Effects of Implicit and Explicit Focus on Form on
EFL Learners’ Anaphor Resolution Performance
in Reading
Ting Fang
National Taiwan Normal University [email protected]
Abstract
Learner attention and the implicitness/explicitness of the relation between anaphor and its antecedent are known to be crucial in anaphor resolution performance. This reflects the rationale of focus on form in language teaching, applied implicitly or explicitly to direct learners’ attention to specific grammatical form. The present study thus examined effects of implicit and explicit focus on form on L2 learners’ anaphor resolution performance in reading and the relationship among focus on form, anaphors, and reading comprehension, thereby helping to explicate the role of focus on form in L2 learners’ coherence construction in reading literacy and how their attention should be best directed to the targeted form. One hundred and two Taiwan EFL learners underwent four different treatments, including textual enhancement and explicit rule conditions. Participants’ responses were compared with respect to (a) their ability to identify correct English anaphor relations and (b) their reading comprehension degree. The results revealed that implicit, rather than explicit focus on form aided EFL learners’ anaphor resolution performance. Moreover, learners’ attentional resources for meaning and form processing were not competing, but may have been mutually transferable. These findings are discussed in light of the characteristics of anaphors, the effect of focus on form, and learners’ attentional resources with implications.
142
INTRODUCTION
It was reported that anaphoric reference is critical in L1 reading literacy (Britton & Gulgoz, 1991; Oakhill & Yuill, 1986) and is important for non-native speakers to comprehend English texts (Cohen & Fine, 1978). Although literature investigating anaphor teaching for EFL learners is still lacking so far, some factors influencing readers’ anaphor resolution performance were identified, one of which is the implicit/explicit degree of the relation between the anaphor and its antecedent (Gallini & Spires, 1992). Researchers suggest that the more transparent the relation is, the better readers are able to infer the references in texts. The saliency of the relation and readers’ attention to it are thus important in anaphor resolution performance. This feature reflects the rationale of employing focus on form in language teaching, which is often applied implicitly or explicitly to direct learners’ attention to specific grammatical forms. Previous studies on the effect of implicit and explicit focus on form showed mixed results (Alanen, 1995; Doughty, 1991; N. Ellis, 1993; Williams & Evans, 1998), which were possibly influenced by many factors, e.g., different forms. In addition, the targeted form in most focus on form literature is restricted to the teaching of grammatical rules only. The present study, therefore, aims to investigate the effect of implicit and explicit focus on form on L2 learners’ anaphor resolution performance, helping depict the generalizability of focus on form and the uniqueness of the targeted form. The relationship among focus on form (i.e., learner attention), anaphors, and reading comprehension is also examined in this study.
143
LITERATURE REVIEW
Anaphor and Reading Comprehension
Anaphors function as references indicating a concept that was already mentioned earlier. They are used “for reference to a relation between two linguistic elements, in which the interpretation of one (called an anaphor) is in some way determined by the interpretation of the other (called an antecedent)” (Y. Huang, 2006, p. 231). For example, the anaphors like the car, the vehicle, and it could refer back to their antecedent a car (Dell, McKoon, & Ratcliff, 1983). The relation between anaphor and its antecedent could come across or within sentences. They occur “when an element of underlying clause structure refers to an entity which has already been established, directly or indirectly, in the preceding discourse (discourse anaphora) or is being established in the same clause (sentence anaphora)” (Dik, 1997, p. 215). In L1 reading, anaphoric reference is crucial in text coherence construction and in successful general reading comprehension (Ehrlich, Remond, & Tardieu, 1999; Garrod & Sanford, 1990; Givón, 1995; Peterson, 1993). It was also reported that non-native speakers had common problems in integrating textual information with referential devices (Cohen & Fine, 1978). This inability to construct coherence in texts often led to failure of reading comprehension (Demel, 1990), and less skilled readers often had difficulties in anaphor comprehension for textual coherence (Oakhill & Yuill, 1986).
144
Anaphor and Focus on Form
One of the factors affecting readers’ anaphor resolution process is the transparent relation between anaphor and its antecedent perceived by readers. According to Gallini and Spires (1992), low inference ability is required when anaphor-antecedent relation is more explicit, and high inference ability is needed when this relation is implicit. Their findings showed that readers performed better when low inference ability was utilized, especially for less skilled readers. Since “anaphora increases coherence by signaling relationship between ideas” (Shapiro & Milkes, 2004, pp. 166-167), it is important to understand how this relation is signaled and to what extent the signal is salient enough to be perceived by readers. In this regard, readers’ attention is directed implicitly or explicitly to the linguistic form, i.e., the anaphor-antecedent relation, in the reading comprehension process, which reflects the fundamental core of focus on form. Focus on form is a way to allocate our focal attentional resources, and learners’ attentions to forms and meaning. It directs our attention to linguistic code features when perceiving difficulties in comprehension or production (Long & Robinson, 1998, p. 23).
