The Effects of Textual Enhancement and Topic Familiarity ...



The Effects of Textual Enhancement and Topic Familiarity on Korean EFL Students’ Reading Comprehension and Learning of Passive Form

Sang-Ki Lee

University of Hawai‘i at Mānoa

This quasi-experimental study attempts to incorporate grammatical elements into meaning-focused reading classes by attracting learner attention to form with minimal interruption to meaning comprehension. Two hundred and fifty-nine Korean EFL students underwent four different treatments—involving textual enhancement and topic familiarity conditions. The responses of the students were compared with respect to (1) their ability to identify and correct English passive errors and (2) their degree of reading comprehension. The results revealed that textual enhancement aided the learning of the target forms, while having unfavorable effects on meaning comprehension. Topic familiarity, by contrast, aided the students’ comprehension, but was ineffective in terms of their learning of form. I discuss the findings in light of the theoretical relationship between acquisition and comprehension and conclude with implications for meaning-oriented lessons.

The widespread reaction against form-oriented language instruction (e.g., Grammar Translation Method, Audiolingual Method, Total Physical Response, etc.) in the 1980s prompted a shift in preference towards meaning-oriented pedagogies (e.g., Communicative Teaching, Immersion Instruction, the Natural Approach, Content-based Instruction, etc.). As a result, language teachers have been urged to pursue the goals of communicative competence and fluency, while grammar instruction that does not contribute to communicative goals has been regarded as counterproductive. However, the 1990s brought some doubts to the validity of teaching approaches that focus primarily on meaning. These concerns have stemmed mainly from studies examining the French immersion situation in Anglophone Canada, which since the mid 1980s documented students’ difficulty with the accurate usage of the second language (L2) even after undergoing a substantial period of study in a meaning-first program (Day & Shapson, 1991; Hammerly, 1987; Harley, 1993; Harley & Swain, 1985; Swain, 1985; Swain & Lapkin, 1989). Such an observation has led instructed second language acquisition (SLA) researchers to investigate how to integrate grammar instruction efficiently without hindering the aims of meaning-oriented instruction. One solution that was introduced in the 1990s to address this problem is focus on form, an approach for teaching grammar in which learners’ attention is drawn briefly to a linguistic form in a meaningful context (Long, 1991; Long & Robinson, 1998). One of the fundamental assumptions of focus on form is that learners’ attention can be allocated to the learning of form as well as content.

Many SLA researchers have reached a general consensus that paying focal attention to items in the input is a necessary condition for its sustained processing (Leow, 1999; Long, 1991; Robinson, 2003; Rosa & O’Neill, 1999; Schmidt, 2001; Sharwood Smith, 1993; Tomlin & Villa, 1994; VanPatten, 1990, 1996, 2002). Yet, the actual psycholinguistic mechanisms that would explain the relationship between learner attention and L2 learning continue to be debated. VanPatten (1990, 1996, 2002) predicted that learners’ limited attentional resources would be selectively directed at meaning before form. His 1990 study suggested that it is difficult, especially for beginning level L2 learners, to attend to both form and meaning at the same time. Further studies investigating his prediction have found that a competitive relationship exists between a learner’s capacity to attend to form and his or her ability to attend to meaning (Bransdorfer, 1991; Greenslade, Bouden, & Sanz, 1999; Wong, 2001). VanPatten’s (1996, 2002) theory of input processing, the predictions of the limited capacity of human attention (e.g., Skehan & Foster, 2001), and the psycholinguistic rationale of focus on form (Doughty, 2001) caution that excessive intervention may not guarantee learners’ comprehension of meaning and, therefore, suggest that the right amount of obtrusiveness might be crucial to balancing the instructor’s intention to teach form with the learner’s pursuit of meaning comprehension.

Textual enhancement was chosen in the present study as an optimal intervention point for fulfilling the objective of focus on form. This choice was based on previous research that reported a positive role for textual enhancement in promoting learners’ attention to form (Alanen, 1995; Doughty, 1991; Jourdenais, Ota, Stauffer, Boyson, & Doughty, 1995; Shook, 1994; White, 1998). As a less obtrusive technique of focus on form, it is likely to be less burdensome on learners’ attention to meaning than other more explicit techniques also available under the focus on form approach (Doughty & Williams, 1998; Sharwood Smith, 1981; White, 1998). In contrast to natural reading contexts, where learners tend to prioritize meaning before form for comprehension (Skehan, 2003; Skehan & Foster, 2001), the interventions used within the context of this study pose learners the additional demand of drawing their attention to form in addition to their primary task of meaning comprehension. Thus, when reading enhanced texts in their L2, the L2 learners are inherently engaged in a quasi-dual-task: on the one hand, they are reading for content, while on the other they are reading grammatical forms for acquisition.1 It is noteworthy that this is not the case with native speakers. Their knowledge of language forms is proceduralized, and thus “can be executed automatically without requiring much, if anything, in the way of attentional resources” (Skehan & Foster, 2001, p. 189). In sum, the present study addresses one of several questions that have not been satisfactorily answered to date: How can learners cope with quasi-dual-task situations where they must attend to form and meaning simultaneously? The question has significance not only for the focus on form agenda that posits the feasibility of pedagogical interventions with a specific grammar target, but also for curricula that aim largely at learners’ understanding of content, rather than of language itself.

Another way to address the attentional imbalance created by the tension between form and meaning during L2 processing might be to employ culturally familiar topics. According to common assumptions in the L1 and L2 reading literature, text reading is an interactional process between a reader and the text (Rumelhart, 1977), and familiar topics are able to bridge the reader and the text by providing prior knowledge (or content schemata; see Carrell, 1987). Thus, familiar topics are thought to promote the reader’s comprehension as well as his or her retention of textual material (Anderson & Pearson, 1984; Johnson, 1981, 1982; Pritchard, 1990; Reynolds, Taylor, Steffensen, Shirley, & Anderson, 1982; Weber, 1991). Building on these rationales and integrating schema theory of reading with SLA information processing theories of attention, it is possible to hypothesize that the ease of comprehension achieved through a familiar topic will free up attentional resources away from meaning and make them available for attending to form.

Empirical Studies with Textual Enhancement

Textual enhancement, an implicit technique of focus on form that uses visual enhancement methods such as underlining, bold facing, color-coding, italicizing, CAPITALIZING, using different fonts, has been investigated as a means to promote the processing of linguistic items in a number of studies, summarized in Table 1 (see also Simard, 2002, pp. 250-251, for a review of the literature up to 2002). Five of the thirteen studies reviewed in Table 1 yielded results that support positive effects of textual enhancement on grammar learning2 (Alanen, 1995; Doughty, 1991;3 Jourdenais et al., 1995; Shook, 1994; White, 1998). The remaining eight, however, failed to reach the same conclusion (Jourdenais, 1998; Izumi, 2002; Leow, 1997, 2001; Leow, Egi, Nuevo, & Tsai, 2003; Overstreet, 1998, 2002; Wong, 2003). Some researchers pointed out that textual enhancement alone might not be sufficient for interlanguage development and argued that other supplementary instructional elements should be added to this technique. For example, White (1998) suggested the accompaniment of more explicit information and Izumi (2002) recommended output activities for further processing and subsequent learning.

-------------------------------------------------

Table 1 here

-------------------------------------------------

While Jourdenais et al. (1995), Jourdenais (1998), and Leow (2001) explored the effects of textual enhancement as an independent variable with no other intervening variables, the other studies in Table 1 investigated textual enhancement in combination with other variables. According to the available empirical evidence, explicit directions to pay attention to form (Shook, 1994), text length (Leow, 1997), topic familiarity (Overstreet, 1998), extensive reading and listening (White, 1998), saliency of target forms (Leow et al., 2003), and simplified input (Wong, 2003) have been discovered to have no statistically significant effects on learners’ form intake or ensuing acquisition. In contrast, rule instruction (Alanen, 1995) and output tasks (Izumi, 2002) were proven to be beneficial for learners’ form acquisition. With regard to the effects of these moderating variables on meaning comprehension, these same studies found that text length and simplified input had main effects on comprehension, but that topic familiarity and saliency of target forms did not have significant effects. In addition, Overstreet (2002) found that although the modality of enhancement (whole-word vs. morphology-only enhancement) had no effect on either meaning comprehension or form processing, the relative communicative value of the targets had a significant main effect on the intake of forms. Of particular relevance for later discussion is to note that Overstreet (1998) adopted the same two independent variables as in the current study, and that his findings regarding topic familiarity contradicted the assumptions of both schema theory of reading and the theory of input processing (VanPatten, 1996, 2002).

