Seeing What They See: Elementary EFL Students Reading Science Texts

Abstract

Science texts use various text features and multiple representations to communicate meaning to their readers. English science texts are challenging for elementary-level English as a foreign language (EFL) learners in Taiwan because they are familiar with reading language-controlled texts from textbooks. Teaching students to make use of various text features and visual representations will help them achieve a more successful science text reading experience. In this study, 27 Grade 6 Taiwanese students were instructed in science text reading strategies that included understanding text features, creating imagery, and using visual representations. Before and after the instruction, they took an English reading and writing test. Their eye movements during science text reading were recorded before and after the instruction to more fully understand their visual attention while reading English science texts. Eye movement performances such as number of fixations, mean fixation duration, and saccade size were examined. The findings showed that although the participants’ English reading and writing performance improved in the post-test, they focussed more on the written language than the visuals in both tests. More visual representation reading strategies should therefore be taught to help young EFL students read and learn from science texts.

Keywords

Science text reading strategy eye movement EFL learners

This study examined Taiwanese Grade 6 students’ ability to read English science texts. Students in Taiwan learn English as a foreign language (EFL), and science texts are a rare choice for English teaching materials. Little research has been conducted to understand the processes and difficulties of young EFL students’ reading of English science texts. To understand the visual attention of young EFL students when they read science texts with multiple representations, this study examined their reading comprehension and eye movements before and after instruction in science text reading comprehension strategies.

Background

The design for this research is based on three rationales, the first of which is acknowledging the importance of reading in learning science. The role of reading and writing is well recognized and highly emphasized in the field of science education. In a special issue of Science that addressed literacy skills in learning science, Hines, Wible and McCartney (2010: 447) noted that ‘good literacy skills make it easier to learn science, but science topics can also be used to teach literacy skills that will translate well to other subjects’. Many science educators have stressed the value of learning science through reading and writing. Davies and Greene (1984) stated more than three decades ago that reading and writing are not alternatives but integral to learning science. Using the metaphor of building houses, Osborne (2002) stated that just as there can be no houses without roofs or windows, there is no science without reading or writing. Palincsar and Magnusson (2001) stressed the importance of reading and learning through other researchers’ previous research experiences in science inquiry.

On the side of literacy research, helping children read texts across various disciplines, genre types, and modalities has long been a core belief in literacy education. Disciplinary literacy aims to teach students the types of discourses and comprehension strategies that are valued in various subject areas (Moje, 2008; Shanahan and Shanahan, 2008). Fang (2008) argued for special training in science text reading because young students are often less familiar with scientific expository texts. The recent proliferation of nonfiction reading materials and the push for content integrated literacy instruction by the Common Core State Standards in the United States are all confirmations of this belief.

The second rationale for this research was to address the limitations of reading materials. In places where children are learning English as a second or foreign language, the focus of instruction is often on vocabulary, grammatical rules, and communicative ability. Taiwan is an example of this type of English instruction. Hung (2007) described the craze of English learning and the overemphasis on English oral skills and phonics instruction in Taiwan. Hung (2017) further noted the limitations of the textbooks used in English classes in elementary schools in Taiwan. English textbooks for elementary school students are limited in terms of vocabulary, sentence patterns, and phonics rules. Reading passages in these textbooks are almost always scripts of short conversations that introduce a few target sentence patterns for a specific communicative function. For example, sentences such as: ‘How’s the weather today?’ ‘Today is sunny.’ and ‘Today, it is raining.’ will feature in the unit on asking about the weather. Sentences such as: ‘How do you go to school?’ ‘I go to school by bus.’ and ‘I go to school by car.’ will feature in the unit on using various modes of transportation.

Taiwan is undergoing a major K-12 curriculum reform. Experts have call for more content-based and cross-disciplinary English instruction in a new curriculum launched in 2019. With these goals in mind, the science reading strategies developed and taught in this study and the research findings can provide valuable suggestions for selecting material and the teaching design for elementary school English teachers in Taiwan and in other countries where English is taught as a foreign language.