According to the three criterion proposed by researchers (DeKeyser, 2005; Doughty & Williams, 1998; Hulstijn & De Graaff, 1994), anaphoric references could serve as the prime candidate for focus on form. First, the targeted form should be one with high reliability and wide scope (Hulstijn & De Graaff, 1994). High reliability refers to the consistency of the targeted form in all cases, and wide scope means the availability of the form in many cases. The relation between an anaphor and its antecedent has high reliability
145
since it is held true in all cases, without being affected by the distance or direction. It also has wide scope because the concept of reference exists not only in English, but also in many other languages, e.g., Chinese. Second, according to Doughty and Williams (1998, p. 212), the forms “that appear to be problematic for a particular group of learners” were the targets for many teachers. It was found that many learners have difficulties handling anaphoric references in reading and writing (Garrod & Sanford, 1990; Shapiro & Milkes, 2004). Third, the ideal targeted form is the one with formal and functional simplicity for learners to acquire (DeKeyser, 2005). The function of anaphoric references is simply to guide readers to the concept established earlier. The form of anaphors is more flexible since it can be presented as pronouns or repeated nouns. However, as maintained by R. Ellis (1990), a form could still be learnable when learners are developmentally ready for it, even though it is relatively complex. Implicit and Explicit Focus on Form
Based on previous studies on focus on form, Leow (2001) classified three conditions for drawing learners’ attention to the input. The first condition provides learners with implicit or incidental means to notice target linguistic features, such as typographical enhancement (TE). It was found that different textual enhancement formats might have effects on ESL learners’ intake of English linguistic form (Simard, 2009). The second one provides more explicit cues, including explicit instructions or feedback. The third one lay somewhere in between, such as processing instruction. The distinction between implicit and explicit was suggested by Doughty and
146
Williams (1998, pp. 230-232) that implicit meant “implied or understood though not directly expressed” and “contained in the nature of something though not readily apparent.” Explicit meant “fully and clearly expressed, defined or formulated” and “readily observable.” They further explained different purposes of implicit and explicit instructions by claiming that implicit focus on form aimed to “attract learner attention and to avoid metalinguistic discussion, always minimizing any interruption to the communication of meaning.” Explicit teaching aimed to “direct learner attention and to
exploit pedagogical grammar in this regard.” It should be noted and
kept in mind that although both Leow (2001) and Doughty and Williams (1998) made distinctions between implicitness and explicitness, Leow was referring to exposure while Doughty and Williams were referring to instruction. Exposure (as in reading) is not the same as instruction (as performed by the teacher). In sum, compared with explicit rule presentations, TE is a more implicit technique promoting learners’ attention to form, even though TE is more explicit than the input flooding (White, 1998).
Effects of implicit and explicit focus on form in language teaching have been studied by researchers (Alanen, 1992, 1995; Doughty, 1988, 1991; Lyddon, 2011; Williams & Evans, 1998). Lyddon (2011) examined the effect of TE and implicit/explicit corrective feedback on university French learners’ grammar learning. The participants who read either textually enhanced or unenhanced target forms received one of the four corrective feedback types, including meaning-focused, implicit form-focused, non-metalinguistic explicit form-focused, and metalinguistic explicit form-focused. The
147
author claimed that no significant differences were found among these conditions and suggested teachers focus more on learners’ processing opportunities than the form itself. On the other hand, some researchers emphasized the use of explicit cues over implicit ones for interlanguage development (White, 1998). For example, Alanen (1992, 1995) investigated the effect of explicit explanation and implicit TE on ESL learners’ learning of targeted linguistic forms. Participants were instructed to read texts containing the targeted form and answer comprehension questions afterwards. The participants were grouped into four types, each reading different text versions, including the original text, the enhanced text, the original text with explicit rule explanations, and the enhanced text with explicit rule explanations. The results showed that in the sentence completion and consonant production task, two explicit explanations given (explicit focus on form) groups outperformed the other two (meaning-based) groups. Moreover, no significant difference was found between the two explicit focus on form groups and the meaning-based groups. In other words, the researcher found that explicit rule presentation was more beneficial to learners’ acquisition of form than TE alone. In N. Ellis’ (1993) study, it was also found that explicit rather than implicit instruction led to greater short-term learning on complex L2 rules. Similar findings were shown by DeKeyser (1995), who claimed that implicit learners did not have advantages over explicit learners on complex rules, and explicit learners outperformed implicit learners on simple rules in short-term learning. Other studies, however, indicated the advantage of both implicit and explicit focus on form. For example, Doughty (1988, 1991) analyzed 20 adult ESL learners’
148
acquisition of relative clauses. These participants were randomly assigned to three groups, each reading one text and answering follow-up comprehension questions. The control grofollow-up read the unenhanced text, the TE group read the text with relative clauses capitalized and enhanced, and the rule-oriented group read the unmodified text with additional written instruction. The results showed that the TE group performed better than the other two groups in the comprehension tests, and both the TE and rule-oriented groups outperformed the control group in relativization knowledge. Moreover, Williams and Evans (1998) investigated how implicit and explicit focus on form treatments affected 33 ESL learners’ learning participial adjectives and the passive. For the input flood group, participants were given a flood of positive input containing the target form without any explicit rule instructions. For the explicit group, participants were given the same input flood but with the extra explicit instructions and corrective feedback. The control group used the same material without particular flood of the target form. The results showed that both focus on form groups (i.e., the input flood group and the instruction group) outperformed the control group in the grammaticality judgment task, the sentence completion task, and the narrative task. Also, instructions on the passive, rather than on the participial adjectives, seemed to benefit learners more, i.e., focus on form instruction may be more effective in one form than another.