Textual Enhancement vs. Comprehension

As White (1998) has pointed out, textual enhancement is considered more explicit than input flooding but less explicit than rule explanation (p. 86). In terms of learners’ attention to meaning, the relative unobtrusiveness of textual enhancement would be empirically demonstrated if comprehension (as well as language gains) is tested and no adverse results are found. Wong (2003) properly stated that “the role of TE [textual enhancement] in SLA cannot be complete without information about how comprehension is affected (or not affected) as learners’ attention is directed at form” (p. 21). Yet, the available evidence is inconclusive. While some researchers reported no debilitating effects on comprehension (Jourdenais, 1998; Leow, 1997, 2001; Overstreet, 2002; Leow et al., 2003; Wong, 2003), the finding reported by Overstreet (1998) indicated that textual enhancement may hinder students’ comprehension processing. The bulk of extant studies, on the other hand, have not given adequate consideration in their design to the impact of textual enhancement on learners’ comprehension (Alanen, 1995; Doughty, 1991; Izumi, 2002; Jourdenais et al., 1995; Shook, 1994; White, 1998).4 Thus, it is important to further investigate directly and by design whether inducing learner attention to a linguistic form via typographical enhancement compromises comprehension.

Task Specificity

The wide range of results obtained in previous studies may be better understood if the task specificity of each study is carefully inspected. As Leow et al. (2003) recently put it, “Task types and task demands, which are learner-external factors that may constrain noticing … are clearly variables that need to be considered” (pp. 10-11). Table 1 indicates that diverse tasks have been used to measure acquisition (or intake) as well as comprehension. Tasks frequently used to assess the processing of form encompass grammaticality judgment tasks (Alanen, 1995; Doughty, 1991; Izumi, 2002), sentence completion tasks (Alanen, 1995; Doughty, 1991; Izumi, 2002), fill-in-the-blank tasks (Shook, 1994; Leow, 2001), multiple-choice form recognition tasks (Leow, 1997, 2001; Leow et al., 2003; Overstreet, 2002; Shook, 1994;), form correction tasks (White, 1998; Wong, 2003), and think-aloud protocols (Alanen, 1995; Jourdenais et al, 1995; Leow, 2001; Leow et al, 2003).5 Among these, form correction tasks have an advantage in that they may prevent learners from random guessing by asking them to supply (their own) correct forms to incorrect sentences (Wong, 2003, p. 23). This would require students to demonstrate their linguistic competence to switch their attention from meaning to form as they identify errors and correct them. With regards to the measures of comprehension, commonly used tasks in the previous literature are multiple-choice tests (Jourdenais, 1998; Leow et al., 2003), short-answer type tests (Doughty, 1991; Leow, 1997), and true or false type tests (Overstreet, 1998). All these types of tests seem to fail in effectively assessing the learners’ comprehension, for they can be taken after only a shallow reading of the text and may allow the learners to make wild guesses without deeper comprehension (or with no comprehension at all; see Bernhardt, 1983). Overstreet (2002) claimed that “[a free recall task] will best test the subjects’ ability to recall the information in the text by allowing them to recall anything and everything they can without any researcher bias created by test questions” (p. 47).6 Taking these considerations into account, the present study chose to adopt a form correction task for assessing learners’ acquisition and a free recall task for assessing comprehension.

Research Questions

The present study examined the feasibility of incorporating grammatical elements into meaning-focused reading classes by attracting learner attention to form with minimal interruption to meaning comprehension. Textual enhancement was employed as an appropriate focus on form technique for the allocation of learner attention to both form and meaning. As a second variable, the role of familiar topics was explored as an additional aid to the successful quasi-dual-task performance of learners. This study was motivated by the following three questions:

1. What are the effects of textual enhancement on the students’ acquisition of form (as measured by a form correction task) and comprehension of meaning (as measured by a free recall task)?

2. What are the effects of topic familiarity on both acquisition of form and comprehension of meaning?

3. What is the relationship between performance on the correction task and comprehension scores, if any?

Based on theoretical and empirical rationales of focus on form, I expected that students experiencing the textual enhancement condition would have their attention more optimally directed to the enhanced passive verb forms and hence would perform better on the post-test correction task than students experiencing the same texts in baseline form. In addition, I hypothesized that these students’ attentional resources would be depleted from having to decode the enhanced parts, resulting in a poorer understanding of the meaning.

With regard to the second research question, and given the rationale provided by the schema theory of reading and by empirical results in L1 and L2 reading studies, I predicted that the groups given culturally familiar topics would recollect more idea units in the free recall task than the groups given culturally less familiar topics. In addition, I expected that the processing of a text with a familiar topic would demand less attentional resources than that with a less familiar topic, making more cognitive resources available for attending to form. In other words, I predicted that the benefits of topic familiarity would not only result in better comprehension but also in better correction scores on the post-test correction task. Extending the same rationale to research question 3, I expected that there would be a correlation between correction and comprehension scores.

Method

Participants

Twelve intact classes in four different schools located in Seoul City and Kyung-gi province participated in this quasi-experimental study. They comprised a total of 259 (grade 11) high school juniors and their four teachers.7 At the time of the research, most of these students were 17 years old, and had begun learning English as a foreign language approximately four years earlier, in middle school (grade 7). As is typical of the Korean EFL context, the participants’ English learning happened primarily via four to five hours of classroom instruction per week, in classrooms where students’ reading skills and explicit grammar teaching are mostly emphasized, due to logistical reasons such as a shortage of native-speaking teachers. They had previously been taught the target structure, namely, the passive voice in English, through explicit rule presentations by their teachers, but pre-test results (M = 2.20, out of 10 possible points, SD = 1.88) showed that they had failed to internalize the target form.

Reading Materials for Treatment Sessions

Whereas the previous studies reporting no significantly beneficial effects of textual enhancement (see Table 1) generally gave their participants one period of exposure to the enhanced texts, the students in the 12 intact classes participating in this study were provided with three exposure sessions in total. Experiencing multiple exposure sessions enabled the students to become accustomed to the reading conditions of typological enhancement. In addition, the relatively ample exposure to the target forms was expected to facilitate a more efficient allocation of their attentional resources during form processing.

The reading text for the first exposure session was an expository passage about Korean Jokpo, or a genealogical record of important historical events and achievements of ancestors. The passage for the second exposure session briefly dealt with the physical process of mummification.8 In these first two exposure sessions, all participants experienced the same texts containing the same number of tokens of the target form, but half of them experienced a typographically enhanced version and half of them a baseline version. For the third exposure session, two expository texts differing in the extent of familiarity were administered. Thus, in this final exposure session, a fourth of the participants each experienced a familiar or an unfamiliar topic with or without enhancement. A short descriptive passage about the birthday celebration customs of Korea was selected as the familiar text, since these customary feasts are a basic part of daily Korean life. The less familiar passage discussed the traditional Egyptian belief that humans were composed of three essential elements to be preserved for eternal life. Two steps were taken to ensure that these two experimental texts were equivalent in all respects except for the degree of topic familiarity. The researcher and two other reading research experts adjusted reading difficulty levels by carefully considering the complexity of structures as well as vocabulary choices of the two texts. In addition, using readability measures in Microsoft Word both texts were modulated so as to fit the 7th Flesch-Kincaid grade level, meaning that the texts were appropriate for the 7th graders in the American school system (7.8 for the familiar passage and 7.7 for the less familiar passage).9 The total number of words was 309 for the familiar passage and 315 for the less familiar passage.

For the enhanced texts in all three sessions, the words containing the target forms were placed in larger, boldfaced letters in different fonts to operationalize textual enhancement. Students encountered 12 tokens of the target form in each text, for a total of three texts, one per session. Four of the 36 total tokens were duplicated up to four times across the three texts and the rest of the tokens (27 for familiar conditions and 25 for less familiar conditions) were single instances (see Appendix A for the complete enhanced familiar and less familiar texts).