The third rationale for this research was to address the importance and value of teaching children to read multi-representational and multimodal texts. Most people read more informational than non-informational texts in their daily lives (Hoyt et al., 2003), and by Grade 6, 75% of the material students read in school is informational text (Moss, 2005). Miller (1998) estimated that in journals such as Science and Nature, visual elements occupy approximately one half to one-third of the space on a page. It is thus crucial for students to develop skills to read various visual representations and multimodal texts in the digital age. The incorporation and juxtaposition of verbal text and visual images is well recognized (Kress and van Leeuwen, 1996; Safafini, 2014), and students must ‘learn how to read between the borders of visual images as much as how to read between the lines of written text’ (Safafini, 2014: 3). Empirical studies (e.g. Doolittle, 2002; Mayer and Anderson, 1992; Mayer and Gallini, 1990; McKay, 1999) discuss and demonstrate the potential and conditions for visual and multimedia illustrations to be helpful in comprehending science texts. However, more research is required on how students make sense of visuals in the reading process (Unsworth et al., 2004).

In English-speaking countries the concept and practice of integrating language learning and content learning is common in English language arts education. As aforementioned, the Common Core State Standards (http://www.corestandards.org/) in the United States places a high value on reading nonfiction and cross-disciplinary texts in English language arts curriculums. Research on teaching English as a second or other language also supports content-based English learning because such learning contextualizes the English language and makes language learning more purposeful and functional. Content and language integrated learning is another pedagogy that lends support to the instructional innovation experiment in the current study. The integration of EFL learning and content learning has often been reported as promoting EFL students’ language learning motivation as well as achievements (Lasagabaster, 2010; Pladevall-Ballester, 2018).

Eye Movement Measures for Understanding Reading

Eye tracking technology was adopted in this research because of its wide application in research on attention, human cognition, and more recently, human–computer interaction. In past decades eye movement studies on reading have revealed the cognitive processes involved in reading comprehension (Conklin et al., 2018; Just and Carpenter, 1984; Rayner, 1995). Eye movements are strong reflections of the comprehension processes in reading (Paulson and Freeman, 2003; Rayner, 1998; Rayner, 2009; Rayner et al., 2006).

Three types of eye movement measures are commonly used as indications of visual attention in science text reading. The first is the total number of fixations in various areas of interest (AOI). Research on eye movement has long recognized that a higher number of fixations suggests more visual attention and cognitive processing (Conklin et al., 2018; Just and Carpenter, 1980; Paulson and Freeman, 2003; Rayner, 2009; Rayner and Liversedge, 2011). Because each AOI is a different size, the total number of fixations is examined in relation to the size of the AOI, indicated in pixel numbers. Therefore, information on the number of fixations is presented as a pixel-to-fixation ratio, where a higher ratio suggests a lower fixation density.

The second measure is the fixation duration in various AOIs. This information is presented as mean fixation duration in milliseconds (ms). The longer the fixation duration is, the higher the cognitive demand and efforts are in processing the meaning (Conklin et al., 2018; Rayner et al., 2006; Rayner and Liversedge, 2011; Samuels et al., 2011). Rayner (2009) in a review of eye movement research on reading, scene perception, and visual search found that the range of mean fixation duration for fluent adult silent reading was 225–250 ms, and factors such as text difficulty and the reader’s reading skill affect the mean fixation duration. We predicted longer mean fixation durations in this study than those reported by Rayner (2009) because our participants were reading an unfamiliar genre in an unfamiliar language.

The third measure is the size of the saccades, which are movements from one fixation to the next. A larger saccade size, or longer eye movement, suggests the reader is more actively moving and integrating visual information from various sections of the page. When the text becomes more difficult, the saccades become shorter (Rayner et al., 2006). The saccade size is presented in visual degrees (deg) in this article.

This study sought to investigate not just the results of comprehension but also the processes of the participants’ reading of English science texts in various forms. Eye movement information made it possible for researchers to ‘see’ what the readers saw and to learn what they were paying attention to.

Little research on young English learners’ reading and visual attention to science texts of multiple representations has been conducted. Second language researchers have used the eye tracking method to study lexical access and representation, syntactic ambiguity resolution, and attention and cognitive processes of various language tasks (Winke et al., 2013), but little research has been done on examining the online moment-to-moment processes of processing print and visuals among second language learners when reading science texts. Mason, Tornatora and Pluchino (2013) examined Italian fourth graders’ online processing of text and graphics of illustrated science texts with eye tracking technology, and the purpose was to understand patterns of viewing in relationship to variables such as comprehension, prior knowledge, and spatial ability. They found that high-integrators, students who more often integrated information from text and from visuals also had better comprehension. This is a confirmation of multimedia principles. No research has tried to understand the effects and possible changes of viewing patterns and reading abilities after students received science text comprehension strategy instruction. The current study may help to fill this research gap.

