Effect of screen size on multimedia vocabulary learning

The objective of this study was to investigate the effects of three different screen sizes (small, medium and large) and two types of multimedia instruction (text only and text with pictorial annotation) on vocabulary learning. One hundred thirty-five Korean middle school students learning English as a foreign language were randomly distributed into six groups and were given a pretest, a self-study multimedia instruction, a posttest and a retention test online. The pretest, posttest and retention test were identical and included 30 vocabulary questions. Results show that the large screen multimedia instruction helped the students to learn English vocabulary more effectively than the small screen instruction as demonstrated on both the posttest and retention test. However, there was little difference in vocabulary learning between the text-only and text-with-pictorial annotation instructions. Although visual perception can be influenced by each learner’s expectations and knowledge, using a smaller screen for instruction causes more challenges for learners to perceive and comprehend vocabulary learning. Introduction Two-year-old Jonah watches his favourite movie, Dinosaur, on an 80-gig iPod in his living room. It is not surprising to see young children using small portable devices and even electronic toys as learning tools. Older children often view their favourite digital content (eg, movies, TV shows, news, books) on computers, iPods, e-book readers or cell phones. Students spend hours reading emails, blogs, text messages, news and reports. They even use various digital devices to complete school projects. Moreover, many newer portable devices enable their owners to view and create digital content online. Hearther (2008) notes that ‘reading doesn’t have to involve cover-to-cover, word-for-word activity’ (p. 34). This reading trend requires educators and instructional material developers to change the way they use electronic content to teach young students. Such content often embraces text and graphics but also makes use of multimodal features including multimedia and hyperlinks (Larson, 2008). Many educators believe that all of these features should be taken into consideration when working with technology-based education. Literature review In this era of digital devices, mental processes are strongly related to tool-mediated activity (Vygotsky, 1986). Visual text and graphics are already popular tools in distance learning environments, where graphics are often used to represent important information and support British Journal of Educational Technology (2010) doi:10.1111/j.1467-8535.2010.01145.x © 2010 The Authors. British Journal of Educational Technology © 2010 Becta. Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA. traditional text (Newby, Stepich, Lehman & Russell, 1996). The properties of such tools are inseparable from the cognitive information processes of the uses of the tools (Rogoff, 1990). Some researchers (Paivio, 1986; Sadoski & Paivio, 2001) believe that verbal stimuli (ie, visual text) and non-verbal stimuli (ie, graphics) are processed by the sensory systems. Other researchers have shown that higher cognitive processes occur when learners interact with verbal and nonverbal information (Mayer & Moreno, 2002; Schnotz, 2001, 2005). For example, Kim and Gilman (2008) adopted the cognitive theory of multimedia learning (CTML) from Mayer and Moreno (2002) to show that visual text (verbal) and graphics (non-verbal) enhanced the vocabulary learning performance of English as a foreign language (EFL) students better than visual text (verbal) and audio (verbal) in web-based instruction. Kim and Gilman explained that because many EFL learners are familiar with memorising new English vocabulary without knowing how the word is pronounced, audio creates an unnecessary distraction and thus requires a heavier cognitive load, as suggested by Sweller (Schnotz & Kürschner, 2007; Stiller, Freitag, Zinnbauer & Freitag, 2009; Sweller & Chandler, 1994; Sweller, van Merriënboer & Paas, 1998). However, adding graphics such as pictorial annotations in multimedia instruction is not always effective. For example, Schnotz, Bannert and Seufert (2002) argued that some learners paid less attention to visual text when pictures were added. Acha (2009) also found that children’s vocabulary learning performance was better when they received verbal annotations than when they received either both verbal and pictorial annotations or pictorial annotations only. From the cognitive psychology perspective, most cognition involves a mixture of bottom-up processing (stimulus– attention–perception–thought processes–decision–response or action) and top-down processing, which is ‘influenced by the individual’s expectations and knowledge rather than simply by the stimulus itself ’ (Eysenck, 2001, p. 2). Although visual perception depends on the information that is presented to the eye and involves bottom-up processing, constructivist theorists emphasise the top-down process (Eysenck, 2001). The growth of new technologies also demands a broader