Assessment Measures

L2 reading proficiency test. In order to ensure that participants were at comparable levels as L2 readers, a reading proficiency test was administered at the onset of the study. The test items were sampled from the 2002 National Assessment of Education Achievement, a measure used to rate the scholastic achievement levels of Korean high school students.10 The test contained 20 items, most of which asked test takers to identify the main ideas in a text and sort out correct and incorrect descriptions of reading contents.

Form correction tasks. Constructing the passive voice in English is relatively simple and straightforward: be + -(e)d. However, many Korean EFL learners tend to omit either the be or -(e)d in their passive construction, and they often have difficulty with the appropriate usage of adjectival participles (Kim, 2002; Yeo, 2002).11

Each form correction task contained 10 items associated with passive-related form errors plus 5 distracter items. Table 2 describes the three types of form errors included in the task. Type A incorrectly uses a present participle instead of a past participle; in Type B, a bare verb form is used in place of a past participle; in Type C, the auxiliary be is missing. Pilot-testing of materials and procedures with 53 participants drawn from the main sample population indicated that the pre- and post-test versions of the form correction tasks to be used in the main study had comparable difficulty levels (t = .53, df = 52, p = .60).

-------------------------------------------------

Table 2 about here

-------------------------------------------------

Students were awarded one point for each appropriate correction of the passive-related form errors. The students’ responses were deemed acceptable when it was apparent that they understood the target forms well enough, even if they gave a different response from the expected answer. For instance, in response to the question item “The solution to that question can explained only in this way,” both the fully accurate response “The solution to that question can be explained only in this way” and the less accurate response “The solution to that question was explained only in this way” were accepted. The rationale was that both responses equally demonstrate students’ knowledge of the target forms (although the second one has been changed from present to past tense).

Free recall task. In order to assess students’ understanding of the text content, a free recall task was administered immediately after the third reading task (the Korean birthday celebration passage for half of the participants and the Egyptian beliefs passage for the other half). Students were directed to write down every idea they could remember from what they had read, from general ideas to smaller details, without referring back to the reading or adding any non-textual information. The students were allowed to write in Korean, their L1, in order to avoid a situation in which the pressure of an L2 writing task would discourage them from producing recalled ideas (for detailed suggestions on the use of L1 recall protocol for L2 readers, see Lee, 1986; and Wolff, 1987).

The idea units of the two (familiar and unfamiliar topic) reading texts were analyzed following Carrell’s (1985) criteria. Great care was taken to make the necessary adjustments to analyze the idea units of the recalls in Korean. This analysis was done through an iterative process of discussion and consensus by three raters, who established the total number of idea units in English using pilot data and later determined the corresponding possible total number of idea units in Korean for the main study. Each reading text contained 33 L1 Korean idea units (see Appendix B, for samples of idea unit analysis). Participants received one point for each L1 Korean idea unit in their recall protocol that was clearly judged to be related to the pre-analyzed L1 English idea unit in the text.

Procedures

The study was conducted from the first week of March 2003 to the end of the month with the consent of the students as well as their teachers. Each intact class participated in the study over a two-week period during four regular 50-minute English class sessions. Table 3 displays the overall design of the study.

-------------------------------------------------

Table 3 here

-------------------------------------------------

The 12 intact classes were randomly assigned to one of the four treatment groups, according to the two manipulated variables of Enhancement (+/-E) and Topic Familiarity (+/-F): -E/+F (n = 62), -E/-F (n = 65), +E/+F (n = 70), and +E/-F (n = 62). In the first session, two pre-tests (i.e., L2 reading proficiency test and form correction task) were administered to measure the comparability of the groups. The overall results are shown in Table 4. A one-way Analysis of Variance (ANOVA) revealed that the four groups were not statistically different from one another in terms of their general reading ability, F(3, 255) = .35, p = .79; neither did they differ from one another in their performance on the form correction task at the onset of the study, F(3, 255) = 1.01, p = .39. Therefore, any measurable changes in the post-test were unlikely to be induced by any preexisting differences among the groups and instead could be attributed with a reasonable degree of confidence to the different treatments that the various groups experienced.

The second and third class sessions were devoted to the first and second exposures. Students individually read the given passage and immediately discussed the content of the passage in a whole-class-discussion format. By contrast, on the last day of the experiment, which was devoted to the third exposure session, students individually read the (familiar or less familiar topic) passage and then took the post-tests (i.e., a 15-minute free recall task and a 10-minute form correction task). The total exposure to the targeted forms lasted 60 minutes in total (i.e., 20 minutes by three exposure sessions). This represents 30 percent of the total instructional time over the period of the study (i.e., 60 / 50 X 4 = 30%).12 During these sessions, when questions about the intention of the visually enhanced parts in the texts were raised by any students in the enhancement conditions, their teachers just informed them that some careful attention should be allocated to them. Also, teachers were instructed to avoid any reference or mention to the targeted forms during the two-week period of the experiment.

-------------------------------------------------

Table 4 about here

-------------------------------------------------

Results

A two-way ANOVA was conducted, with textual enhancement and topic familiarity as the independent variables and the two post-test scores (correction task and free recall task) as the dependent variables, to determine whether there were any significant effects of the two main variables on the students’ performance. In addition, the students’ scores on the pre- and post-tests of the form correction task were compared by a paired t-test to determine the significance of the difference in their performance before and after the treatments. All the statistical analyses were carried out using SPSS ver. 10.0 with the level of significance set at 0.05.13

Results of the Post-Treatment Form Correction Task

As shown in Table 5, the +E/-F group scored best in the form correction task, and the students with enhanced texts (the two +E groups) performed better than those with the baseline version texts (the two -E groups). A two-way ANOVA indicated a significant main effect for the enhancement condition, F(1, 255) = 65.37, p = .00, whereas the effect for topic familiarity was negligible, F(1, 255) = .39, p = .53. There was no interaction effect, F(1, 255) = 2.56, p = .11. Moreover, a post hoc Least Significant Difference (LSD) test confirmed that significant differences existed only between the enhancement condition groups and the baseline condition groups. There was no significant difference between the -E/+F and -E/-F groups and between the +E/+F and +E/-F groups. The magnitude of the effects was inspected by calculating Cohen’s d for the reported ANOVA contrasts. The results further support the statistical significance outputs: The enhancement condition resulted in a large effect size (d = 1.02), whereas the effects of topic familiarity and the interaction effects were negligible (d = 0.10 and d = 0.20).14

-------------------------------------------------

Table 5 about here

-------------------------------------------------

These findings were further corroborated by pre- and post-test comparisons of the same data. Overall, compared with their pre-test scores (M = 2.20), students appeared to perform better on the post-test (M = 3.72). When pre- to post-test changes for each of the four groups are compared visually, in Figure 1, the lines of the two +E groups have noticeably steeper slopes than those of the two -E groups.

-------------------------------------------------

Figure 1 about here

-------------------------------------------------

Paired t-tests were conducted to determine the statistical significance of the students’ score changes from the pre-test to the post-test. As shown in Table 6, the two +E groups improved statistically significantly in the post-test of the form correction task. Conversely, there were no measurable differences between pre- and post-test scores for the two -E groups that read the baseline texts.

-------------------------------------------------

Table 6 about here

-------------------------------------------------

Cohen’s (1988) pre-to-post d effect sizes showing the standardized mean difference between pre-test and post-test performance were also calculated and are shown in Table 6.15 Clearly, the magnitude of pre-to-post change was much larger for the +E/+F (d = 0.99) and +E/-F (d = 1.23) groups than for the -E/+F (d = 0.20) and -E/-F (d = 0.17) groups.

Results of the Free Recall Task

The overall results of the four groups’ recall are shown in Table 7. The groups who were given the familiar topic texts (the two +F groups) scored better than the groups who read the less familiar topic texts (the two -F groups). In addition, the students with the baseline texts (the two -E groups) performed better than those with the enhanced texts (the two +E groups).