Although these studies provided mixed results in the form processing by implicit or explicit intervention, possibly due to their methodological idiosyncrasies (Han, Park, & Combs, 2008), many studies seem to agree that focus on form might have a trade-off effect
149
between form and meaning processing. For example, in Overstreet’s (1998) experiment, the form enhanced by TE affected readers’ global comprehension performance negatively. According to the explanation given by Overstreet, this interference in meaning processing occurred when readers’ attention was drawn to these salient forms. Lee (2007) also indicated that although focus on form facilitated learning of the linguistic form, this obtrusive intervention brought about unfavorable effects on the global comprehension of texts. It was suggested the balance between obtrusiveness and advantage in learning of focus on form be reached by teachers and researchers (Lee, 2007).
RESEARCH QUESTIONS
The present study examines effects of implicit and explicit focus on form on L2 learners’ anaphor resolution performance in reading and the relationship among focus on form, anaphors, and reading comprehension. This study is important in helping understand the role of focus on form in L2 learners’ coherence construction in reading literacy and how their attention are best directed to the targeted form. The research questions are as follows.
1. What are the effects of textual enhancement (implicit focus on form) on the L2 learners’ anaphor resolution performance?
2. What are the effects of explicit rules (explicit focus on form) on the L2 learners’ anaphor resolution performance?
150
3. What is the relationship among L2 learners’ anaphor comprehension, reading comprehension scores, and focus on form?
METHOD
Participants
Four intact classes in one senior high school located in northern Taiwan participated in this study. Participants comprised a total of 102 high school students, who began to learn English as a foreign language approximately five years ago in the junior high schools. As in a typical Taiwan EFL classroom in senior high schools, these participants attended 6 English classes per week, and each class consisted of 50 minutes. Due to the shortage of native speakers and preparation for the coming college entrance exam, reading skills and grammar instruction are more emphasized than other skills. These students had learned the target concept (anaphor reference in English) earlier from examples given by their teacher, but they did not seem to fully master the target concept according to pretest results (M = 7.59, out of 10 possible points, SD = 1.58).
Reading Materials for Experimental Sessions
The 102 participants in the study were provided with two reading tasks, serving as pretest and posttest, respectively. The reading text for the first task included two passages about relationships and jet lag. The reading text for the second task was about human brains and a computer company. Readings in both tasks
151
were informative in nature. Unlike task 1, there were four versions for task 2, including the TE version (anaphor and its antecedent in the text were typographically enhanced by bolding and increasing the font size), the Explicit rule version (the explicit rule and function of anaphors were provided in addition to the reading passage), the Combination of TE and Explicit rule version, and the baseline version (without the target form being enhanced). It should be noted that the same token of target forms and approximately equal word numbers were contained in task 1 (517 words) and 2 (512 words) readings. In addition, based on Flesch reading ease formula, the overall readability of the paragraphs is 57.5 in task 1 (Flesch-Kincaid grade level = 9.1) and 59.7 in task 2 (Flesch-Kincaid grade level = 8.9). Thus, readings in the pretest and posttest were of the same level and comprehensible for the participants in this current study. The 102 participants were assigned randomly to four groups. In task 1, all participants read the same text (relationships and jet lag) without the target form being enhanced. In task 2, participants read different versions of readings (human brains and a computer company) based on different groups. Group 1 read the typographically enhanced version only; Group 2 read the baseline version only (the control group); Group 3 read the baseline version, but they were given a piece of paper with explicit instruction on it; Group 4 read the typographically enhanced version and they were also given a piece of paper with explicit instruction on it. The target forms in the enhanced text were marked with boldfaced letters and larger fonts (see Appendix A and B). Both passages were followed by reading comprehension questions and reference
152
identification questions. All participants were required to answer these questions after they finished the reading.
Assessment Measures
L2 reading proficiency test. In order to ensure the participants were comparable, a reading proficiency test was conducted in the very beginning of the study. The researcher adopted samples from the General Scholastic Ability Test as test items. The General Scholastic Ability Test is a nationwide measure in Taiwan to examine high school students’ scholastic achievement. Participants needed to answer 16 reading comprehension questions from four reading passages, each followed by four comprehension questions. They were asked to identify main ideas and detailed information mentioned in the readings. To ensure the test has required reliability, two experienced, high school English teachers and one university English professor were invited to evaluate the test. The Cronbach’s alpha of the test, determined as .82, was generated from the data in the pilot study.
Reading comprehension questions. To measure the effect of TE and explicit instructions on participants’ reading comprehension, all participants were required to answer eight reading comprehension questions after they finished the passage reading in both task 1 and task 2. The reading comprehension questions were designed based on the reading materials in task 1 and 2. The eight reading comprehension questions for each reading material aimed to test participants’ overall understanding of the passages, including main idea and detailed information identification. All test items were
153
scored by giving one point for correct response, and zero for incorrect or no response.