Research Purpose

The purpose of this study was to understand the influence of instruction in science text reading comprehension strategies on Taiwanese Grade 6 EFL students’ English reading abilities. Specifically, this study employed eye-tracking technology to inspect how the students read and viewed various types of text and visual representations before and after the instruction was delivered. The primary purpose of this research was to demonstrate these elementary EFL readers’ comprehension and selective visual attention when reading a science text. A secondary purpose was to examine the effectiveness of the science text reading comprehension strategy instruction. The current study can help literacy researchers and EFL educators further understand how young EFL students read informational science texts. The results of this study also have wider implications for English reading instruction in other EFL education contexts.

Methods

A one-group pre- and post-test study with an experimental teaching intervention between the two tests was designed. The dependent variables included scores in the English reading and writing pre-test and post-test and eye movements across print and visual areas when reading a science text.

Participants

The participants were 27 students (14 boys, 13 girls) from a Grade 6 class in a public elementary school in central Taiwan. Grade 6 is the last year of elementary school in Taiwan. English instruction commenced in Grade 1 at the selected school, but class time was limited. The participants had only 40 minutes of English instruction each week in Grades 1–2 and 80 minutes in Grades 3–6. The authors of the present study worked together to choose the reading materials and develop the teaching activities, and one of the authors was the instructor in the experimental teaching intervention. A letter was sent to the parents of the participants explaining fully and clearly the research ethics and participants’ rights. Informed consent to participate in this study was obtained from all of the students’ parents.

Instruments

The Movers level of the Cambridge English for Young Learners test (http://www.cambridgeenglish.org/exams-and-tests/movers/) was chosen for the English reading and writing test. Corresponding to the A1 scale in the Common European Framework of Reference for Languages (CEFR) standard, it comprises three parts: listening, speaking, and integrated reading and writing. Only the scores from the integrated reading and writing section were analysed for the purpose of this study. There were 40 items in the integrated reading and writing test, with a possible maximum score of 40. Two equivalent tests were used in this study: one for the pre-test and the other for the post-test.

The EyeLink 1000 model eye tracker was used for collecting eye movement data. The sampling rate was 500 Hz, meaning that eye movements were recorded 500 times per second. The entire eye tracking unit was moved to the participants’ school for their convenience and comfort. The remote mode of eye tracking was selected to increase the authenticity of the reading event. The participants sat 60 cm in front of a 19-inch computer monitor, where the reading material was displayed. Each reader was fully informed of the data collection procedure, and all questions from the participants about the purposes or procedures of eye movement data collection were answered before they started reading.

Reading Strategy Instruction

The reading materials used in the experimental teaching were 12 science texts selected from three books in the Discover More series published by Scholastic: See Me Grow, Farm, and Puppies and Kittens. Permission from Scholastic in Taiwan was obtained to use these materials in this study. The reading strategy experimental teaching lasted 12 weeks, with two class meetings per week, lasting 30 minutes each.

Three reading strategies were taught. First, students learned about various text features in science texts and their functions. These included the names, uses, and functions of headings, captions, visuals, page numbers, text boxes, bold and italic print, the table of contents, and the index. Second, students learned to make mind pictures to facilitate their understanding. They practised drawing and depicting their understandings of the text in illustrations. Last, they learned about various visual representations and how to use them to support their comprehension. The functions of decorational, representational, and interpretational illustrations were taught as well as various forms of visual representations, such as photos, organization charts, graphs, tables, and three-dimensional graphs.

We followed the gradual release of responsibility model (Pearson and Gallagher, 1983) to plan the teaching activities. The instructor, who was one of the authors of this study, started with explicit teaching of the reading strategy and provided a demonstration of how to use it with an exemplar text. Then, students worked in small groups and practised with another text in class. Finally, students completed individual practise at home. After 12 weeks of teaching, students were interviewed to understand their responses and comments on the reading materials and learning activities in this experimental teaching programme.

