of Gloucestershire

 
Abstract

This research project aimed to investigate the possible application of interactive multimedia to the mnemonic keyword method of second language vocabulary acquisition and to discover any potential benefits to the long-term retention of vocabulary when compared to the traditional, non-multimedia approach of the keyword method. An interactive multimedia application was built to display a series of mnemonic images supporting researcher-created keywords, which was tested for word recall against a traditional, non-multimedia, presentation of the same keywords in a longitudinal study including immediate and 1 week recall. The statistical findings of the experiments were analysed using an unrelated ‘t’ test, the results show that the multimedia supported keyword method improved immediate recall p<0.15, and dramatically improved delayed recall p<0.0089. Qualitative observations were recorded during the learning and testing phases of the experiments to support the analysis of the statistical data. The potential advantages and disadvantages of using interactive multimedia to support the keyword method are outlined and suggestions are made for further study.

 

Chapter 1: Introduction

1.1 - Overview

Many second language learners appear to use only a limited number of language learning strategies, though many learners who want to be more effective are looking for some system that can combine as many proven strategies, good ideas and resources as possible to help improve their vocabulary and language learning skills more quickly and effectively.

 

The mnemonic keyword method (KWM) is a popular method of vocabulary acquisition, however, although supporters of the keyword method have suggested that it has useful applications in education (Atkinson & Raugh, 1975; Pressley, Levin & Miller, 1981; Merry, 1980; Gruneberg, 1996), many other authors (Hall, Wilson & Patterson, 1981; Thomas & Wang, 1996; Wang & Thomas, 1995; Hulstijn, 1997; Willerman & Melvin, 1979) contest that the technique has a number of unresolved problems. Interactive multimedia is being used increasingly for second language acquisition, not only because of the practical and motivational benefits to learners, but also the range of modalities available to interactive multimedia applications make them ideal for second language instruction. The purpose of this research project is to investigate whether interactive multimedia can be used to resolve some of the problems associated with the mnemonic keyword method.

 

1.2 - Aims And Objectives

The main aim of this research project is to discover whether word recall is improved when the mnemonic keyword method of second language vocabulary acquisition is displayed using interactive multimedia. To achieve this aim, a review of the literature will be conducted to investigate the theories surrounding second language acquisition (SLA), the benefits and limitations of the keyword method, and to investigate the possible ways multimedia can solve the limitations of the keyword method. To test the advantage of using interactive multimedia to support the keyword method, a multimedia application will be built to display a series of mnemonic keyword images as graphics and animations, together with sound and interaction. A small-scale experiment will then be conducted to collect quantative statistical data and qualitative observational data.

 

 

 

 

 

 

 

 

 

Chapter 2 - Literature Review

2.1- Introduction

The aim of this literature review is to identify some of the principle aspects of second language learning, how the mnemonic keyword method is used to improve vocabulary acquisition, the potential limitations of the keyword method, and to investigate the use of multimedia in second language learning in order to discover the potential applications and benefits of using multimedia representations to enhance the mnemonic keyword method of vocabulary acquisition.

 

2.2 – Second Language Acquisition

When adult learners of a foreign language are unsuccessful they tend to blame themselves and are resigned to the fact that their memory is not what it used to be. It is the view of many learners and teachers that the traditional, ‘tried and tested’, methods of language learning are sufficient, and when these methods fail, lack of hard work or natural ability are suggested as the cause. Cognitive psychology suggests that people learn in different ways, that, what may be appropriate method for some, may not work for others. What is to be learnt is rarely disputed, but learners are rarely taught how to learn the material on offer using the most appropriate methods for them.

 

The most prevalent form of second language acquisition (SLA) tools available to language learners have traditionally been books and audiotapes. They normally rely on rote repetition to drum the words into memory. Rote repetition requires the learner to repeat to themselves, the L2 (target language) word followed by its L1 (learner’s language) translation. This forms a link between the memory codes for the L1 word and the L2 word, as figure 1. shows, this link is strengthened with each additional repetition. This method can be effective for many learners, but only if the learner has the motivation and focus to stick with it, because many language learners may find this method laborious and boring. Thomson (1999) suggests that the laboriousness of traditional methods such as rote repetition is one of the main factors that lead to many people failing to become proficient in a second language. It is clear that, to any SLA method, improving motivation is an important part of improving learning.