-------------------------------------------------

Table 7 about here

-------------------------------------------------

A two-way ANOVA revealed that the students’ recall was affected statistically significantly by both textual enhancement, F(1, 255) = 34.78, p = .00, and topic familiarity, F(1, 255) = 115.33, p = .00. The interaction of textual enhancement and topic familiarity, while not statistically significant, approached significance, F(1, 255) = 3.83, p = 0.051. The results of the LSD analysis demonstrated that the four groups differed statistically from one another in their recall performance. Figure 2 visually shows the statistical significance results in the form of 95% confidence intervals (CIs) around each group mean for the free recall task scores. The means in Figure 2 reveal the exact average value for each group, and the CIs around the means show where the true mean may fall at 95% probability (see discussion of CI interpretation in Norris & Ortega, 2000; Thompson, 2002). The CIs for the four group means do not overlap at all, indicating the trustworthiness of the observed mean differences among the four treatment groups. As with the results of the correction performance, Cohen’s d was calculated for the reported ANOVA contrasts on the recall performance. The magnitude results further support the statistical significance outputs: The textual enhancement condition resulted in a detrimental medium-sized effect on recall scores (d = -0.57), whereas topic familiarity resulted in a sizable benefit in comprehension (d = 1.24), and the effect size for the statistically not significant interaction between the two variables was small (d = 0.25).

-------------------------------------------------

Figure 2 about here

-------------------------------------------------

Correlations Between Correction Scores and Recall Scores

Table 8 summarizes the correlation coefficients between the students’ correction task scores and recall task scores, which show the effects of textual enhancement and topic familiarity from a different perspective.

-------------------------------------------------

Table 8 about here

-------------------------------------------------

As indicated in Table 8, the two -E groups who read baseline texts with no manipulations showed a statistically significant positive correlation between their form correction scores and recall scores (r = 0.42 and 0.35). In other words, the students who, after reading the baseline texts, performed better in the recall task tended to perform better in the form correction task as well. Conversely, in the case of the two +E groups, no statistically significant correlations were found, suggesting that the two sets of scores were dissociated. In other words, the students who comprehended more after reading the enhanced texts did not necessarily correct better (or worse) than the students who comprehended less under the same experimental condition of enhancement.16

Discussion

Based on the theoretical rationale of focus on form and on empirical evidence from previous research (Alanen, 1995; Doughty, 1991; Jourdenais et al., 1995; Shook, 1994; White, 1998), I predicted that textual enhancement would make certain parts of the texts perceptually salient and that, consequently, the students who experienced the textual enhancement condition would have closer attention directed to the enhanced passive verb forms. The statistical analyses revealed that textual enhancement brought about better performance on the form correction task. Students in the textual enhancement condition (M = 4.91, out of 10 possible points) outperformed the others (M = 2.48) by about one standard deviation unit (d = 1.02). Therefore, it can be inferred that textual enhancement aided the students to attend to the target grammar. While this result supports the prediction, it runs counter to a number of studies that reported no such facilitative effects (Jourdenais, 1998; Izumi, 2002; Leow, 1997, 2001; Leow et al., 2003; Overstreet, 1998, 2002; Wong, 2003).

A variety of factors might have intervened in the present study, as well as in the previous studies, which makes it a delicate task to compare each study on the same basis. These factors include, among others, measures of attention, types of enhancing methods, characteristics of the targets, and nature and duration of the treatment conditions. In addition, important differences can also be found in the age of the learners, program differences, learners’ previous learning experience, L1/L2 differences affecting learnability of target features, and learner readiness. Two key differences merit close consideration. One important difference between the present study and previous ones is that in the present study the participants had multiple opportunities to experience the treatment conditions, whereas many of the previous studies offered one short period of exposure to the enhanced targets. Specifically, the Korean high school students in this study had three periods for familiarization with the structure (for all conditions) and with textual enhancement as a technique (for the +E conditions), involving three different texts, 36 tokens, and about 60 minutes total of opportunity for exposure to the targets. It may be that the lengthier, more intensive treatment more effectively drew their attention, and hence strengthened their learning of the passive forms to a degree that was not possible in the briefer treatments featured in previous studies. Another main source of difference is the distinctive learning experience of the Korean participants with the target forms. Although the students’ knowledge of the passive forms in English did not seem to be stable at the time of the pre-test, this target form had been taught to them since a relatively early stage of their learning and had been presented to them repeatedly as an important rule in English (the passive construction is first introduced with explicit rule explanations in 8th grade school textbooks). In contrast, the previous studies that did not witness any meaningful effects of textual enhancement either (1) targeted structures that were entirely or relatively new to the students (Leow, 1997, 2001; Leow et al., 2003; Wong, 2003) or (2) had participants who were in their first or second year of learning the target language (Jourdenais, 1998; Overstreet, 1998, 2002). As Jourdenais (1998) put it, “the implicit nature of the enhancement … was more likely to be beneficial to learners who already had some initial awareness of the forms and their use” (p. 92).

The present study also found that the students who were presented with the enhanced texts recalled statistically significantly less (M = 8.39, out of a maximum 33 idea units) than the students who were given the baseline texts (M = 12.46), while the magnitude of the negative mean difference was medium-sized (d = -0.57). This result is similar to Overstreet (1998), who also reported detrimental effects of textual enhancement on comprehension, but runs counter to a number of studies that reported no such negative effects (Jourdenais, 1998; Leow, 1997, 2001; Overstreet, 2002; Leow et al., 2003, Wong, 2003). The students could have been distracted by the visual elements while exposed to the enhanced texts. Such behavior appears to be traceable to the limited and selective nature of attention. Following VanPatten’s (1996, 2002) theory of input processing, the students’ attentional resources may have been depleted from having to decode the enhanced parts, resulting in a poor understanding of the meaning. The statistically significant lower comprehension scores for the enhanced groups warn teachers that when the goal of the lesson is to learn the content, comprehension may be sacrificed as students must pay focal attention to linguistic elements. Research evidence to date informs us that acquisition of linguistic forms does not always happen naturally without instruction (Norris & Ortega, 2000). However, assumptions regarding the limited and selective capacity of human attention caution that excessive intervention may also be undesirable in the meaning-oriented content-based lessons. This study demonstrates that even the relatively less obtrusive intervention of textual enhancement (Doughty & Williams, 1998; Sharwood Smith, 1981; White, 1998) may be unfavorable for effectively drawing learner attention to meaning. Future studies, then, should deliberately strive for the right balance both in terms of obtrusiveness and acquisitional benefits of attention-drawing techniques.

Given the robust rationale provided by the well-established schema theory of reading and by numerous empirical results in L1 and L2 reading studies, it was reasonable to expect that the groups given culturally familiar topics would remember more idea units in the free recall task. The results supported this hypothesis. The 132 Korean EFL students who read the familiar topic texts exhibited far better recall (M = 14.20, out of a possible total of 33 idea units) than the 127 peers in the two -F groups (M = 6.43), who were at a similar level of overall L2 reading proficiency but worked with less familiar topics. The mean difference was statistically significant and its size was very large (d = 1.24). In addition, based on VanPatten’s (1996, 2002) claim that attention is consumed towards L2 meaning before it can be directed towards L2 form, I predicted that the familiar topic groups would earn better scores in the form correction task as well as in the comprehension recall task. This is because a text with a familiar topic would demand less attentional resources than that with a less familiar topic, leaving more cognitive resources available for allocation to the L2 form. This in turn would foster more noticing and therefore lead to larger benefits on the correction task. The results, however, did not support this hypothesis. With the aid of background knowledge, the students presumably did not have to spend many resources on text meaning comprehension. Even so, the results demonstrate that the unconsumed resources were not reallocated towards the target forms.

Several explanations can be invoked to account for this part of the results. It might be argued that textual enhancement was not strong enough to draw the leftover resources. However, this interpretation seems to render a self-contradiction; that is, if textual enhancement was strong enough to induce attentional resources overall (as shown in the large effects on acquisition of target forms), it should logically have exerted an influence on the remaining resources of equal or at least measurable strength. Alternatively, it is plausible to raise the possibility that comprehension of meaning and acquisition of linguistic forms are independent and dissociated processes. This interpretation is reinforced by the observed weak relationship between the scores on the form correction and on the free recall for the enhancement groups, a correlational pattern that does not resonate with that obtained for the baseline groups (cf. Table 8). These results, together with the overall positive effects of textual enhancement on the correction task, indicate that enhancement had the effect of making the forms salient to a point where intake was facilitated, but independently from how well the text was understood. This dissociation between acquisition and comprehension has been indirectly attested in previous studies. For example, both Doughty (1991) and Loschky (1994) found that experimental groups who experienced various meaning and form-focused conditions comprehended to differing degrees but learned to similar extents.17 From the standpoint of a relative dissociation between comprehension and acquisition components of processing, it may be that topic familiarity facilitated learners’ focus on meaning within the component of comprehension, whereas textual enhancement may have been involved in both comprehension and acquisition components. Specifically, by conditioning a quasi-dual-task situation, textual enhancement may have entailed both a decrease in the amount of resources necessary for form processing and an increase of resources required for meaning processing. Therefore, it is possible that learners’ attentional resources for meaning comprehension were nontransferable to the acquisition component.