Data Analysis

The scores of the integrated English reading and writing tests were converted to a percentage first, and subsequently, a pairwise t-test was conducted to understand whether any significant differences existed between the pre-test and post-test scores.

For eye movement pre-test and post-test data collection, a text titled ‘Animal Babies’ from See Me Grow was chosen. This text was not chosen for the strategy instruction; therefore, it was new to the participants and appropriate for eye movement data collection. The text was divided into various areas of interest (AOIs) for further analysis. The two primary types of AOIs are print and visuals, with print including headings, text, and captions, and visuals including deictics, illustrations, and page numbers. Figure 1 illustrates the reading text and its AOIs with various text features and visual presentations.

Figure 1.

Reading text used in eye movement pre-test and post-test data collection. Heading, Text, and Caption AOIs belong to the print category; Deictic, Illustration, and Page Number AOIs belong to the visual category. Fixations that do not fall on any of these AOIs belong to the blank category.

A descriptive analysis of the eye movements was conducted first to understand the number, density, and size of the fixations and saccades. Then, analysis of variance (ANOVA) was conducted to investigate whether any significant differences existed among eye movements in the various AOIs in the pre-test and post-test and between eye movements in the same AOIs across the pre-test and post-test.

Results

Reading and Writing Test

The results of the pairwise t-test comparing the pre-test and post-test scores of the integrated reading and writing section of the Cambridge English for Young Learners test are provided in Table 1. The difference between the mean score of the pre-test (45.00) and that of the post-test (51.57) is statistically significant (p < .01). In the absence of a control group, to state that the experimental teaching caused the growth in the post-test is unfounded. It is more appropriate to say that, in the specific context of this research, the instruction helped this specific group of participants to progress in their English reading and writing ability after 12 weeks of experimental teaching.

Table 1.

Integrated reading and writing pre-test and post-test t-test results.

	n	Mean	SD	t	p
Pre-test	27	45.00	26.04	−2.95	.007**
Post-test	27	51.57	27.10	−2.95	.007**

p < .01.

Heat Maps

Before we examined the eye movements statistically, fixation time based heat maps of the pre-test and post-test were created to provide a more holistic view of the readers’ visual attention. The hotter, or darker, an area on a heat map is, the more time readers spent on it. As illustrated in Figure 2, the two heat maps are similar in that print AOIs are generally darker than visual ones. The results are both surprising and expected. They are surprising because after receiving 12 weeks of science reading comprehension strategy instruction that targeted the use of visual information to assist comprehension, the participants still focussed more on the print than the visuals. The results are, however, expected because the participants were EFL learners, and they wanted to focus on the words and sentences to understand their meanings. A deeper look into the eye movements can reveal more about the influence the strategy instruction had on the participants’ reading performance. Table 2 provides the results of the descriptive analysis of the eye movement data, based on which we checked for any significant differences among the various AOIs within the pre-test and post-test as well as possible significant differences within the same AOI across the pre-test and post-test.

Figure 2.

Fixation time based heat maps (n = 27).

Table 2.

Descriptive analysis of eye movement data.

AOI		Total no. of fixation (%)		Area pixel (%)	Pixel-to-fixation ratio		Mean fixation duration (ms)		Mean saccade length (deg)
AOI		Pre-test	Post-test	Area pixel (%)	Pre-test	Post-test	Pre-test	Post-test	Pre-test	Post-test
Print	Heading	758 (12%)	442 (8%)	76785 (10%)	101	174	358	285	3.47	3.85
	Text	2341 (39%)	2342 (43%)	161196 (21%)	69	69	434	395	3.01	3.17
	Caption	248 (4%)	278 (5%)	34229 (4%)	138	123	454	426	3.16	2.78
	All	3347 (55%)	3062 (56%)	272210 (35%)	81	89	418	382	3.13	3.23
Visual	Page no.	3 (0%)	2 (0%)	3470 (0%)	1157	1735	201	242	8.24	5.39
	Deictic	12 (0%)	9 (0%)	1305 (0%)	109	145	310	329	3.17	6.52
	Illustration	966 (16%)	992 (18%)	164052 (21%)	170	165	323	319	4.15	4.13
	All	981 (16%)	1003 (18%)	168827 (21%)	172	168	323	319	4.15	4.16
Blank	Blank	1757 (29%)	1437 (26%)	340283 (44%)	194	237	280	283	5.02	5.97

Number of Fixations

The fixation counts, as presented in the pixel-to-fixation ratios, in the print, visual, and blank AOIs in the pre-test were 81, 172, and 194, respectively; those in the post-test were 89, 168, and 237 (see Table 2). This means that in both the pre-test and post-test, the print AOIs had the highest density of fixations. The written language area is where the participants paid the most attention in both the pre-test and post-test.