Figure 1. Strengthening the links between memory codes using rote repetition

2.2.1 - The Importance of Vocabulary

Whichever method is used to learn a foreign language, the foundation of success is a good range of vocabulary. Gruneberg & Jacobs (1991) argues that it is possible to get by in another language using just 100 of the most common words, and that knowing 3000 of the most frequently occurring words will give a 90% understanding of the language.

 

Of course, language is not just vocabulary. Semantics and grammar form an integral part of languages and an understanding of these nuances is needed if a learner is to become fluent. However, the acquisition of a solid vocabulary may give inexperienced learners the confidence and motivation to start practicing the language with others, and should give all learners a base from which to explore the more intimidating aspects of the language such as grammar and semantics.

 

2.3 - Mnemonics and Vocabulary Acquisition

Mnemonics are techniques developed to aid the memorisation of information. According to Paivio & Desrochers (1981) mnemonics work by developing a retrieval plan during memory encoding so that a word can be recalled through verbal and visual prompts. They rely on the formation of associations between an easy to remember construct, such as a keyword, verse or mental image and the subject to be learnt, such as words, numbers or a list.

 

 

Despite the extravagant (and often groundless) claims of authors such as Buzan (1989) and Gruneberg (1987; 1991) mnemonic techniques do not improve memory as a whole, and may not be the answer to all memory demands experienced by L2 learners, but, as suggested by Hauptmann (2004), mnemonics should be used for specific memory tasks, and there are sensible and effective uses that can be made of these strategies.

 

2.3.1 - Mnemonic Keyword Method

There have been many studies into the use of mnemonic techniques, and there are many methods for second language acquisition that involve mnemonics in one form or another. One of the most flexible and powerful mnemonic methods used in language acquisition is the keyword method. The mnemonic keyword method was formalised by Atkinson in 1975, who describes the method in two stages; an acoustic stage, and a visual stage:

 

“By a keyword we mean an English word that sounds like some part of the foreign word. In general, the keyword has no relationship to the foreign word except for the fact that it is similar in sound. The keyword method divides vocabulary learning into two stages. The first [acoustic] stage requires the subject to associate the spoken foreign word with the keyword, an association that is formed quickly because of acoustic similarity. The second [visual] stage requires the subject to form a mental image of the keyword, ‘interacting’ with the English translation; this stage is comparable to a paired-associate procedure involving the learning of unrelated English words” (Atkinson, 1975: 821).

 

For example, the Thai word ‘Bahn’ meaning ‘house’ can be learned by first establishing an acoustic link with the phonetically similar English keyword ‘barn’ and then linking the keyword visually with the L1 word by imagining a mental image of a man sitting on a huge comfy sofa inside a barn saying, “I don’t live in a house, I live in a barn”.

 

In order for the learner to remember the associations used in mnemonics, the associating mental images themselves need to be as memorable as possible. Buzan (1989) suggests the following techniques to make the mnemonics more memorable;

 

1)      Exaggerate the size of important parts of the image

2)      Use humour, funny or peculiar things are easier to remember than normal ones.

3)      Similarly rude or sexual themes are very difficult to forget

4)      Symbols (e.g. red traffic lights, pointing fingers, etc.) can be used in mnemonics.

5)      Vivid, colourful images are easier to remember than drab ones.

6)      Use all the senses to code information or dress up an image. Remember that your mnemonic can contain sounds, smells, tastes, touch, movements and feelings as well as pictures.

7)      Bringing three dimensions and movement to an image makes it more vivid. Movement can be used either to maintain the flow of association, or can help to remember actions.

8)      Locate similar mnemonics in different places with backgrounds of those places. This will help to keep similar images distinct and unconfused.