Reference identification questions. Reference identification questions were developed to examine participants’ comprehension of the antecedent-anaphor reference relations. After reading in both task 1 and task 2, all participants answered ten reference identification questions, designed according to the reading materials participants read. These ten questions following each reading passage were to investigate whether readers were able to identify the antecedent-anaphor reference relationship in the texts. Since the purpose of this study is to investigate L2 readers’ anaphor resolution performance, it is hoped that these questions used can really measure whether participants comprehend the anaphoric relation. Therefore, the researcher adopted the testing format from literature (S. H. Huang, 2005) and asked the participants to identify this relation “directly” (e.g., what does “they” in line 16 refer to?) to ensure the test is measuring what is supposed to be tested, i.e., with high validity. This testing format is seen as direct measure because it involves questions requiring readers to name the targeted form directly. In contrast, other measures that “indirectly” tackled participants’ knowledge of anaphoric relation were less preferred in the present study. These measures are seen as indirect because they often require the researcher’s indirect inference for readers’ ability. For example, the reading text is presented as John and Bill loved the car, but Bill had
no money; he bought it immediately and the anaphoric relation
between John and he is measured by the indirect question such as who
154
higher face validity can be reached by the second more direct question because the measure “looks like” it is the anaphoric relation being measured. The first testing question, however, can be argued that it measures more students’ reading comprehension than their anaphoric resolution ability, even though these two may be strongly correlated. Participants were told to write down the antecedent of its targeted anaphor after finishing the reading. One point was given when readers could correctly identify the antecedent of the targeted anaphor in the paragraph, or when they could clearly indicate the key words of the antecedents. Like reading comprehension questions, reference identification questions served as the offline measure to detect participants’ noticing phenomenon.
Both the reading comprehension questions and the reference identification questions together play important roles to ensure the internal validity of the offline measure. With the comprehension questions, participants were asked to engage in a meaningful reading process so as to avoid the possibility that any group has lower scores on reference identification questions because they noticed the relationship but did not process the information deeply. Similarly, with the pretest reference identification questions, the possibility that any group has high scores because of high prior knowledge of anaphoric relationships can be avoided.
Procedure
The study was administrated in December 2011. Participants in the study had 6 English classes (50 minutes for each class) per week. The participants were assigned randomly to four groups according to
155
different treatments, i.e., TE or/and explicit instructions. Consequently, there were four groups in this study, including the groups using TE only (＋T－E), the group using explicit instructions only (－T＋E), the group using TE and explicit instructions (＋T＋E), and the group using neither (－T－E). Participants who did not finish the test were eliminated. At the onset of the study, an L2 reading proficiency test was conducted to measure participants’ reading abilities and to ensure these four groups were comparable, as shown in Table 1. The ANOVA result showed that no significant difference was found between these groups, F(3, 98) = .114, p = .95
Table 1
Descriptive Statistics of L2 Reading Proficiency Test
L2 Reading Proficiency Test Group n Mean SD －T－E 26 9.88 2.21 ＋T－E 22 9.55 2.81 －T＋E 24 9.50 2.99 ＋T＋E 30 9.53 2.78 Total 102 9.62 2.67
Note. The maximum possible scores for the L2 reading proficiency test were 16. In the experimental sessions, participants in all groups were given two reading tasks. In task 1, participants read individually and answered reading comprehension questions and reference identification questions. In task 2, participants were given another reading passage. The ＋ T － E group read the passage with the targeted form typographically enhanced only; the － T ＋ E group
156
received the passage with written explanations on the concept of anaphors only; the ＋T＋E group received the passage with both TE and written explanations; the －T－E group (the control group) read the original passage. In case the two +E groups carelessly ignored the written instruction and thus undermined the reliability of the experiment, participants were reminded of the presence of the written instruction after they received it. The teacher then read the definition of anaphor and explained the written instruction with examples provided, especially the relation between anaphor and its antecedent and how to comprehend such relation in the examples by using linguistic and world knowledge. All four groups were asked to answer reading comprehension questions and reference identification questions upon completion of their reading, which was controlled for equal durations.
RESULTS
First, participants’ pretest scores on reading comprehension and reference identification questions in task 1 were reported in Table 2.
A one-way ANOVA was conducted to determine whether there is any significant difference among the four groups. The results showed that no significant difference was found in their reading comprehension scores F(3, 98) = .267, p = .85 and reference identification scores F(3, 98) = .075, p = .97. In other words, these four groups were comparable in the sense that all participants had similar reading comprehension and reference identification scores
157
Table 2
Descriptive Statistics of Pretest Scores
Reading comprehension
questions Reference identification questions Group Mean SD Mean SD －T－E (n = 26) 5.69 2.04 7.65 1.32 ＋T－E (n = 22) 5.23 1.90 7.68 1.46 －T＋E (n = 24) 5.71 2.14 7.50 2.21 ＋T＋E (n = 30) 5.50 2.21 7.53 1.33 Total (N = 102) 5.54 2.06 7.59 1.58
Note. The maximum possible scores for reading comprehension questions were 8,
and 10 for reference identification questions.
prior to the treatment. Therefore, any significant difference in the scores of posttest can be more possibly attributed to the effect of treatment than to participants’ preexisting differences.
The current study further analyzed whether there were significant effects of implicit or explicit focus on form on the readers’ anaphor resolution performance and reading comprehension in the posttest with a two-way ANOVA. The two-way ANOVA was used because there were two independent variables in the research design, including textual enhancement and explicit explanations. The independent variable of textual enhancement had two levels: with and without textual enhancement. The independent variable of explicit explanations also had two levels: with and without explicit explanations. In brief, each independent variable had two levels. The dependent variables were the two posttest scores on reference identification and reading comprehension questions. In addition, a
158
paired t-test was conducted to compare participants’ reading comprehension and reference identification scores on the pretests and posttests to determine whether there exists any significant difference before and after the focus on form.
Results of Posttest of Reading Comprehension and Reference Identification
The results of participants’ scores of posttests in task 2 were presented in Table 3.