When we compared the number of fixations in the AOIs across the pre-test and post-test by running ANOVA, we found that the headings were the only AOI that exhibited a significant difference (F[1, 52] = 4.22, p < .05). The pixel-to-fixation ratio of the headings in the pre-test was 101, and it was 174 in the post-test, meaning that significantly fewer fixations on heading AOIs occurred in the post-test than in the pre-test. In the experimental teaching sessions, students were taught about various text features and their functions, and we had expected to see more viewing of these text features in the post-test reading. However, the eye movement data indicated that in the post-test, the participants fixated on the headings less than they did in the pre-test. We speculate that when the participants read the same text for the second time, they did not navigate the various text features as much as they did when reading it for the first time. This might be especially true for headings because they are often used to preview the entire text.

Mean Fixation Duration

The mean fixation duration of the various AOIs in the pre-test was examined first. ANOVA tests indicated significant differences between print (418 ms), visual (323 ms), and blank (280 ms) (F[2, 6082] = 100.58, p < .001). Bonferroni post-hoc analyses revealed that the mean fixation duration of print was significantly longer than that of visual and blank, and that of visual was longer than blank, all p < .016. Results in the post-test were similar (F[2, 5499] = 53.85, p < .001). The mean fixation duration of print was longer than that of visual and blank while that of visual was longer than blank, all p < .016. Thus, in both tests, fixations on the written language were longer than those on visual areas and blank space. Even after receiving the instruction, the participants’ fixations on the written language were still longer than on the visuals in the post-test.

If we examine the pre-test and post-test and compare the mean fixation duration of each of the print, visual, and blank AOIs, the analysis reveals that the mean fixation duration of the print AOI in the pre-test (418 ms) was significantly longer than that in the post-test (382 ms) (F[1, 6407] = 16.65, p < .001). Therefore, in the post-test, students did not fixate on the print as much as they did in the pre-test.

Saccade Size

Saccade size data were analysed using the same method as for the mean fixation duration data. First, ANOVA tests were conducted to check for any significant differences among the print, visual, and blank AOIs in the pre-test and post-test, and then within the same AOI across the pre-test and post-test. The results were similar to those of the mean fixation duration. In the pre-test, significant differences were found among print (3.13 deg), visual (4.15 deg), and blank (5.02 deg) (F[2, 6058] = 97.30 p < .001). Bonferroni post-hoc analyses revealed that the saccade size of print was significantly shorter than that of visual and blank, and that of visual was shorter than blank, all p < .016. Results in the post-test were the same (F[2, 5479] = 128.10, p < .001). The saccade size of print was significantly shorter than that of visual and blank, and the saccade size of visual was shorter than that of blank, all p < .016. Longer fixations and smaller saccade sizes usually indicate print oriented reading behaviour. Eye movement information in terms of fixation duration and saccade size revealed that the participants focussed more on the written language than on the visuals in both the pre-test and post-test.

When we look across the pre-test and post-test and compare the saccade sizes of each of the print, visual, and blank AOIs, we find that the saccade size of the blank AOIs in the pre-test (5.02 deg) was significantly smaller than that in the post-test (5.97 deg) (F[1, 3311] = 16.31, p < .001). Other than the longer saccade size, the blank AOIs in the post-test also had a larger pixel-to-fixation ratio than in the pre-test (237 vs 194), although nonsignificantly. That is, the participants made fewer fixations on the blank AOIs in the post-test, and the fixations were further apart, suggesting that in the post-test, the participants were more efficient in guiding their eyes to the print and visual AOIs that they considered important.