 

The original experiments conducted on the mnemonic keyword method of L2 vocabulary acquisition employed Russian (Atkinson, 1975; Raugh & Atkinson, 1977) and Spanish (Atkinson & Raugh, 1975). These experiments found extremely positive results using the keyword method to teach L2 vocabulary to university students when compared to other methods including rote repetition. The efficacy of Atkinson and Raugh’s original studies have been investigated by numerous researchers since (Brigham & Brigham, 1998; Pressley, Levin & Miller, 1981; Sweeney & Bellezza, 1982). While these later studies did not tend to find such emphatic results, most demonstrated an advantage in using the keyword mnemonic to teach vocabulary. However, there have been a number of studies that have discovered conflicting results from the aforementioned studies.

 

2.4- The Limitations of the Keyword Method

2.4.1 - Keyword Method and Experienced Learners

Several researchers (Wang & Thomas, 1995; McDaniel & Pressley, 1984) have suggested that teaching the KWM to more able learners proves less effective than teaching it to inexperienced learners, perhaps because experienced learners already have a variety of effective strategies that they use. A study conducted by McDaniel & Pressley (1984) that failed to find any superiority in the KWM among university students, pointed to the high academic grades of their subjects. They suggested that other, similar studies that have found a KWM superiority, have used subjects who were less linguistically able. This suggests that the method may not be as applicable to experienced language learners as it is to learners who have not developed their linguistic skills.

 

2.4.3 - The Importance of Instruction and Practice

Wang & Thomas (1995) suggests that before a learner can effectively use the KWM, they need to understand how it works, and that durable mnemonic images take a lot of practice to create. Tellingly, the KWM is shown to be more effective in studies where participants are given individual instruction in the method, rather than in classroom-based experiments. Hall, Wilson, and Patterson (1981) conducted studies involving groups of students in classroom conditions comparing the use of the keyword and rote methods to learn Spanish nouns. When recall was tested, the control method was shown as slightly superior to the KWM in immediate recall, and significantly superior in delayed recall. Pressley, Levin, Digdon, Bryant, and Ray (1983) attempted to confirm the results of Hall et al. (1981) in laboratory experiments. They used a sample of students who were asked to learn the same nouns, using the same methods as the Hall et al study. When recall was tested, the participants who used the KWM showed significantly superior recall to those in the control group. The main difference between the studies of Pressley et al. (1983) and Hall et al. (1981) is that Pressley et al. applied the different conditions to individuals, whereas Hall et al. worked with groups, and, as pointed out by Levin et al. (1979), the KWM requires a high level of instruction for it to be used effectively.

 

2.4.5 – The Keyword Method and Backward Recall

A number of studies (Wang & Thomas, 1992, Wang & Thomas, 1997) have found that the KWM is not as effective as rote repetition in backward recall (recalling the L2 word from the L1 word), although similar studies (Pressley et al, 1981; Hulstijn, 1997) found that the method is no better and no worse at backwards recall than rote repetition, but significantly better in forward recall. Because the L1 meaning and the keyword are tied together in the mnemonic image the studies showed that the subjects remembered the keyword when given the L1 word as frequently as they remembered the L1 word when given the keyword. The problem is- remembering the unfamiliar L2 word from the keyword is much harder than remembering the familiar keyword from the L2 word.

2.4.6 - The Keyword Method and Delayed Recall

One of the main limitations of the KWM is that word recall has been proven to drop more dramatically over time than with other methods (Wang & Thomas, 1996). As can be seen from figure 1. the link between memory codes for the L1 and L2 words is strengthened over time by repetition, this means that the link is durable, though there still remains only the one direct link between the 2 words. As can be seen from figure 2. the KWM forms a number of links between the L1 and L2 words, although individually they are not reinforced in the same way as the link in rote learning. This illustrates why the KWM is more effective in immediate recall, as there are many links between the 2 words helping the learner to recall the target word. It also illustrates why the KWM is weaker in delayed recall as these links may fade quicker than the strong singular link formed using the rote method.

Figure 2. The Keyword Method

 

In a recent debate between Gruneberg (1998) and Wang & Thomas (1999) about the effectiveness of the KWM for enhancing long-term recall, a number of questions remain unresolved. Gruneberg claimed that the fact that Thomas and Wang (1996), Wang and Thomas (1992), and Wang, Thomas & Ouellette (1992) did not obtain better recall with the keyword method than they did with the rote method was because the participants were not tested immediately after exposure, but several days later. Wang and Thomas (1999) argued that the KWM is not effective for enhancing long-term recall because the original keyword encoding is fragile, and alternative representations may be evoked, particularly after a delay.