Table 3
Descriptive Statistics of Posttest Scores
Reading comprehension
questions Reference identification questions Group Mean SD Mean SD －T－E (n = 26 ) 5.42 2.08 7.69 1.64 ＋T－E (n = 22) 5.55 1.54 8.64 1.40 －T＋E (n = 24) 5.50 1.77 7.50 2.72 ＋T＋E (n = 30) 5.60 1.83 8.97 1.22 Total (N = 102) 5.52 1.80 8.23 1.89
Note. The maximum possible scores for reading comprehension questions were 8,
and 10 for reference identification questions.
As shown in Table 3, participants’ reading comprehension performance was roughly equal. As for their performance in the reference identification task, the ＋T＋E group had the highest scores. The two group with the targeted form enhanced (＋T－E and＋T＋E groups ) also outperformed the control group and the group with
159
written explanations only ( － T ＋ E group). A further two-way ANOVA revealed that there was a main effect for TE on readers’ reference identification F(1, 98) = 11.144, p = .00, but not for explicit explanations F(1, 98) = .037, p = .85. Moreover, no interaction effect was found F(1, 98) = .524, p = .47.
The comparison of participants’ pretest and posttest scores also indicated an increase in their overall reference identification performance (Mean from 7.59 to 8.23), but not in their overall reading comprehension (Mean from 5.52 to 5.54). Moreover, when the pretest and posttest scores on reference identification were compared for each of the four groups on the trend chart, the two TE groups had steeper slopes than the other two groups, as revealed in Figure 1.
Figure 1
Pretest and Posttest Scores of Anaphor Identification To further identify whether any significant difference exists in participants’ reference identification from pretests to posttests, a paired t-test was employed to all the four groups. As revealed in
160
Table 4, participants in both TE groups had significant improvement in the posttest of reference identification questions. In contrast, no significant difference was found between the pretest and posttest for the control group and the participants who received explicit explanation only.
Table 4
Paired T-Tests for the Pretest and Posttest Scores of the Anaphor Identification Group Mean difference SD t df p －T－E -.04 1.84 -.11 25 .92 ＋T－E -.95 1.29 -3.47 21 .00 －T＋E .00 2.13 .00 23 1.00 Pretest-posttest ＋T＋E -1.43 1.07 -7.32 29 .00
Correlations Between Reading Comprehension and Reference Identification Scores
The study investigated the correlation between participants’ reading comprehension and their anaphor resolution performance, as indicated in Table 5.
The results showed that participants’ reading comprehension scores are positively correlated with their reference identification scores, from r = .49 to r = .58 in all the four groups. In other words, whether the participants received TE or explicit explanations, the more they comprehended the text, the better they processed anaphoric references, or vice versa. It also shows that correlation coefficient is
161
Table 5
Correlations Between Anaphor Identification and Comprehension Scores
Group Correlation coefficient p n
－T－E .50 .01 26
＋T－E .52 .01 22
－T＋E .49 .02 24
＋T＋E .58 .00 30
Total .47 .00 102
the highest in the ＋T＋E group, implying such a positive correlation might be reinforced by the collaboration of enhanced texts and explicit explanations. Another possibility for this positive correlation between anaphor identification and comprehension, however, may be due to participants’ little form processing in all four groups.
DISCUSSION
According to the empirical results of focus on form from literature (Alanen, 1992, 1995; Doughty, 1988, 1991; White, 1998), both implicit TE and explicit rule explanations were hypothesized in the present study to have beneficial effects on EFL readers’ anaphor resolution performance. Moreover, it was also predicted, based on evidence from previous studies (Leow, 2001; Long & Robinson, 1998), that readers who received treatment of focus on form would have their attention directed to the targeted form, i.e., the anaphor-antecedent relation in texts. From the statistical analyses in this study,
162
implicit focus on form, rather than explicit focus on form, led to better performance on anaphoric reference identification tasks. Readers who received TE only showed scores as high as readers who received both TE and explicit rule explanations. On the other hand, readers who were only given explicit rule explanations during the task had relatively lower scores as the control group did. When participants were given texts of TE, their anaphor resolution performance (M = 8.80, out of 10 maximum possible scores) was better than the others (M = 7.59). This result, therefore, supports the prediction that implicit focus on form, i.e., TE, facilitates anaphor resolution performance in L2 reading. However, it does not support the idea that explicit focus on form alone, i.e., rule explanation, helps promote readers’ anaphoric reference identification as some studies claimed in language learning (Doughty, 1991; N. Ellis, 1993).