Discussion

Students in this study reported positive responses to the experimental teaching. They had favourable impressions of the reading texts, instruction content, and teaching activities. A separate paper would be necessary to report the design and delivery of the experimental teaching, and therefore this paper focusses on the results of the teaching. The English reading and writing pre-test and post-test results revealed that the students performed significantly better after receiving 12 weeks of science text reading comprehension strategy instruction. Thus the science text and reading strategy instruction were effective. Other than recording the results of the reading and writing tests, we also examined what influences the strategy instruction had on how the participants read the science text. Eye movement information allowed us to see what the readers saw and what they were focussing on.

One major conclusion that can be drawn from the eye movement data is that students paid more attention to the print AOIs than to the visuals in both the pre-test and post-test. This finding is supported by the evidence that the print AOIs were hotter (darker) than the visual AOIs, as illustrated on the heat maps; the print AOIs had lower pixel-to-fixation ratios (i.e. a higher density of fixations); the mean fixation duration was larger; and the saccade size was smaller in the print AOIs in both the pre-test and post-test. The participants in this study were EFL learners, and before we began the research, we were unsure whether they would focus more on print sections that they were not confident of understanding or on visuals that provided nonlinguistic meaning. The findings showed that even after receiving strategy instruction that taught them how to use various text features and visual representations when reading science texts, the participants still spent more time on processing the text. Thus, they relied more on the text to understand meaning, even though the text was written in a language they were still learning. This print orientated reading of science texts is also reported in the eye movement reading research of Hung (2014), which involved Taiwanese Grade 5 students reading a science text written in their first language, Chinese. It is thus unsurprising to find that participants in this study, who were EFL learners, focussed primarily on the written language rather than the visuals in the post-test. The eye tracking research by Pellicer-Sánchez, Tragant, Conklin, Rodgers, Llanes and Serrano (2018) also discovered that without auditory support, both children and adult second language learners spent more time reading the text than the images.

The English reading and writing test results revealed that the EFL Grade 6 students in this study performed significantly better in the post-test. The eye movement data indicated that they paid more visual attention to the verbal language than the visuals in both the pre-test and post-test. These results are not contradictory. The integrated reading and writing test in Cambridge English for Young Learners does not target language skills required to read and write informational science texts. Although picture-prompted questions are often used, the test items do not generally require readers to understand and process various text features and visual presentations or to create mental images to achieve comprehension. The higher score in the post-test indicates that the science texts and the science reading comprehension strategy instruction in the experimental teaching were effective and helped the young EFL learners in this study to improve their English reading and writing ability. Nevertheless, 12 weeks of science text reading comprehension strategy instruction is probably insufficient to note a clear difference in the students’ viewing patterns and visual attention when reading science texts with complex print and visuals.

Another major finding from the statistical analysis of the eye movement data is that the mean fixation duration of the print AOIs in the post-test was significantly shorter than that in the pre-test. Given the mean fixation duration of the visual AOIs in the post-test was also shorter than that in the pre-test, though nonsignificantly, we consider this difference to be a result of the shorter reading time taken by the participants in the post-test. Moreover, the larger saccade size for the blank AOIs and the lower density of fixations on the headings in the post-test could also be attributable to the fact that the students read the same text for the second time in the post-test. When the students realized they were reading the same text as the one in the pre-test, they might have depended less on the headings to obtain a preview of what each text box or section would say. This meant they could also direct their eyes more efficiently towards text or visual areas they wanted to see or view, which would explain the longer saccade size in the post-test. Although using the same text for both the pre-test and post-test during eye movement data collection is convenient for the purpose of comparison, the cognitive processes and visual attention involved in reading a text for the first time are not the same as those involved in reading it for the second time, even if the two instances of reading are 12 weeks apart. For future research, different reading text selections and research design methods are suggested.

Preparing students to read from a book as well as digital devices is necessary. Many researchers have discussed the value of visual literacy in using modern digital technology (Arizpe and Styles, 2016; Hin and Subramaniam, 2009; Kress et al., 2005; Serafini, 2014). The current study comprises an initial effort to develop instructional reading strategies that can help students draw on non-verbal visual representations to enhance their comprehension. Future research should employ a longer experimental teaching intervention to more fully understand how science reading strategy instruction might assist EFL learners’ comprehension of multi-representational and multimodal science texts. More in-depth analyses of the utilization and integration of verbal and visual information should also be pursued to understand the complex viewing and comprehension processes of EFL learners reading science texts.