 

However, both Gruneberg (1998) and Wang & Thomas (1999) agree it is not the keyword itself that failed to be remembered in delayed recall, it is the image. This is confirmed by a number of experiments into the performance of the KWM in delayed recall (Wang & Thomas, 1992; Atkinson, 1975; Hall et al, 1981). Clearly then, the main weakness of the method is remembering the mnemonic image from either the L1 word or the keyword. It is these links, and the memorability of the image itself, that need to be strengthened.

 

2.5- Using Multimedia to Enhance the Keyword Method

2.5.1 – Defining Interactive Multimedia

This research project will use the definition of interactive multimedia presented by Vaughan (1993):

“Multimedia is any combination of text, graphic art, sound, animation and video that is delivered by computer. When you allow the user to control that and when these elements are delivered, it is interactive multimedia.” (Vaughan, 2004, p. 3).

 

2.5.2 –Benefits of using Interactive Multimedia for Language Learning

Interactive multimedia is being used increasingly in an educational context, and especially for language learning (Brett, 1996). There are many CD-ROMs and DVD-ROMs on the market that utilise interactive multimedia elements such as images, video, animation and audio alongside text to teach languages. Of course, the use of images, video, audio and text in language learning is not the sole domain of multimedia applications, many traditional methods of language learning make good use of all these media in varied combinations. However, as Canning (1998c) suggests, it is when all these elements are presented in one source, the computer, and combined and orchestrated by one application that their full potential to educate is unleashed. This, combined with the highly interactive nature of many multimedia learning applications, means that the combination of media used should be greater than the sum of its individual parts.

2.5.3 - The Application of Multimedia to the Keyword Method

If used correctly, interactive multimedia and the mnemonic keyword method can both be powerful methods of language learning. Despite the amount of research that has gone into the use of the keyword method and the use of interactive multimedia in language learning, there has been little research into the potential application of a combination of these two branches of SLA theory.

 

2.5.4 - How Multimedia Can Help Beginner Learners

Motivation is an important issue in language learning, as Brett (1996) suggests, a motivated learner will get more from the time they spend using a learning method, they will be more focused on the task at hand, and will likely retain more of what they have learnt. According to Astleitner & Wiesner (2004) carefully designed, multimedia instruction can be very effective as a method to increase motivation and alertness. Multimedia language learning applications have distinct motivational advantages over human teachers. Many learners may feel intimidated to practice their language skills in front of others until they have gained a certain level of proficiency in the language. Multimedia language learning applications are non-judgemental, the learner can practice their language at their own speed, in their own homes, without the fear that they may appear slow. They are also untiring and infinitely patient, the learner can repeat sections that they are having trouble with as many times as they want.

 

2.5.5 - Researcher-Generated Mnemonics & Line Drawings

In many of the experiments conducted on the keyword mnemonic method, the particpants are requested to construct their own mental image linking the keyword to the target word (Atkinson, 1975; Hall et al. 1981; Gruneberg & Pascoe. 1996). Levin & Miller (1981) suggest that in experiments where pre-constructed mnemonic keywords are presented to participants, the way the mnemonic keywords are presented to the participants can have an affect on recall. Merry (1980) demonstrated that similar efficacy can be obtained by using line drawings supplied by the researchers instead of by asking participants to generate their own mental images. In his study, Merry found an improvement in immediate and delayed recall when researcher generated line drawings were used instead of participant generated mental images.

 

Although the subjects of Merry’s study were children and his findings may not be entirely relevant to adult language learning, the results of Campos, Amor & Gonzalez’s (2002) study support his findings. They found that the use of pre-constructed mnemonic drawings using adult subjects demonstrated increased delayed recall when compared with rote methods, there was no significant increase in immediate recall. This is not surprising, as, according to Astleitner & Wiesner (2004) visual media take advantage of the visual cortex to clarify abstractions such as language and meaning. The fact that pre-constructed, simple line drawings can be used just as, or more effectively than participant generated mental images suggests that more expressive and memorable visual representations (such as animations) of the mnemonic image may be used, with the possibility of further increasing recall.