view of learning. In technologyenriched environments, learning occurs not only through interactive learning activities with others but also through the procedures of internalisation with the use of digital devices. Instances of learning through internalisation are illustrated by learning experiences that are shaped by computer-assisted instruction. In computer-assisted language learning, for instance, vocabulary learning has been facilitated through portable devices, the Internet and multimedia technology. Researchers have recently begun to investigate the use of various mobile technologies such as mobile phones, MP3 players and personal digital assistants for learning (Stockwell, 2007). An extensive amount of research on vocabulary learning via mobile phones has been conducted (eg, Chen, Hsieh & Kinshuk, 2008; Lu, 2008; Motiwalla, 2007). The findings of these studies suggest that mobile learning (m-learning) by means of these technologies has great potential in providing EFL learners with rich learning experiences anytime and anywhere (Lu, 2008). Cavus and Ibrahim (2009) claimed that students expressed their satisfaction and enjoyment of learning new English words with short message service text messaging through their mobile phones. These mobile devices that enhance communication and social interactions can be potential learning tools for language learners. Rationale for the current study Although mobile technologies that utilise wireless Internet connection could potentially make a difference in m-learning, the characteristics of small screens (eg, resolution, display size and text/image density) certainly create problematic challenges for the development of multimedia instruction. Most previous studies, which emphasised the importance of multimedia in learning, were conducted without comparing different display sizes. A few studies focusing on comparing different screen sizes have been conducted (Chen et al, 2003; Maniar, Bennett, Hand & Allan 2 British Journal of Educational Technology © 2010 The Authors. British Journal of Educational Technology © 2010 Becta. 2008; Reeves, Lang, Kim & Tatar 1999). For instance, some of the previous studies emphasising the importance of screen size were focused mainly on the effect of screen size on user attention. Reeves et al (1999) found that people tend to pay more attention when they receive a media message on a large screen. Chen et al (2003) raised a similar concern about screen size; that is, a small screen size may create problems with attention and visual perception. In a related study, Maniar et al (2008) stated that a small screen size may be problematic for learning paper folding through video-based instruction because the small space often displays less data at a given time and may create difficulties for users when they use the device for complex tasks. In spite of the interdependence of screen size and learning, there has been little investigation of the effects of screen size on students’ vocabulary learning. Students live in a rapidly changing and increasingly technological society where they may be exposed to numerous types of digital instruction. Because ‘the human mind is limited in the amount of information it can process’ (Miller, 1956, as cited by Sorden, 2005, p. 264), it is very important that educational researchers understand the effects of different screen sizes on multimedia vocabulary learning to reduce redundant memory load and to increase the effectiveness of instruction. The main focus of our study was to extend Kim and Gilman’s (2008) experimental idea with three screen sizes (small, medium and large) and two types of multimedia instruction (text-only and text-with-pictorial annotation mode). Method Research questions Our interest in investigating the effectiveness of screen size and two different instructions on students’ vocabulary learning led to the following research questions: 1. What are the differences in vocabulary learning among students who received the small, medium and large computer screen multimedia instructions? 2. Is there a difference in vocabulary learning between the text-only and text-with-pictorial annotation mode of instruction? Participants and materials Five middle school classes in South Korea were recruited, yielding a sample of 135 students to participate voluntarily in the study during July 2009. Participants were randomly assigned to one of the six experimental groups in the study as shown in Table 1. We randomly assigned the numbers 1–6 to the participants in order to divide them into six groups. The retention test groups were smaller than the posttest groups because only students who took both the pretest and posttest were analysed for retention test results. We adopted Kim and Gilman’s (2008) instructional design with three common screen sizes (or screen resolutions): 320 ¥ 240 pixels (small screen; eg, iPod), 480 ¥ 320 pixels (medium screen; eg, smart phone) and 600 ¥ 800 pixels (large screen; eg, Kindle). For each screen size, two representation modes were considered as the type of multimedia instruction: visual text only and visual text and pictorial annotation. Table 