Conclusion

This study compared 27 Taiwanese Grade 6 EFL students’ English reading and writing performance and their visual attention when reading an English science text with multiple print and visual representations before and after 12 weeks of instruction in science text reading comprehension strategies. The results of English reading and writing proficiency tests indicated that the participants performed significantly better in the post-test, potentially confirming the effectiveness of the strategy instruction. However, due to the limitations of the research design, it is difficult to fully attribute the reading and writing proficiency growth to the strategy instruction. Nevertheless, we suggest that science texts, texts with interdisciplinary content, and comprehension strategies for reading these cross-disciplinary and multi-representational texts be integrated with EFL instruction. In future research, different comprehension strategies and learning activities should be incorporated in experiments. Furthermore, an experimental research design with a control group is also suggested.

The eye movement data gathered in this study also indicated that the students focussed most on the print AOIs, including the headings, text and captions in both the pre-test and post-test. After 12 weeks of strategy instruction, the students still paid more attention to the verbal language than the visuals. This finding is unsurprising, considering that English was a foreign language for the young participants in this study. Therefore, future research should employ a longer experimental teaching intervention to more fully understand how such instruction might influence young EFL learners’ use and integration of information from various text features and visual representations when reading science texts. In-depth analysis and investigation of how EFL readers utilize and integrate information from print and visuals is also suggested.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Ministry of Science and Technology of Taiwan under Grant MOST 106-2918-I-142-001.

References

Arizpe

Styles

(2016) Children Reading Picture Books: Interpreting Visual Texts (2nd Edition). London and New York: Routledge.

Conklin

Pellicer-Sánchez

Carrol

(2018) Eye-tracking: A Guide for Applied Linguistics Research. Cambridge: Cambridge University Press.

Davies

Greene

(1984) Reading for Learning in the Sciences. Edinburgh: Oliver & Boyd.

Doolittle

(2002) Multimedia learning: empirical results and practical applications. Paper presented at the Irish Educational Technology Users’ Conference, Carlow, Ireland.

Fang

(2008) Going beyond the fab five: helping students cope with the unique linguistic challenges of expository reading in intermediate grades. Journal of Adolescent & Adult Literacy 51(6): 476–87.

Hin

LTW

Subramaniam

(eds) (2009) Handbook of Research on New Media Literacy at K-12 Level: Issues and Challenges, Volume 1. Hershey and New York: Information Science Reference.

Hines

Wible

McCartney

(2010) Learning to read, reading to learn. Science 328(5977): 447.

Hoyt

Mooney

Parkes

(eds) (2003) Exploring Informational Texts: From Theory to Practice. Portsmouth, NH: Heinemann.

Hung

(2007) Critical issues in early foreign language literacy instruction: Taiwan experience. In: Goodman

Martens

(eds) Critical Issues in Early Literacy Development: Research and Pedagogy. Mahwah, NJ: Lawrence Erlbaum Associates, 263–271.

10.

Hung

(2014) What are you looking at? An eye movement exploration in science text reading. International Journal of Science and Mathematics Education 12(2): 241–260.

11.

Hung

(2017) Theme-based English instruction at the elementary school—A three year journey. In: Liu

(ed.) Theories and Practice of Teaching Materials and Methods [Jiao cai jiao fa li lun yu shi wu]. Taipei: National Taiwan University Center for Teacher Education, 109–125.

12.

Just

Carpenter

(1980) A theory of reading: from eye fixations to comprehension. Psychological Review 87(4): 329–54.

13.

Just

Carpenter

(1984) Using eye fixations to study reading comprehension. In: Kieras

Just

(eds) New Methods in Reading Comprehension Research. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc., 151–82.

14.

Kress

Jewitt

Bourne

, et al. (2005) English in Urban Classrooms: A Multimodal Perspective on Teaching and Learning. London and New York: Routledge Falmer.

15.

Kress

van Leeuwen

(2006) Reading Images: The Grammar of Visual Design (2nd ed.). London: Routledge Falmer.

16.

Lasagabaster

(2010) English achievement and student motivation in CLIL and EFL settings. Innovation in Language Learning and Teaching 5(1): 3–18.