 

2.5.6 – Using Media to Visualise Mnemonic Imagery

As illustrated above, the keyword method is highly visual in nature and lends itself to visual representations such as drawings. The next logical step in this strategy is to use a medium that can suitably illustrate the interaction of the keyword and the target word in the mnemonic image, i.e. animations and computer graphics. The use of static visuals, such as drawings, as used in the aforementioned studies, have limitations when it comes to the depiction of the abstract concepts that often form part of mnemonic images. As Winn (1989) suggests, complex abstract concepts can be depicted in a much more concrete manner as animations. Rieber (1990) supports this view, suggesting that computer graphics and animation can create metacognition by illustrating and linking a presupposed concept map of the target subjects (L1 word and the keyword) interacting.

 

Animations and other visual multimedia have been used in other methods of language learning for decades, although the value of using visual media to aid language learning is often questioned by academics. Omaggio (1979) claims that there is virtually no empirical basis for promoting the use of visuals as aids to comprehension in the second language, because practically nothing is known about how students benefit from visuals. Canning-Wilson (1998b) disagrees, arguing that using visuals such as illustrations, photographs, animations, colours, charts, or anything else used to help the learner see an immediate meaning in the language may benefit by helping to clarify the message, provided the visual works in a positive way to enhance or supplement the language point.

 

2.5.7 - Multimedia and Memory

The use of visual aids to language learning and retention is supported by psychological research. Standing (1973) reported extensive research on the superiority of visual memory over verbal memory. However, Paivio (1975) believes that combinations of modalities of information are better for memory performance than when these elements are presented individually. This belief is the basis for his dual-coding theory. Paivio hypothesizes that there are two independent codes for creating memory: a verbal and a visual ("imaginal"), one activated by words and the other by pictures. Retention of material can be increased if the information is initially coded in two codes rather than one.  Thus, presentation in visual (including animation) and verbal modes simultaneously facilitates comprehension and memory formation. According to Gardner’s (1993) theory of multiple intelligences people learn in many different ways, depending on their particular proficiency. A simplified view of Gardner’s theory divides learning styles into visual (learn by seeing), auditory (learn by hearing) and kinaesthetic (learn by doing). Interactive multimedia can facilitate all these learning styles and types of memory by inclusion of text, images and animations, sounds and voices, and interaction.

There is an important distinction between short-term memory and long-term memory. Short-term memory stores the most recent memory codes (such as a word that is being learnt) for a short time, approximately 6 to 60 seconds. Long-term memory stores memory codes for a longer time, eventually permanently (Thompson, 1987). The purpose of learning is to pass these memory codes from short to long term memory, although the memory codes must remain easily accessible in long term memory for learning to be effective. The way a learning method is presented can have a major effect on the transfer of a memory from short to long term memory, and its subsequent recall. A popular maxim in the field of educational multimedia states that learners remember 10% of what they read, 30% of what they see, but 80% of what they hear, see, and do. Although these figures have never been proven empirically, they highlight the difference to memory retention that combining modalities can make.

Making the memory codes of L2 words more accessible is not solely achieved by making the link to the memory code of the word stronger. Also important is making multiple and varied connections to the memory code. According to McDaniel, Pressley & Dunay (1987) memory codes are held in a network, and the learner locates a particular word in their memory by following a trail of linked codes. Therefore, using multiple modalities, such as audio, visual and interaction to create more trails and links to the memory code of the target words, keywords and mnemonic image should increase the chances of the learner locating it. Astleitner & Wiesner (2004). found that the more sensory modes in which mental representation is stored, the more likely they will be remembered. Figure 3. shows the multiple connections between the memory code for the L1 word and the L2 word created by multimedia links and the mnemonic keyword.

Figure 3. Using multimedia to strengthen the links between the keyword, L1 word, L2 word, and the mnemonic image.