1: Numbers of students in each group in pretest, posttest and retention test Screen size Group Pretest Posttest Retention test Small (320 ¥ 240) Text only n = 22 n = 20 n = 16 Text with pictorial annotation n = 22 n = 19 n = 14 Medium (480 ¥ 320) Text only n = 23 n = 20 n = 15 Text with pictorial annotation n = 20 n = 23 n = 13 Large (600 ¥ 800) Text only n = 21 n = 21 n = 16 Text with pictorial annotation n = 27 n = 27 n = 23 Multimedia learning and screen size 3 © 2010 The Authors. British Journal of Educational Technology © 2010 Becta. The key design of the instruction was based on the following criteria: • The items of English vocabulary were of appropriate difficulty level for Korean students. • Pictorial annotations supported visual text. • Pictorial annotations were available for cueing the meaning of vocabulary items from static or animated images. • Example sentences were available for linguistic cues. A web-based self-instruction programme was used for student vocabulary learning. The length of each lesson was a maximum of 30 minutes, and students controlled the amount of time they spent on each instruction. Procedures The experiment was conducted via the Internet. Students were required to take a pretest, participate in multimedia instruction, take a posttest and take a retention test for the study. The 30 vocabulary questions on the pretest, posttest and retention test were identical and were delivered in the same format. The pretest was administered to the participants approximately 1 week before they received the multimedia instruction. During the following week, the students received the self-study multimedia instruction based on their assigned groups. Items of vocabulary were projected on the computer screen through the multimedia instruction. Immediately after students finished the instruction, they were asked the same questions as in the pretest. Approximately 1 week after the posttest, the retention test, which contained the same questions as the posttest, was conducted. Results The analysis that follows is guided by the two research questions listed above. Question 1: What are the differences in vocabulary learning among students who received the small, medium and large computer screen multimedia instructions? The data were analysed to evaluate the difference of the small, medium and large screen sizes regardless of multimedia presentation modes as shown in Table 2. Analysis of variance (ANOVA) was conducted to determine if differences existed among the groups on the pretest, posttest and retention test. Scores on the pretest did not differ significantly across the screen size groups on a one-way ANOVA, F(2, 132) = 0.61 and p = 0.544. However, the comparison of the three groups showed a significant group difference, both in the posttest, F(2, 127) = 11.34, p < 0.001, and in the retention test, F(2, 94) = 5.53, p < 0.01. Eta squared (h) was also calculated as a measure of effect size. The resulting h values in the posttest and retention test were 0.15 as a large effect and 0.11 as a medium effect respectively. Tukey’s Honestly Significant Difference (HSD) test indicated that students who received the medium and large screen multimedia instructions earned higher scores than students who received the small screen instruction on the posttest and retention test. In the retention test, however, scores of students who studied content on the medium screen (M = 14.71) were not significantly better than those of students who studied content on the small screen (M = 12.27). Table 2: One-way analysis of test scores of three different screen sizes on vocabulary learning Test Small (screen size) Medium (screen size) Large (screen size) F p Pretest (mean score) 8.98 9.95 9.50 0.61 0.544 Posttest (mean score) 16.62 20.95 23.92 11.34 0.000* Retention test (mean score) 12.27 14.71 18.87 5.53 0.005* *The score difference is significant at the 0.01 level. 4 British Journal of Educational Technology © 2010 The Authors. British Journal of Educational Technology © 2010 Becta. Question 2: Is there a difference in vocabulary learning between the text-only and text-withpictorial annotation mode of instruction? In order to respond to the second research question, the data were separated into ‘text-only’ and ‘text-with-pictorial annotation’ groups. Table 3 shows that the overall difference between the two groups was also not significant. In other words, the performance of the visual text-only group and the text-with-pictorial annotation group was similar. In addition, each screen size group was separated into the text-only and text-with-pictorial annotation groups. The results of a t-test also revealed that there was no significant difference in the posttest between the two groups within the same screen size (small, medium and large). Discussion In this study, we sought to investigate the effects of three different screen sizes and two different modes of instruction on Korean EFL students’ vocabulary learning. In sum, we identified the effects of screen size on multimedia vocabulary learning. More specifically, the mean of students who studied English vocabulary on the large