17.

Mason

Tornatora

Pluchino

(2013) Do fourth graders integrate text and picture in processing and learning from an illustrated science text? Evidence from eye-movement patterns. Computers & Education 60(1): 95–109.

18.

Mayer

Anderson

(1992) The instructive animation: helping students build connections between words and pictures in multimedia learning. Journal of Educational Psychology 84(4): 444–52.

19.

Mayer

Gallini

(1990) When is an illustration worth ten thousand words? Journal of Educational Psychology 82(4): 715–26.

20.

McKay

(1999) An investigation of text-based instructional materials enhanced with graphics. Educational Psychology 19(3): 323–35.

21.

Miller

(1998) Visual persuasion: a comparison of visuals in academic texts and the popular press. English for Specific Purposes 17(1): 29–46.

22.

Moje

(2008) Foregrounding the disciplines in secondary literacy teaching and learning: a call for change. Journal of Adolescent & Adult Literacy 52(2): 96–107.

23.

Moss

(2005) Making a case and a place for effective content area literacy instruction in the elementary grades. The Reading Teacher 59(1): 46–55.

24.

Osborne

(2002) Science without literacy: a ship without a sail? Cambridge Journal of Education 32(2): 203–18.

25.

Palincsar

Magnusson

(2001) The interplay of firsthand and text-based investigations to model and support the development of scientific knowledge and reasoning. In: Carver

Klahr

(eds) Cognition and Instruction: Twenty Five Years of Progress. Mahwah, NJ: Lawrence Erlbaum Associates, 151–94.

26.

Paulson

Freeman

(2003) Insight from the Eyes: The Science of Effective Reading Instruction. Portsmouth, NH: Heinemann.

27.

Pearson

Gallagher

(1983) The instruction of reading comprehension. Contemporary Educational Psychology 8(3): 317–44.

28.

Pellicer-Sánchez

Tragant

Conklin

Rodgers

Llanes

Serrano

(2018) L2 Reading and Reading-while-listening in Multimodal Learning Conditions: An Eye-tracking Study. London: The British Council.

29.

Pladevall-Ballester

(2018) A longitudinal study of primary school EFL learning motivation in CLIL and non-CLIL settings. Language Teaching Research. Published online first. DOI: 10.1177/1362168818765877

30.

Rayner

(1995) Eye movements and cognitive processes in reading, visual search, and scene perception. In: Findlay

Walker

Kentridge

(eds) Eye Movement Research: Mechanisms, Processes, and Applications. New York: Elsevier Science, 3–21.

31.

Rayner

(1998) Eye movements in reading and information processing: 20 years of research. Psychological Bulletin 124(3): 372–422.

32.

Rayner

(2009) Eye movements and attention in reading, scene perception, and visual research. Quarterly Journal of Experimental Psychology 62(8): 1457–506.

33.

Rayner

Chace

Slattery

Ashby

(2006) Eye movements as reflections of comprehension processes in reading. Scientific Studies of Reading 10(3): 241–55.

34.

Rayner

Liversedge

(2011) Linguistic and cognitive influences on eye movements during reading. In: Liversedge

Gilchrist

Everling

(eds) The Oxford Handbook of Eye Movements. Oxford: Oxford University Press, 751–66.

35.

Samuels

Rasinski

Hiebert

(2011) Eye movements and reading: what teachers need to know. In: Samuels

Farstrup

(eds) What Research Has to Say about Reading Instruction (4th ed.). Newark, DE: International Reading Association, 25–50.

36.

Sarafini

(2014) Reading the Visual: An Introduction to Teaching Multimodal Literacy. New York and London: Teachers College Press.

37.

Shanahan

(2008) Teaching disciplinary literacy to adolescents: rethinking content-area literacy. Harvard Educational Review 78(1): 40–61.

38.

Unsworth

Thomas

Bush

(2004) The role of images and image-text relations in group ‘Basic Skills Tests’ of literacy for children in the primary school years. Australian Journal of Language and Literacy 27(1): 46–65.

39.

Winke

Godfroid

Gass

(2013) Introduction to the special issue: eye-movement recordings in second language research. Studies in Second Language Acquisition 35(2): 205–12.