 

2.5.8 – Supporting Language Learning with Audio

Although there has been no empirical research into the use of audio and the mnemonic keyword method, Gruneberg (1989) makes good use of voice media to support his Linkword method, and there have been numerous positive empirical studies into the use of interactive audio in computer aided language learning. The benefits of using audio media in the target language are obvious, there is no other way to effectively convey pronunciation and improve accent than hearing the word spoken (Brett, 1996). But there may also be benefits of using audio to support the keyword method. As can be seen from Figure 3, vocals can be used to enhance the acoustical link between the keyword and the L2 word and the direct link between the L1 and L2 words, and sound effects or music can be used to enhance and support the vividness of the visualisation of the mnemonic image.

 

2.5.9 - Mnemonics And Grammatical Rules

Mnemonics can be used to support many aspects of language learning, not just the memorisation of vocabulary. Many languages have complex grammatical rules that may differ drastically from the learner’s mother tongue. Mnemonic methods such as the KWM can include prompts to help the learner recall the correct language rule for the target word. An example of this is presented by Desrochers, Gélinas & Wieland (1989) who proved in their study that incorporating exaggerated elements associated with gender in the mnemonic image, such as a boxer to suggest a male noun or a little girl to suggest a female noun in German, helped learners to recall the gender of nouns. Other gender prompts can be included using multimedia, such as colours, sounds, music and imagery associated with gender. Pronunciation prompts can also be included in some mnemonic images using multimedia to help the learner remember the correct pronunciation of L2 words. For example, an IMM application could help the learner remember the pronunciation of the Thai word ‘mahk’ (‘many’) because it sounds like the English ‘Mark’ in a strong Liverpool accent, and the application could include an audio track of someone with a Liverpool accent saying ‘hello Mark’.

 

2.5.10 - Making the Keyword Method More Accessible

Because the keyword method requires imagination and practice to be effective, it may not be accessible to all learners in the way that rote and other methods are. Using multimedia to generate effective pre-designed mnemonics and display these mnemonics in an appealing way could make the method more accessible to all learner types and different proficiencies. According to Vaughan (2004) interactive multimedia helps all learners because of easy, intuitive access to information, presented in a way that provokes interest and requires interaction.

 

The use of a multimedia based mnemonic keyword application may help to increase the ability of the learner to use the mnemonic keyword method by providing initial instruction, and then by showing numerous examples of keywords and supporting mnemonic images, this is a much more intuitive way to learn how to use the keyword method than from text or audio commentary. The experience gained by using this method may help the learner to form their own keyword associations when they encounter a new L2 word.

 

2.5.11 - Limitations Of Using Multimedia

Perhaps one of the reasons there has been little research into the use of multimedia and the mnemonic keyword method is that some keywords do not lend themselves to visual/audio representation, they may be too complicated to be given effective visualisations or may rely on the learner forming highly personalised mental images, which would be impossible to represent outside the mind of the individual learner. However, this does not mean that all keywords cannot benefit from multi-modal representation.

Chapter 3: Methodology

3.1- Research Question

Since the major flaw in the mnemonic keyword method is often the weakness of the mnemonic image (Wang & Thomas, 1992, Wang & Thomas, 1989), and the use of line drawings in the mnemonic keyword method has been proven to increase recall (Merry, 1980), multimedia representations of the interaction between the keyword and the L1 word should produce better recall than the keyword method using participant generated mental images alone. Thus, the focal research question was:

 

Can interactive multimedia be used to enhance immediate and delayed word recall with the mnemonic keyword method of second language vocabulary acquisition?

3.2- Approach

This research project is looking for empirical evidence that a interactive multimedia based mnemonic keyword method application can be used to increase word recall when compared to traditional presentation of the keyword method. Because of the unresolved issue of the effectiveness of the keyword method in delayed recall and the conflicting results from the various studies into this area, it was decided to conduct a longitudinal study to gauge the comparative effectiveness of the two methods in immediate and delayed recall.