One of the fundamental differences between this study and previous focus on form research is the characteristics of the targeted form. Most focus on form literature narrowed the research scope within grammar learning, and researchers analyzed how these grammatical forms could be mastered by learners when their attention was directed to them. In this study, however, the focused form is perceived by learners as one relation that is important in text comprehension and coherence. On the basis of this different characteristic of the targeted form, established principles and empirical evidence from literature are compared with the results in the current study for two reasons. First, the generalizability of focus on form, i.e., whether focus on form has equal or similar effects on grammar learning and other aspects of language learning. Second, the
163
uniqueness of targeted form, i.e., differences between a certain grammatical form and anaphoric reference learning when the input is enhanced. This study confirms the facilitating effect of TE and provides support for previous studies (Izumi, 2003; Jourdenais, Ota, Stauffer, Boyson, & Doughty, 1995; Lee, 2007; Shook, 1994; White, 1996). On the other hand, this study runs counter to other studies claiming the less positive effect of TE (Alanen, 1995; DeKeyser, 1995; N. Ellis, 1993; Leow, Egi, Nuevo, & Tsai, 2003; Overstreet, 1998). Moreover, the result that the implicit, rather than the explicit focus on form group, outperformed the control group is not consistent with Doughty’s (1988, 1991) finding that both the implicit and explicit focus on form groups had improvement in relativization knowledge. This result could be accounted by some possible explanations. First, in grammatical rules learning, explicit learning has the advantage on complex rules as implicit learning does, but has greater facilitating effects on simple rules within a short period of time than implicit learning (DeKeyser, 1995). In learning and comprehension of anaphors, whose function is considered simple and form to be relatively flexible, implicit focus on form outperforms the explicit intervention. Second, this finding may suggest that TE is more salient than the written rules in anaphor comprehension, compared with other grammatical forms. Previous focus on form research (Leow, 2001) tends to see TE as a more implicit intervention than rule instruction in grammar learning. However, in the present study where the targeted form is anaphor, the anaphoric relation seems more transparent and easily perceived by learners when the anaphor and its antecedent were marked by TE. Conversely, although
164
explicit rules given by the teacher clearly states the idea of anaphors, these rules may not be easily consumed by learners without sufficient and concrete examples since these written instructions provide more definition than explanation of anaphors. When the teacher’s explanations are not systematic or processed by readers, these rules may not be easily used by students. Therefore, TE is seen as implicit focus on form in grammar learning, but may be considered more explicit in anaphor comprehension. The definition and function of implicit and explicit in focus on form is thus not fixed for all forms, and what is implicit for certain forms may be explicit for others. This greater salience of TE in anaphor comprehension may explain why the two ＋T groups performed better in the identification task.
The results in this study also indicate that learners’ reading comprehension is positively correlated with their anaphor resolution performance. This correlation was shown the strongest when anaphor knowledge was simultaneously enhanced explicitly and implicitly by providing rules to the readers and by TE, respectively. From a view of bottom-up learning, when readers have sufficient anaphoric reference knowledge, their general reading comprehension may be facilitated. This positive correlation between anaphor knowledge and reading comprehension supports the results of previous studies in L1 reading (Demel, 1990). The result may explain Oakhill and Yuill’s (1986) reports that children of native speakers who were considered poor readers often had difficulty in anaphor resolution. In this regard, such a correlation exists in both L1 and L2 reading. This correlation, however, may also imply a more top-down processing, i.e., the understanding of the text promotes readers’ anaphor comprehension.
165
As suggested by Kurby, Britt, and Magliano (2005), both top-down and bottom-up mechanisms are needed by readers in texts integration. Moreover, it was indicated in this study that learners’ reading comprehension scores did not decline when they experienced either implicit or explicit focus on form (i.e., anaphoric reference). This finding is unlike other previous studies whose targets are certain linguistic forms in that the processing of linguistic forms may impede the processing of meaning, i.e., a trade-off effect between form and meaning processing (Lee, 2007; Overstreet, 1998). According to VanPatten (1990), when L2 learners’ attentional resources are employed to decode the form, their processing of meaning is challenged. In the current study, this trade-off effect does not occur.
These two findings that anaphor knowledge is correlated with reading comprehension and that no trade-off effect exists between form processing and global meaning processing may imply that learners’ global reading comprehension is mainly associated with their knowledge of anaphors, and although focus on form enhances this knowledge, it does not directly affect learners’ global comprehension of texts. That a trade-off effect between form and meaning processing reported in literature was not found in this study could be explained by the strong association between anaphoric reference and general comprehension. Although VanPatten’s (1990) input processing theory claims that L2 learners cannot focus on both the form and meaning simultaneously, and that their meaning processing is hindered by form processing, Lightbown (1998, p. 192) suggests that learners do benefit from both foci on form and meaning when “the form in focus is an important carrier of the meaning in
166
focus.” According to Lightbown, learners have problems with targeted forms when these forms do not contribute crucial meaning in focus. In this study, unlike other linguistic forms, the processing of anaphoric reference in texts is crucially associated with the processing of textual meaning because the function of anaphors is to construct text coherence. Therefore, learners’ attentional resources for meaning and form processing are not competing, but may be mutually transferable.
The generalizability of this study is constrained by some factors, including the certain type of anaphoric references chosen in this study, the specific language level and characteristics of the participants (e.g., high prior knowledge of anaphoric relationships in four groups, as shown in the pretest), and other textual factors that might affect readers’ anaphor resolution, e.g., the distance between the anaphor and its antecedent. Moreover, it is difficult to determine the durability of participants’ performance in the enhanced form and their attention with a lack of a delayed post-test and follow up interviews. In this regard, future studies analyzing how L2 learners’ attentional resources are located in their anaphor comprehension and acquisition are needed. It is expected that findings in the present study can provide some insights for the future exploration.