screen was significantly higher than that of students who studied English vocabulary on the small screen. This difference was observed in both the posttest and retention test. Our findings also indicated that there was little difference between the text-only and text-with-pictorial annotation instruction on the same screen size. Screen size Our study shows that small screen instruction can be less effective in assisting Korean EFL students to learn and retain English vocabulary than large screen instruction. Figure 1 shows the cognitive process of vocabulary learning, which is modified from Mayer’s CTML (Mayer 2001, p. 59). Although providing either text only or text and pictorial annotation allows learners to select relevant information (S), organise it into coherent representations (O) and integrate it with prior knowledge (I) as meaningful learning, showing the information on a small screen may lead to increased cognitive load. Increased cognitive load (ie, extraneous cognitive load) may occur because the small screen (groups A and B) affects students’ attention and visual perception (Chen et al, 2003; Maniar et al, 2008). This small screen effect can appear in the transition from sensory memory to working memory. For example, the small screen often shows information with limitations (ie, distortions in brightness, colour, font, and spacing between characters, lines, and words) compared to the large screen. In the cognitive process of vocabulary learning, learners must read the information closely and carefully and then try to translate it in a manner that is meaningful to them. When students receive words in visual text or pictorial annotations that ‘gorge’ means ‘to eat greedily’ on the small screen, they are easily distracted because the information with limits of the viewing screen does not have sufficient saliency and meaning to hold their attention, which may lead to perceptual errors. Moreover, spending too much time manipulating the environment itself by closely reading words or graphics on the screen may distract them from the concepts to be learned (eg, spelling and definition). Although visual perception can be Table 3: Means of test scores with t-test results for two different instructions on vocabulary learning Test Text only (mean score) Text with graphics (mean score) t-test p Pretest (mean score) 9.09 9.84 1.60 0.208 Posttest (mean score) 19.57 21.78 0.12 0.729 Retention test (mean score) 14.68 16.52 0.48 0.490 Multimedia learning and screen size 5 © 2010 The Authors. British Journal of Educational Technology © 2010 Becta. influenced by each learner’s expectations and knowledge, working memory activities can be overloaded by the limitations of the small screen before the learner gets to a series of information processing systems (eg, thinking process, decision and response or action) that enables meaningful retention of vocabulary. Furthermore, the small screen may create obstacles in cognitive processes for users who are accustomed to learning on a bigger computer screen. For instance, some English definitions on the screen may not be translated correctly and their meanings may be distorted for students who are habituated to scanning information quickly rather than reading word by word on bigger computer screens. The scanning activity may make it difficult to perceive and comprehend vocabulary as students try to locate information rapidly by moving their eyes on the screen. Especially in the scanning process of an unrelated piece of information, students may often make careless mistakes and have difficulty in cognitive activities. One implication from our findings is that multimedia developers should consider perceptual errors and obstacles in cognitive processes in the use of small screens to reduce unnecessary memory load and increase the effectiveness of instruction. Our findings suggest one more interesting consideration: that the limitations of small screen may have an effect on the nature of cognitive processes in working memory of vocabulary learning. According to Vygotsky (1986), ‘the word is a direct expression of the historical nature of human consciousness’ (p. 256). In other words, consciousness can be investigated in a word. In addition, thought and language in a word are inseparable. That is, the properties of language are inseparable from the thinking processes related to the word use (Rogoff, 1990). If one assumes that vocabulary learning is a tool-mediated activity, our findings can also indicate that screen size may affect not only language learning but also thinking processes and reasoning in language learning. In the domains of vocabulary learning, a bigger screen may effectively develop students’ thinking process and reasoning because of its smaller cognitive load. Future researchers should take all components of cognitive load into consideration in assessing the outcomes of multimedia vocabulary learning. At the same time, there is a need to study how we can deal with all components to ensure a sound foundation on which to investigate thinking processes and reasoning in vocabulary learning. Instructions on small screen Sensory memory