 

The biggest influence on the choice of study method is the previous work done in this area. The majority of previous studies into the effectiveness of mnemonic keyword methods have been conducted using classroom (Pressley et al, 1983; Atkinson & Raugh, 1975; Merry, 1980; Hall et al, 1981) or laboratory (Brigham & Brigham, 1998; Hulstijn, 1997) experiments. This approach is confirmed by Robson (1999) who suggests that an experiment is an appropriate method for an explanatory study. Because of the findings of Levin et al. (1979), who suggests that the keyword method is less effective for classroom use than for individual use, it was decided to test participants individually.

 

3.3- Study Design

A multi-method experiment was proposed to collect quantative and qualitative data. The quantitative data collected from the experiments was supported by the qualitative data collected from observations and post-test interviews. As suggested by Robson (2002), using a range of qualitative and quantative methods of data collection gives a richer picture when interpreting the numerical data.

 

Two conditions were tested in the experiments, Condition 1 (the control group who were presented with the non-multimedia list of keywords) and Condition 2 (who were presented with the multimedia supported keywords). To test whether the vocabulary had been retained and recalled, each participant was given a pre-test, testing their ability to remember words using rote, to provide a gauge for the post-tests. According to Robson (2002) a pre-test also enables the allocation of participants to groups according to ability, and enables the researcher to adjust the results to account for any discrepancies between the two conditions. The immediate post-test tested the number of words remembered immediately after the mnemonic keyword presentation. The delayed post-test tested the number of words that were remembered 1 week after the keyword presentation.

 

Qualitative data was collected using observations of the responses and comments of the participants during the learning and testing phases, and from a short interview after the immediate post-test.

 

3.4- Validity and Reliability

To maintain the validity and reliability of the experiments, a number of factors were considered and any possible extraneous variables were limited. Any possible inconsistencies between the testing methods of the two conditions were examined and dealt with. The tests of the two conditions were constructed using the same words, and (with the exception of the multimedia elements) were presented in as similar a way as possible. The tests were conducted in the same environment and all participants were given the same amount of time to complete the tests. The words in all the tests appear in a different order from that which they appear in the learning phases, this helps avoid participants learning the words as a numbered list.

 

3.5 - Pilot-tests

It was crucial that both the control and experimental tests were reliable. Pilot-tests were conducted to ascertain whether the experimental procedures were sound. A number of issues were raised and the experimental procedures were altered accordingly. The pilot-tests found that there were too many questions in the initial design. The participant used in the first pilot-test became bored after 10 minutes of memorising a list of 30 words, so it was agreed to reduce this number to 16 and the memorisation time to 5 minutes. Also, the pilot-tests showed that most of the words in the pre-test were too easy to remember and presented little challenge because they were single syllable words, the list was revised to contain a mixture of simple and difficult to remember words. The participants in the pilot tests found it hard to understand what was required of them, especially when using the non-multimedia presentation of the KWM. It was decided to conduct more thorough instruction at the beginning of the learning phase of the experiments.

3.6- Research Hypotheses

Hypothesis 1 (H1) - one tailed hypothesis predicting that the mean score in the immediate recall tests of Condition 2 (using interactive multimedia) will be significantly higher than the mean score of Condition 1 (using traditional presentation).

 

The null hypothesis is that no significant difference will be found in the scores.

 

Hypothesis 1 (H2) - one tailed hypothesis predicting that the mean score in the delayed recall tests of Condition 2 (using interactive multimedia) will be significantly higher than the mean score of Condition 1 (using traditional presentation).

 

The null hypothesis is that no significant difference will be found in the scores.

Chapter 4: Testing

4.2- Participants

Based on the findings of the literature review, the focal population of this research project will be inexperienced adult learners, who have been proven to benefit the most from the KWM (Wang & Thomas, 1995; McDaniel & Pressley, 1984). A total of thirty adults (14 women, 16 men; mean age 22 years, range 18-29 years) took part in this experiment. To avoid inequality between the conditions, and because of the suggestion of Wang & Thomas (1995) that the KWM is less appropriate for experienced language learners, all participants were volunteers who classed themselves as monolingual native English speakers, had no knowledge of Thai, and little or no experience of using the mnemonic keyword method or CALL. They were all told that the purpose of the experiment was multimedia and language research.