CONCLUSION
This study provides both important theoretical and pedagogical implications in the effect of focus on form on L2 learners’ anaphoric reference comprehension. It was found that implicit focus on form
167
(TE), rather than explicit one (rule explanation) facilitated L2 learners’ anaphoric reference comprehension, at least for the learners with high prior knowledge of anaphoric relationships in this study. It was also found that L2 learners’ attentional resources for the meaning processing were not sacrificed when they decode the enhanced form of anaphors. These results thus imply two things. First, supported by previous studies maintaining the facilitating effect of TE on grammatical forms, TE might also make the enhanced anaphors salient to L2 readers and improve their understanding of antecedent-anaphor relation in texts. In antecedent-anaphor comprehension, the enhanced antecedent-anaphor relation by TE may be more salient than the explicit rule given by the teacher. Second, unlike other focus on form studies that reported the trade-off effect between form and meaning processing, the positive correlation between anaphor comprehension and reading comprehension shows that learners’ attentional resources are not necessarily depleted through competition.
In terms of pedagogical implications, it is suggested that teachers integrate TE in anaphoric reference teaching. As the study suggests, explicit rule explanations for anaphors may not be as effective as TE. Teachers’ use of TE in anaphor teaching also helps three things (Zhang, 2010). First, TE unites form and meaning in texts, which is the core in language teaching nowadays. Second, TE is helpful in discourse-based language teaching, and it compensates what sentence-based grammar teaching cannot achieve. Third, TE is more easily manipulated by teachers than other teaching techniques. Moreover, since anaphoric reference comprehension is correlated with reading comprehension and the form of anaphor is more flexible
168
than its function, it is suggested that teachers develop individual learner-centered activities (e.g., problem-solving tasks) and attention-focusing devices (e.g., TE designed reading materials) to raise learners’ attention of the targeted anaphors.
REFERENCES
Alanen, R. (1992). Input enhancement and rule presentation in
second language acquisition (Unpublished master’s thesis).
University of Hawaii at Manoa, Honolulu, HI.
Alanen, R. (1995). Input enhancement and rule presentation in second language acquisition. In R. Schmidt (Ed.), Attention and
awareness in foreign language learning and teaching (pp.
259-302). Honolulu, HI: University of Hawaii Press.
Britton, B. K., & Gulgoz, S. (1991). Using Kintsch’s computational model to improve instructional texts: Effects of repairing inference calls on recall and cognitive structures. Journal of
Educational Psychology, 83, 329-345.
Cohen, A. D., & Fine, J. (1978). Reading history in English: Discourse analysis and experience of native and non-native reader. Working Papers on Bilingualism, 16, 55-74.
DeKeyser, R. (1995). Learning second language grammar rules: An experiment with a miniature linguistic system. Studies in
Second Language Acquisition, 17, 379-410.
DeKeyser, R. (2005). Beyond focus on form: Cognitive perspectives on learning and practicing second language grammar. In C. Doughty & J. Williams (Eds.), Focus on form in classroom
169
second language acquisition (pp. 42-63). New York, NY:
Cambridge University Press.
Dell, G., McKoon, G., & Ratcliff, R. (1983). The activation of antecedent information during the processing of anaphoric reference in reading. Journal of Verbal Learning and Verbal
Behavior, 22, 121-132.
Demel, M. C. (1990). The relationship between overall reading comprehension and comprehension of coreferential ties for second language readers in English. TESOL Quarterly, 24, 267-292.
Dik, S. C. (1997). The theory of functional grammar. Berlin, Germany: Mouton de Gruyter.
Doughty, C. (1988). The effect of instruction on the acquisition of
relativization in English as a second language (Unpublished
doctoral dissertation). University of Pennsylvania, Philadelphia, PA.
Doughty, C. (1991). Second language instruction does make a difference: Evidence from an empirical study of ESL relativization. Studies in Second Language Acquisition, 13, 431-469.
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. 197-262). New
York, NY: Cambridge University Press.
Ehrlich, M. F., Remond M., & Tardieu, H. (1999). Processing of anaphoric devices in young skilled and less skilled
170
comprehenders: Differences in metacognitive monitoring.
Reading and Writing, 11, 29-63.
Ellis, N. (1993). Rules and instances in foreign language learning: Interactions of implicit and explicit knowledge. European
Journal of Cognitive Psychology, 5, 289-319.
Ellis, R. (1990). Instructed second language acquisition. Oxford, England: Basil Blackwell.
Gallini, J. K., & Spires, H. A. (1992). The influence of anaphoric relations and departures from story grammar structure on text processing. Reading Psychology, 13, 107-130.
Garrod, S., & Sanford, A. (1990). Referential process in reading: Focusing on roles and individuals. In D. Balota, G. B. F. d’Arcais, & K. Rayner (Eds.), Comprehension processes in
reading (pp. 465-484). Hillsdale, NJ: Erlbaum.
Givón, T. (1995). Functionalism and grammar. Philadelphia, PA: John Benjamins.
Han, Z. H., Park, E. S., & Combs, C. (2008). Textual enhancement of input: Issues and possibilities. Applied Linguistics, 29, 597-618. Huang, S. H. (2005). Assessing the relationship between referential
understanding and academic reading comprehension among EFL college students (Unpublished master’s thesis). National
Yunlin University of Science and Technology, Yunlin, Taiwan. Huang, Y. (2006). Anaphora, cataphora, exophora, logophoricity. In K. Brown (Ed.), Encyclopedia of language and linguistics (pp. 231-237). Boston, MA: Elsevier.
171
Hulstijn, J., & De Graaff, R. (1994). Under what conditions does explicit knowledge of a second language facilitate the acquisition of implicit knowledge? AILA Review, 11, 97-112. Izumi, S. (2003). Visual input enhancement as focus on form. Sophia
Linguistica, 51, 1-30.