Numerous factors constrain the generalizability of the interpretations put forth, of which the most obvious are the focus on one single type of attention-drawing technique and one single target rule, and the specific characteristics of the participants and the tasks. In addition, the lack of a follow-up study for gauging the durability of the observed outcomes makes it difficult to know the extent to which the attention that was allocated to textually enhanced passives led to lasting changes in the learners’ interlanguage. Thus, we will need more research that explores the relationship among attention, comprehension, and acquisition and that includes a wider range of instructional interventions and L2 targets. The generalizability and scope of the present findings will be better understood if future studies are conducted with similar as well as varying population and tasks. Most needed are future studies that document long-term effects of enhancement on the developing L2 system.

Conclusion

This study generates important theoretical implications that provide further insight into how learners utilize attentional resources when faced with textual enhancement as a type of focus on form instructional intervention. It was found that a rich-exposure regime of textual enhancement aided these L2 learners to attend to formal aspects of the English passive voice, a grammatical target which they had already encountered during years of instruction but which they needed yet to master. It was also found, however, that these students’ attentional resources may have been depleted from having to decode the enhanced parts, resulting in a poorer understanding of the meaning of the texts. As expected, participants in the enhanced condition who read culturally familiar topics were able to understand the content of the reading better, but the resources freed-up from comprehension were not reallocated towards the attending to or learning of the target forms, at least as measured by the form correction post-test. This pattern of results thus suggests two things. One, and in agreement with predictions from information processing theory espoused by VanPatten and Skehan, among others, L2 learners’ meaning processing may be undermined when their attention is drawn to formal aspects of input, for they may have difficulty distributing their resources simultaneously in the two directions of form and meaning. Two, evidence was found of a possible dissociation between comprehension and acquisition processes, in that enhancement may have the effect of making the forms salient to a point where intake is facilitated independently from how well meaning is understood.

With reference to grammar pedagogy, the present study offers some evidence that textual enhancement would be an effective focus on form technique, at least for the target structure and the EFL population investigated. More accurately stated, however, and considering that textual enhancement was a follow-up to the previous explicit rule instruction that the learners had had over a period of several years, the findings bolster a recommendation for some combination of explicit instruction plus textual enhancement. On the other hand, although textual enhancement is credited in the SLA literature to be a relatively implicit and hence less burdensome technique than others for inducing learner attention to both form and meaning, it still does not appear to be supportive of meaning comprehension. This can be a pedagogical drawback in content-based curricula. Indeed, when L2 students read texts, they often need to learn language (or form) as much as content (or meaning), and it may not be judicious to teach them language at the cost of overlooking content through reading interventions. Particularly in content-based programs, as well as in a variety of other meaning-oriented educational settings, such as immersion education and school education of linguistic minority students, teachers and educators will need to use their professional judgment when making decisions regarding the use of textual input enhancement in their materials and lessons.

The author note

Department of Second Language Studies.

This paper reports on research conducted for my M.A. thesis submitted to the Department of Foreign Language Education at the graduate school of Seoul National University. I wish to thank Jin-Wan Kim, Oryang Kwon, Hyun-Kwon Yang for their insightful comments and suggestions; Timothy Jordan and Michael I. Choi for their generous assistance in the editing process; and three anonymous reviewers and Rob Schoonen for their helpful comments. Finally, my sincere appreciation goes to John Norris and Lourdes Ortega for their invaluable input and encouragement throughout the writing of this paper.

Correspondence concerning this article should be addressed to Sang-Ki Lee, Department of Second Language Studies, University of Hawai‘i at Mānoa, 1890 East-West Road, Honolulu, HI 96822. Internet: sangki@hawaii.edu

Notes

1I use the term “quasi-dual-tasks” in connection to typographical input enhancement treatments because the process of reading for meaning is challenged with enhanced target forms. This is by comparison to “dual-tasks,” a term often used in cognitive psychology, where two different tasks are being performed simultaneously.

2I would like to thank one of the anonymous reviewers, who suggested that intake happens on one exposure, while acquisition may happen over time and over multiple exposures (as a result of repeated intake of the same forms in a meaningful context). I agree that this operational distinction between intake and acquisition is useful for the present research domain. Accordingly, in Table 1 I classified Doughty (1991), Shook (1994), Alanen (1995), Jourdenais (1998), White (1998), Izumi (2002), Wong (2003) as acquisition studies. However, the distinction, albeit useful, should be treated with caution. Intake is the subset of input available in working memory, whereas acquisition involves not only immediate intake, but further processing and elaboration over extended periods of time. Yet, most designs featured in Table 1 are relatively short-term in nature. Thus, it is open to future theoretical debate whether the treatments classified as targeting “acquisition” in some of these studies (as well as the present one) truly engage learners’ internal systems.

3Cho (2002) pointed out that Doughty’s (1991) study should be interpreted with care, for her treatment was much more elaborate than just a simple provision of textual enhancement (p. 11). Doughty (1991) should be interpreted as some indirect evidence for the positive effects of textual enhancement on the acquisition of English relativization.

4Shook (1999) later analyzed his 1994 data for comprehension and found no significant difference for comprehension between groups.

5Regarding the use of think aloud protocols, Jourdenais et al.’s (1995) study employed offline protocols (during a post-test task), while the other studies employed concurrent protocols.

6The assumed superiority of free recall protocols over multiple-choice tests was challenged by one reviewer, who pointed out that recall protocols are an imperfect measure of L2 reading comprehension because they are subject to memory limitations. However, the same can be said for multiple-choice tests. In both types of measures, memory span may play a role.

7Originally 276 students participated in the study, but those who had not completed all the tasks were excluded from statistical consideration.

8The contents of the stories used in these first two exposure sessions were unrelated to the contents of the test session materials, to exclude the possibility that these two prior exposure sessions could affect the familiarity condition of the test session.

9Readers interested in Flesch-Kincaid readability tests are referred to the following webpage: .

10The National Assessment of Education Achievement is administered nationwide by an authorized institution, the Korea Institute of Curriculum and Evaluation. For detailed information, readers can visit the website at .

11Kim (2002, p. 278) claimed that Korean EFL learners’ confusion between the two participial adjectives can be accounted for by the fact that the present and past participle adjectives of Korean do not show a clear form-function relationship as in the following examples:

(1) How was the movie?

English: It was frightening.

Korean: (younghwa-ka) moosuhwoo-uhttsuh.

(2) How did you feel after the movie?

English: I was frightened.

Korean: (na-nun) moosuhwoo-uhttsuh.

12The reading passages were not taken from the students, except for the duration of the correction and recall tests. Thus, they could continue to refer to them at other points during the two weeks of instruction. This constitutes potential additional exposure to and practice with the target forms, as one anonymous reviewer correctly pointed out.

13Even though there were no statistically significant differences on the pre-test, the use of an ANCOVA would be preferable to an ANOVA because it would enable more exacting comparisons by adjusting for any initial differences on the dependent variables, however small. Unfortunately, this alternative method of analysis cannot be pursued because the raw data are no longer retrievable.

14The formula used to calculate d from F is provided by Norris and Ortega (2000, p. 444) and Rosenthal (1994, p. 238). I follow here Cohen’s (1988) useful rule of thumb for interpreting d-values: a value of d around 0.20 (one-fifth of a standard deviation) can be interpreted as small, around 0.50 (one-half of a standard deviation) as medium, and more than 0.80 (eight-tenths of a standard deviation) as large.