 

4.3- Materials

4.3.1 - Construction of the word lists

To test the effectiveness of applying multimedia technology to the keyword method, Thai was chosen as the L2 language for the study due to the lower chance of the participants having prior knowledge of the words. It is also a tonal language, so methods were devised to test whether multimedia can support aspects of language learning other than basic vocabulary acquisition such as pronunciation.

 

The pre-test was compiled using 16 Thai words and their translations. These words were chosen, after the post-test had been constructed, to form a list of words which mirrored the type of words used in the post-test list, although any words which could naturally lend themselves to the keyword method were rejected to maintain the validity of the pre-tests, for example – ‘boy boy’ (‘often’ in English) was rejected after the pilot-tests because of the ease with which it could be turned into a keyword mnemonic.

 

The word list used in the post-tests was constructed using words chosen for their applicability to the mnemonic keyword method. The words that were chosen were a mixture of concrete nouns and abstract words, cognates were rejected, (i.e. words with similar pronunciation and/or spelling in the two languages) to avoid participants guessing at the meaning of the words.

 

4.3.2 – Design and Development of the Multimedia Application

An interactive multimedia application was created for the purpose of this research project. It was created using Macromedia Flash, Adobe Photoshop, and Audacity. The intention of the application was to display the 16 Thai words along with the mnemonic keyword and English translation in a similar way as the non-multimedia keyword presentation. The only major difference in layout presentation between the two conditions is the inclusion of multimedia elements.

 

The animations, images and sounds used in the application were all created by the researcher. The multimedia mnemonic elements in the application were created using Buzan’s (1989) guidelines to what makes an effective mnemonic image. Each of the animations was created to display the meaning of the mnemonic image as obviously as possible, while keeping the irrelevant content such as background detail to a minimum to avoid distraction. For animations such as ‘bahn’ the animation contains quite a lot of detail, including soundovers on the animals, this is because the imperative factor of the mnemonic image is the barn itself. For mnemonic images where the imperative focus is on an action or sentence such as the animation for ‘mai dai’ there is little or no background detail because the only relevant aspect is the situation and the sentence “No you CANNOT, its MY DYE”.

 

The non-mnemonic elements of the application were created using Vaughan’s (2004) guidelines for instructional multimedia. The style of the interface was kept deliberately simple to avoid any adding to cognitive load that may detract from the content of the mnemonic animations.

 

4.3.3 - Equipment

Condition 2 (the experimental group using the multimedia application) all used the same PC with 128MB Graphics card, 256Mb of RAM, 1.7GHz processor, with screen settings set at 800 x 600. The application was maximised before testing began to avoid distractions from other items on the desktop.

 4.4- Procedure

Each participant was welcomed and thanked for their participation, they were given a brief explanation of the experiment and what was required of them.

 

The procedure for the pre-testing for all participants was conducted as follows:

·        Each participant was sat at a desk and given a sheet containing all the Thai words contained in the pre-test alongside their English translations (see Appendix F).

·        They were then asked to spend 5 minutes reading through the list of words repeating the Thai word to themselves, followed by its English translation (rote repetition).

·        After 5 minutes, the sheet with the translations was taken from them and they were given a sheet with the 16 Thai words in a different order (see Appendix).

·        They were then given 2 minutes to write the English translation by the side of each Thai word.

 

The participants were then randomly allocated to one of the two conditions, the experimental (multimedia) condition, or the control (non-multimedia) condition.

 

The procedure for the immediate post-testing for the control condition participants was conducted as follows:

·        After 5 minutes, the word sheet was taken from them and they were given a sheet with the 16 Thai words in a different order (see Appendix F).

 

The procedure for the immediate post-testing for the experimental condition participants was conducted as follows:

·        After 5 minutes, the sheet with the translations was taken from them and they were given a sheet with the 16 Thai words in a different order (see Appendix F).

 

After the immediate post-test was finished the participants were thanked for their participation and a time was arranged a week later for the delayed recall test.

 

The procedure for the post-testing for the control condition participants was conducted as follows:

 

4.4.1 - Ethical Issues

The testing was conducted in accordance with the University guidelines given in the Research Ethics Handbook. Confidentiality was assured to each participant, they were told that they could withdraw at any time and that they would be fully debriefed at the end of the experiment.