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, HI: University of Hawaii.
Kurby, A. C., Britt, M. A., & Magliano, J. P. (2005). The role of top-down and bottom-up processes in between-text integration.
Reading Psychology, 26, 335-362.
Lee, S. K. (2007). Effects of textual enhancement and topic familiarity on Korean EFL student’s reading comprehension and learning of passive voice. Language Learning, 57, 87-118. Leow, R. (2001). Do learners notice enhanced forms while interacting
with the L2? An on-line and off-line 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.
Lightbown, P. M. (1998). The importance of timing in focus on form. In C. Doughty & J. Williams (Eds.), Focus on form in
classroom second language acquisition (pp. 177-196). New
172
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, NY: Cambridge University Press.
Lyddon, P. A. (2011). The efficacy of corrective feedback and textual enhancement in promoting the acquisition of grammatical redundancies. Modern Language Journal, 95, 104-129.
Oakhill, J., & Yuill, N. (1986). Pronoun resolution in skilled and less skilled comprehenders: Effects of memory load and inferential complexity. Language and Speech, 29, 25-37.
Overstreet, M. (1998). Text enhancement and content familiarity: The focus of learner attention. Spanish Applied Linguistics, 2, 229-258.
Peterson, C. (1993). Identifying referents and linking sentences cohesively in narration. Discourse Processes, 16, 507-524. Shapiro, A., & Milkes, A. (2004). Skilled readers make better use of
anaphora: A study of the repeated-name penalty on text comprehension. Electronic Journal of Research in Educational
Psychology, 2, 161-180.
Shook, D. J. (1994). FL/L2 reading, grammatical information, and the input-to-intake phenomenon. Applied Language Learning, 5, 57-93.
Simard, D. (2009). Differential effects of textual enhancement formats on intake. System, 37, 124-135.
VanPatten, B. (1990). Attention to form and content in the input: An experiment in consciousness. Studies in Second Language
173
White, J. (1996). An input enhancement study with ESL children:
Effects on the acquisition of possessive determiners
(Unpublished doctoral dissertation). McGill University, Montréal, Canada.
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.
91-128). New York, NY: Cambridge University Press.
Williams, J., & Evans, J. (1998). What kind of focus and on which forms? In C. Doughty & J. Williams (Eds.), Focus on form in
classroom second language acquisition (pp. 139-155). New
York, NY: Cambridge University Press.
Zhang, B. (2010). The effects of textual enhancement on grammar teaching. Higher Education Forum, 10, 74-76.
ABOUT THE AUTHOR
Ting Fang got his B.A. in Department of Foreign Languages and Literatures from National Taiwan University, and got his M.A. and is currently a PhD student in English Department (TESOL) of National Taiwan Normal University. His main research interest is second language acquisition, particularly language strategies and psycholinguistics.
174
APPENDIX A
The Explicit Instruction Provided to the Explicit Focus on Form Group 指涉詞是一種使文章句意連貫通順的字詞，它被用來指先前或之 後出現的人事物。例如：「這輛車很貴，所以我不想買它。」這 裡的它即用來指前面出現的這輛車。指涉詞與前／後面出現過的 人事物之間的距離可長可短。例如：「這輛車很貴，加上我很不 喜歡它的顏色跟設計……所以我不想買它。」
英 文 範 例 ：Mary and her friend, Sam, went to the department yesterday. They had a quarrel because Sam did not like the clothes she bought. Mary was very angry because it cost her 1000 dollars and that is all the money she had.
175
APPENDIX B
Samples of the Experimental Texts Enhanced by Implicit Focus on Form and the Comprehension and Reference Identification
Tests The experimental text
What can you do if you ‘lose’ the data from your disks? To find an expert, who would recover the lost information for you, is probably the easiest solution. Jack Olson is one of these experts. Jack and a few of his friends set up a company called “Jack’s Disk Doctor Service” in 1984. They work from home and give all the money they earn to charity. The fees are always the same, no matter how precious the data on the disk is. Some people, however, are so grateful that they send extra money to Jack or to the charities his company supports. One oil company offered him $2000 for his help and an architect even sent him a blank check.
It would be difficult to put a value on the things rescued by the Disk Doctor. There have been disks containing medical research, television scripts, manuscripts of whole books, a lawyer’s papers for a court case, and even Margaret Thatcher’s travel plan for a visit to eastern Europe. For this last case, Jack had to go in person to Thatcher’s office “for security reasons!”
Disks are usually sent to the Disk Doctor by post, but sometimes people are in such a hurry that they cannot wait for the mail to come. For example, some radio scripts had to be rushed by taxi to Jack’s house because they were needed for broadcasting the
176
next day. When the material has been recovered, the disk is returned to the sender with a diagnosis and a prescription for avoiding the problem in the future. One grateful client, an author, put a “thank you” to Jack in the front of his book. “Jack saved me from a heart attack,” he wrote. “But,” says Jack, “most people don’t take any notice of the doctor’s advice!”
Sample of the Comprehension Question Why did the architect send Jack a blank check? (A) The architect did not have any money (B) The architect did it for security reasons
(C) The architect always followed the doctor’s advice (D) The architect thought Jack’s service was priceless Sample of the Reference Identification Question What does “they” in line 16 refer to?
177