15The pre-to-post effect sizes were calculated, following Norris and Ortega (2000, pp. 442-443), by subtracting a given group's mean on the pre-test from the same group's mean on the post-test and dividing the resulting pre-to-post mean difference by the pooled standard deviation (which was obtained using the formula in Norris & Ortega, 2000, p. 443, with the standard deviations obtained on the pre-test and the post-test for that same group). Readers will find more discussions about calculation and use of pre-to-post effect sizes in Cohen (1988), Light and Pillemer (1984), Lipsey and Wilson (2001), and Norris and Ortega (2006).

16To test the statistical significance of differences between the correlations of each pair among the four treatment groups, z-values were calculated based on Fisher’s z-tests. The exact z-test values ranged from 0.38 to 1.68, none of which surpass the critical value at an alpha level of p < .05 (zcrit = 1.96). That is, although the correlations are clearly different in terms of magnitude (cf. r values in Table 8) and probability (cf. p values in the same table), these differences are not statistically significant at p < .05. Therefore, care should be employed in interpreting the observed differences in correlation magnitude, as these are not statistically trustworthy.

17I am indebted to Lourdes Ortega for drawing my attention to this issue (see also Ortega, 2003, p. 39).

References

Alanen, R. (1995). Input enhancement and rule presentation in second language acquisition. In R. Schmidt (Ed.), Attention and awareness in foreign language learning (pp. 259-302). Honolulu: University of Hawai‘i, Second Language Teaching & Curriculum Center.

Anderson, R. C., & Pearson, P. D. (1984). A schema-theoretic view of reading comprehension. In P. D. Pearson (Ed.), Handbook of reading research (pp. 255-291). New York: Longman.

Bernhardt, E. B. (1983). Testing foreign language reading comprehension: The immediate recall protocol. Die Unterrichtspraxis, 16, 27-33.

Bransdorfer, R. (1991). Communicative value and linguistic knowledge in second language acquisition oral input processing. Unpublished doctoral dissertation, University of Illinois at Urbana-Champaign.

Carrell, P. (1985). Facilitating ESL reading by teaching text structure. TESOL Quarterly, 19, 727-752.

Carrell, P. (1987). Content and formal schemata in ESL reading. TESOL Quarterly, 21, 461-481.

Cho, H. (2002). Effects of output and input enhancement on Korean EFL learners’ noticing and learning of a grammatical form. Unpublished MA thesis, Seoul National University.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.

Day, E., & Shapson, S. (1991). Integrating formal and functional approaches in language teaching in French immersion: An enhancement study. Language Learning, 41, 25-58.

Doughty, C. (1991). Second language instruction does make a difference: Evidence from an empirical study of SL relativization. Studies in Second Language Acquisition, 13, 431-469.

Doughty, C. (2001). Cognitive underpinnings of focus on form. In P. Robinson (Ed.), Cognition and second language instruction (pp. 206-257). Cambridge: Cambridge University Press.

Doughty, C., & Williams, J. (1998). Pedagogical choices in focus on form. In C. Doughty & J. Williams (Eds.), Focus on form in classroom second language acquisition (pp. 197-261). New York: Cambridge University Press.

Greenslade, T., Bouden, L., & Sanz, C. (1999). Attending to form and content in processing L2 reading texts. Spanish Applied Linguistics, 3, 65-90.

Hammerly, H. (1987). The immersion approach: Litmus test of second language acquisition through classroom communication. Modern Language Journal, 71, 395-401.

Harley, B. (1993). Instructional strategies and SLA in early French immersion. Studies in Second Language Acquisition, 15, 245-260.

Harley, B., & Swain, M. (1985). The interlanguage of immersion students and its implications for second language teaching. In A. Davies, C. Criper, & A. P. R. Howartt (Eds.), Interlanguage (pp. 291-311). Edinburgh: Edinburgh University Press.

Izumi, S. (2002). Output, input enhancement, and the noticing hypothesis: An experimental study of ESL relativization. Studies in Second Language Acquisition, 24, 541-577.

Johnson, P. (1981). Effects on reading comprehension of language complexity and cultural background knowledge of a text. TESOL Quarterly, 15, 169-181.

Johnson, P. (1982). Effects on reading comprehension of building background knowledge. TESOL Quarterly, 16, 503-516.

Jourdenais, R. (1998). The effects of textual enhancement on the acquisition of the Spanish preterit and imperfect. Unpublished doctoral dissertation, Georgetown University.

Jourdenais, R., Ota, M., Stauffer, S., Boyson, B., & Doughty, C. (1995). Does textual enhancement promote noticing? A think-aloud protocol analysis. In R. Schmidt (Ed.), Attention and awareness in foreign language learning (pp. 183-216). Honolulu: University of Hawai‘i, Second Language Teaching & Curriculum Center.

Kim, B. (2002). The role of focus on form in the EFL classroom setting: Looking into recast and CR tasks. English Teaching, 57, 267-296.

Lee, J. F. (1986). On the use of the recall task to measure L2 reading comprehension. Studies in Second Language Acquisition, 8, 201-211.

Leow, R. (1997). The effects of input enhancement and text length on adult L2 readers' comprehension and intake in second language acquisition. Applied Language Learning, 8, 151-182.

Leow, R. (1999). Attention, awareness, and focus on form research: A critical overview. In J. F. Lee & A. Valdman (Eds.), Form and meaning: Multiple perspectives (pp. 69-96). Boston, MA: Heinle & Heinle.

Leow, R. (2001). Do learners notice enhanced forms while interacting with the L2? An online and offline study of the role of written input enhancement in L2 reading. Hispania, 84, 496-509.

Leow, R., Egi, T., Nuevo, A., & Tsai, Y. (2003). The roles of textual enhancement and type of linguistic item in adult L2 learners’ comprehension and intake. Applied Language Learning, 13, 1-16.

Light, R. J., & Pillemer, D. B. (1984). Summing up: The science of reviewing research. Cambridge, MA: Harvard University Press.

Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. Thousand Oaks, CA: Sage.

Loschky, L. (1994). Comprehensible input and second language acquisition: What is the relationship? Studies in Second Language Acquisition, 16, 303-323.

Long, M. H. (1991). Focus on form: A design feature in language teaching methodology. In K. De Bot, R. B. Ginsberg, & C. Kramsch (Eds.), Foreign language research in cross-cultural perspective (pp. 39-52). Philadelphia: John Benjamins Publishing Company.

Long, M. H., & Robinson, P. (1998). Focus on form: Theory, research, and practice. In C. Doughty & J. Williams (Eds.), Focus on form in classroom second language acquisition (pp. 15-41). New York: Cambridge University Press.

Norris, J. M., & Ortega, L. (2000). Effectiveness of l2 instruction: A research synthesis and quantitative meta-analysis. Language Learning, 50, 417-528.

Norris, J. M., & Ortega, L. (2006). The value and practice of research synthesis for language learning and teaching. In J. M. Norris & L. Ortega (Eds.), Synthesizing research on language learning and teaching (pp. 3-50). Philadelphia, PA: John Benjamins.

Ortega, L. (2003). Interacción y negociación del mensaje en el salón de idiomas: Panorámica crítica de la investigación actual y propuestas para el futuro. [Interaction and negotiation of meaning in the classroom: A critical overview and directions for future research]. MEXTESOL Journal, 27, 37-54.

Overstreet, M. (1998). Text enhancement and content familiarity: The focus of learner attention. Spanish Applied Linguistics, 2, 229-258.

Overstreet, M. (2002). The effects of textual enhancement on second language learner reading comprehension and form recognition. Unpublished doctoral dissertation, University of Illinois at Urbana-Champaign.

Pritchard, R. H. (1990). The effects of cultural schemata on reading processing strategies. Reading Research Quarterly, 25, 273-295.

Reynolds, R. E., Taylor, M. A., Steffensen, M. S., Shirley, L. L., & Anderson, R. C. (1982). Cultural schemata and reading comprehension. Reading Research Quarterly, 17, 353-366.

Robinson, P. (2003). Attention and memory during SLA. In C. Doughty & M. H. Long (Eds.), Handbook of second language acquisition (pp. 631-678). Malden, MA: Blackwell.

Rosa, E., & O'Neill, M. (1999). Explicitness, intake, and the issue of awareness. Studies in Second Language Acquisition, 21, 511-556.

Rosenthal, R. (1994). Parametric measures of effect size. In H. Cooper & L. Hedges (Eds.), Handbook of research synthesis (pp. 231-244). New York: Russell Sage Foundation.

Rumelhart, D. E. (1977). Toward an interactive model of reading. In S. Dornic (Ed.), Attention and performance VI (pp. 575-603). Hillsdale, NJ: Lawrence Erlbaum.

Schmidt, R. (2001). Attention. In P. Robinson (Ed.), Cognition and second language instruction (pp. 3-32). New York: Cambridge University Press.

Sharwood Smith, M. (1981). Consciousness-raising and the second language learner. Applied Linguistics, 2, 159-168.

Sharwood Smith, M. (1993). Input enhancement in instructed SLA: Theoretical bases. Studies in Second Language Acquisition, 15, 165-179.

Shook, D. J. (1994). FL/L2 reading, grammatical information, and the input-to-intake phenomenon. Applied Language Learning, 5, 57-93.

Shook, D. J. (1999). What foreign language reading recalls reveal about the input-to-intake phenomenon. Applied Language Learning, 10, 39-76.

Simard, D. (2002). La mise en évidence textuelle: d’où venons nous et où allons nous? The Canadian Modern Language Review, 59, 236-263.

Skehan, P. (2003). Focus on form, tasks, and technology. Computer Assisted Language Learning, 16, 391-411.

Skehan, P., & Foster, P. (2001). Cognition and tasks. In P. Robinson (Ed.), Cognition and second language instruction (pp. 183-205). New York: Cambridge University Press.

Swain, M. (1985). Communicative competence: Some roles for comprehensible input and comprehensible output in its development. In S. Gass & C. Madden (Eds.), Input in second language acquisition (pp. 235-253). Rowley, MA: Newbury House.

Swain, M., & Lapkin, S. (1989). Canadian immersion and adult second language teaching: What is the connection? Modern Language Journal, 73, 150-159.

Thompson, B. (2002). What future quantitative social science research could look like: Confidence intervals for effect sizes. Educational Researcher, 31, 25-32

Tomlin, R., & Villa, V. (1994). Attention in cognitive science and second language acquisition. Studies in Second Language Acquisition, 16, 183-203.

VanPatten, B. (1990). Attending to form and content in the input: An experiment in consciousness. Studies in Second Language Acquisition, 12, 287-301.

VanPatten, B. (1996). Input processing and grammar instruction: Theory and research. Norwood, NJ: Ablex publishing Corp.

VanPatten, B. (2002). Processing instruction: An update. Language Learning, 52, 755-803.

Weber, R. (1991). Linguistic diversity and reading in American society. In R. Barr, M. L. Kamil, P. Mosenthal, & P. D. Pearson (Eds.), Handbook of reading research (pp. 97-119). New York: Longman.

White, J. (1998). Getting the learners’ attention: A typographical input enhancement study. In C. Doughty & J. Williams (Eds.), Focus on form in classroom second language acquisition (pp. 85-113). New York: Cambridge University Press.

Wolff, D. (1987). Some assumptions about second language text comprehension. Studies in Second Language Acquisition, 9, 307-326.

Wong, W. (2001). Modality and attention to meaning and form in the input. Studies in Second Language Acquisition, 23, 345-368.

Wong, W. (2003). Textual enhancement and simplified input: Effects on L2 comprehension and acquisition of non-meaningful grammatical form. Applied Language Learning, 13, 17-45.

Yeo, K. (2002). The effects of dictogloss: A technique of ‘focus on form.’ English Teaching, 57, 149-167.

Appendix A

Reading Materials for the Test Session

Enhanced and Familiar Text

Birthday Celebration in Korea

On the 100th day after a child's birth, there is a small feast, or 'baek-il feast.' It is to celebrate the child having survived this difficult period. At this time the Samshin Halmoni is honored with offerings of rice and soup. Also, to prevent disaster and to bring the child luck and happiness, red bean cakes are usually placed at every corner within the house.

Such customs are part of the 'dol feast,' or first birthday. Because of the high infant death rate in the past, this celebration is regarded as even more important. The highlight of this celebration is when the child symbolically foretells its future. For this ritual the child is dressed in traditional Korean clothes. Then the child is seated before a table of various foods and objects such as thread, books, notebooks, pencils, and money which have been offered by friends and relatives. The child is urged to pick up an object from the table, and it is believed that the object foretells the child's future. For example, if the child picks up the thread, it is believed that s/he will live a long life.

When Koreans are invited to the 'dol feast,' they usually present a gold ring to the parents. Friends and relatives collect money to buy the gold ring. However, it is not for the child's finger but for the benefit of the child in times of need in the future.

The hwan-gap, or 60th birthday, has been considered especially important. That's because few people lived to be 60 before the advent of modern medicine. With the parents seated at the main table, sons and daughters bow and offer wine to their parents. The family members and relatives indulge in various activities to make the parents feel young. In the past, years after the 60th birthday were regarded as extra years.

Enhanced and Less Familiar Text

The Spirit of the Dead

The Egyptians believed that every person was composed of three essential elements: body, ba, and ka.

The body is the physical body and is unique to each individual. Each person also has a ba. Though the ba is also unique to each individual, it is not a physical entity. Ba is sometimes translated as "manifestation," and can be considered as the sum total of all the non-physical things that makes a person different from others. In this sense, ba is very similar to what is called "personality" or "character." In the afterlife, the ba is represented as a bird, often with a human head. Each person also has what is called a ka, or ‘life-force,’ and it is the ka which makes the difference between being alive and being dead.

When a person dies, the ba and ka are separated from the body, though they do not die. The Egyptians enabled this separation through 'The Opening of the Mouth' ritual, in which the ba and ka are released to go to the next world.

In the next world, to get an eternal life, the ka needs to be summoned back to the body. But since the body is tied in its wrappings, it must rely on its ba to seek out its ka. In seeking a union with the ka, the ba must overcome many potential dangers in the underworld. But if it does succeed, it will reunite with the ka and form what is called akh. The Egyptians believed that there are only three kinds of beings that inhabit the hereafter: the dead, the gods, and akhs. Akhs are those who have successfully made the transition to new life in the next world, where they live with the gods. The dead are those who have failed to make the transition. It is believed that they have "died again," with no hope of renewed life.

Appendix B

Samples of Idea Unit Analysis

Birthday Celebration in Korea

1. On the 100th day after a child's birth, 

2. there is a small feast, or 'baek-il feast.'

3. It is to celebrate the child having survived this difficult period.

4. At this time the Samshin Halmoni is honored

5. with offerings of rice and soup.

The Spirit of the Dead

1. The Egyptians believed that every person was composed of three essential elements:

2. (The elements are) body, ba, and ka.

3. The body is the physical body

4. and is unique to each individual.

5. Each person also has a ba.

TABLE 1

Summary of the Studies on the Effects of Textual Enhancement

|Study |Target form(s) |Duration of the |Participants |Assessing measuresa |Effect on |Effect on |

| | |treatment | | |meaning processing |form processing |

|Doughty (1991) |English relative |10 working days |20 adult intermediate level |1. comprehension questions (c) |not directly tested |positive effects on |

| |clauses | |ESL learners with diverse L1|2. free recall task (c) | |acquisition |

| | | |backgrounds |3. grammaticality judgment task | | |

| | | | |4. sentence combination task | | |

| | | | |5. guided sentence completion task | | |

| | | | |6. oral test | | |

|Shook (1994) |Spanish present |two-day period; |125 adult learners with L1 |1. recognition task |not tested |positive effects on |

| |perfect/ relative|less than one hour |English/1st & 2nd year |2. fill-in-the-blank production task | |acquisition |

| |pronouns | |Spanish | | | |

a(c) = measures for meaning comprehension.

TABLE 1 (continued)

Summary of the Studies on the Effects of Textual Enhancement

|Study |Target form(s) |Duration of the |Participants |Assessing measuresa |Effect on |Effect on |

| | |treatment | | |meaning processing |form processing |

|Alanen (1995) |Finnish locative |two-day period; |36 adult learners with L1 |1. sentence completion test |not tested |some positive effects |

| |suffixes / |less than one hour |English |2. grammaticality judgment task | |on acquisition |

| |consonant | | |3. rule statements | | |

| |gradation | | |4. think-aloud protocols | | |

|Jourdenais et |Spanish preterit/|less than one hour |10 adult learners with L1 |1. think-aloud protocols |not tested |positive effects on |

|al. (1995) |imperfect | |English/2nd semester Spanish|2. picture-based writing task | |intake |

|Leow (1997) |Spanish |less than one hour |84 adult learners with L1 |1. short-answer comprehension task (c) |no effects |no effects on intake |

| |impersonal | |English/2nd semester Spanish|2. multiple-choice form recognition task | | |

| |imperative | | | | | |

a(c) = measures for meaning comprehension.

TABLE 1 (continued)

Summary of the Studies on the Effects of Textual Enhancement

|Study |Target form(s) |Duration of the |Participants |Assessing measuresa |Effect on |Effect on |

| | |treatment | | |meaning processing |form processing |

|Jourdenais |Spanish preterit/|three sessions over|116 adult learners with L1 |1. comprehension questions (c) |no effects |no effects on |

|(1998) |imperfect |one-week period; |English from different |2. multiple-choice task | |acquisition |

| | |30-45 minutes for |proficiency groups |3. narrative essay task | | |

| | |each session | | | | |

|Overstreet |Spanish preterit/|less than one hour |50 adult learners with L1 |1. T/F comprehension quiz (c) |negative effects |no effects on intake |

|(1998) |imperfect | |English/3rd semester Spanish|2. circle-the-verb task | | |

| | | | |3. written narration task | | |

|White (1998) |English |two-week period; |86 6th grade ESL learners |1. passage correction task |not tested |partial effects on |

| |possessive |10 hours |with L1 French |2. multiple-choice test | |acquisition |

| |determiners | | |3. oral picture description task | | |

a(c) = measures for meaning comprehension.

TABLE 1 (continued)

Summary of the Studies on the Effects of Textual Enhancement

|Study |Target form(s) |Duration of the |Participants |Assessing measuresa |Effect on |Effect on |

| | |treatment | | |meaning processing |form processing |

|Leow (2001) |Spanish |less than one hour|38 adult learners |1. short-answer and multiple-choice task (c) |no effects |no effects on intake; no |

| |imperatives | |with L1 English/1st |2. multiple-choice recognition task | |effects on students’ |

| | | |year Spanish |3. fill-in-the-blank production task | |reported noticing |

| | | | |4. think-aloud protocols | | |

|Izumi (2002) |English relative |six sessions over |61 adult ESL learners|1. recall task (c) |not directly tested |positive effects on |

| |clauses |two-week period; |with diverse L1 |2. sentence combination test | |noticing (measured by note|

| | |30-60 minutes for |backgrounds |3. picture-cued sentence completion test | |taking and text |

| | |each session | |4. interpretation test | |reconstruction task); no |

| | | | |5. grammaticality judgment test | |effects on acquisition |

a(c) = measures for meaning comprehension.

TABLE 1 (continued)

Summary of the Studies on the Effects of Textual Enhancement

|Study |Target form(s) |Duration of the |Participants |Assessing measuresa |Effect on |Effect on |

| | |treatment | | |meaning processing |form processing |

|Overstreet |Spanish present |less than one hour |109 adult learners |1. free recall task (c) |no effects; better recall |no effects on intake; |

|(2002) |progressive/ | |with L1 English/3rd |2. form recognition task |of enhanced information in|better recognition of |

| |imperfect | |semester Spanish | |the case of whole word |items with higher |

| |subjunctive | | | |enhancement (cf. no effect|communicative value |

| | | | | |with morphology-only | |

| | | | | |enhancement) | |

|Leow |Spanish present |less than one hour |72 first year |1. multiple-choice comprehension task (c) |no effects |no effects on intake; no |

|et al. (2003) |perfect/ present | |college-level |2. multiple-choice form recognition task | |effects on noticing |

| |subjunctive | |learners |3. think-aloud protocols | | |

a(c) = measures for meaning comprehension.

TABLE 1 (continued)

Summary of the Studies on the Effects of Textual Enhancement

|Study |Target form(s) |Duration of the |Participants |Assessing measuresa |Effect on |Effect on |

| | |treatment | | |meaning processing |form processing |

|Wong (2003) |French past |three-day period; |81 adult learners |1. free recall task (c) |no effects; better recall |no effects on acquisition |

| |participle |less than one hour |with L1 English/2nd |2. error correction task |of enhanced information | |

| |agreement in | |semester French | | | |

| |relative clauses | | | | | |

a(c) = measures for meaning comprehension.

TABLE 2

Types of Passive-related Form Errors

|Type A |The Koreans were exciting about the unexpected results of the soccer game. |

|Type B |He was offer the job but did not accept it. |

|Type C |Snow boarding considered as the most exciting winter sport among teens. |

TABLE 3

Overview of the Study

|Procedure |-E/+F |-E/-F |+E/+F |+E/-F |

| |(n = 62) |(n = 65) |(n = 70) |(n = 62) |

|1st week |1st session: | L2 reading proficiency test (30 min) |

| |Pre-test |Form correction task (10 min) |

| |2nd session: | Reading about Jokpo in Korea (20 min) |

| |Exposure #1 | |

| | |Baseline |Baseline |Enhancement |Enhancement |

|2nd week |3rd session: | Reading about Mummies of Egypt (20 min) |

| |Exposure #2 | |

| | |Baseline |Baseline |Enhancement |Enhancement |

| |4th |Exposure #3 |Birthday celebration|The spirit of the |Birthday celebration |The spirit of |

| |session | |in Korea |dead |in Korea |the dead |

| | | |Baseline |Baseline |Enhancement |Enhancement |

| | | | (20 min) |

| | |Post-test | Free recall task (15 min) |

| | | |Form correction task (10 min) |

TABLE 4

Descriptive Statistics of the Pre-test Scores

|Group |Reading Proficiency Testa |Form Correction Taskb |

| |Mean |SD |Mean |SD |

|-E/+F (n = 62) |54.44 |15.71 |2.42 |1.97 |

|-E/-F (n = 65) |51.46 |16.39 |1.88 |1.41 |

|+E/+F (n = 70) |53.14 |17.24 |2.31 |2.01 |

|+E/-F (n = 62) |52.98 |15.95 |2.19 |2.06 |

|Total (N = 259) |52.99 |16.30 |2.20 |1.88 |

aThe maximum possible scores were 100.

bThe maximum possible scores were 10.

TABLE 5

Descriptive Statistics of the Form Correction Task Scores

| |n |Mean |SD |

|-E/+F |62 |2.82 |2.05 |

|-E/-F |65 |2.15 |1.82 |

|+E/+F |70 |4.77 |2.97 |

|+E/-F |62 |5.06 |2.59 |

|Total |259 |3.72 |2.70 |

TABLE 6

Paired T-tests for the Pre-test and Post-test Scores of the Form Correction Task

| |Group |Mean |Pooled SD |t |df |p |d |

| | |Difference | | | | |(Effect Size) |

|Pre-test |-E/+F |.40 |2.01 |1.91 |61 |.06 |0.20 |

|–Post-test | | | | | | | |

| |-E/-F |.28 |1.62 |1.41 |64 |.16 |0.17 |

| |+E/+F |2.46 |2.49 |6.87 |69 |.00* |0.99 |

| |+E/-F |2.87 |2.33 |8.95 |61 |.00* |1.23 |

*p < .05.

TABLE 7

Descriptive Statistics of the Free Recall Task Scores

| |n |Mean |SD |

|-E/+F |62 |17.29 |7.60 |

|-E/-F |65 |7.86 |5.11 |

|+E/+F |70 |11.46 |6.29 |

|+E/-F |62 |4.94 |4.35 |

|Total |259 |10.39 |7.46 |

TABLE 8

Correlations Between Correction and Comprehension Scores

|Group |Correlation Coefficient |p |n |

|-E/+F |.42 |.00* |62 |

|-E/-F |.35 |.00* |65 |

|+E/+F |.12 |.31 |70 |

|+E/-F |.23 |.07 |62 |

|Total |.08 |.19 |259 |

*p < .05.

[pic]

Figure 1. Pre-test and post-test scores of the form correction task

[pic]

Figure 2. 95% confidence intervals for the means of the free recall task

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download