2. Theoretical Background
2.1. Interactive Alignment
IA was first proposed by Pickering and Garrod
| [14] | Pickering, M. J., & Garrod, S. (2004). Toward a mechanistic psychology of dialogue. Behavioral and Brain Sciences, 27(2), 169-90; discussion 190-226. |
[14]
in order to explore the mechanism underlying language use in human dialogues in which one speaker forms a dynamic context-specific model according to the situation presented by the other speaker
| [41] | Zwaan, R., & Radvansky, G. (1998). Situation models in language comprehension and memory. Psychological Bulletin, 123(2), 162-185. |
[41]
. The two speakers of this given dialogue then coordinate and inspire each other wherein they are constantly constructing situation models that are similar to one another’s so as to understand the meaning of the discourse. This kind of cognitive coordination/alignment that occurs in the interaction process is known as IA
| [22] | Wang, C. (2010). Interactive alignment and foreign language teaching. Foreign Language Teaching and Research, (4), 297-299. |
[22]
. Following ongoing exploration, studies on IA have extended this concept from use in solely human dialogues to other fields, such as that between humans and their non-human environments
| [1] | Atkinson, D., Churchill, E., Nishino, T., & Okada, H. (2007). Alignment and interaction in a sociocognitive approach to second language acquisition. The Modern Language Journal, 91(2), 169-188. |
[1]
or between language learners and their reading materials
| [22] | Wang, C. (2010). Interactive alignment and foreign language teaching. Foreign Language Teaching and Research, (4), 297-299. |
[22]
. Further, Wang
| [22] | Wang, C. (2010). Interactive alignment and foreign language teaching. Foreign Language Teaching and Research, (4), 297-299. |
| [23] | Wang, C. (2012). The continuation task: An effective way to facilitate L2 learning. Foreign Language World, (5), 2-7. |
[22, 23]
has also noted that, in IA, one’s language output tends to be in line with the input language, which then takes the forms of imitation and repeated application of the interlocutor’s language.
In IA, comprehension and production share similar representation
| [14] | Pickering, M. J., & Garrod, S. (2004). Toward a mechanistic psychology of dialogue. Behavioral and Brain Sciences, 27(2), 169-90; discussion 190-226. |
[14]
. This statement then lays a theoretical foundation for further study of the IA effect between language production and comprehension
| [32] | Wang, C., & Wang, M. (2014). The effects of alignment in continuation tasks. Modern Foreign Languages, (4), 501-512+584. |
[32]
in a way that then contributes to the further studying of IA from the perspective of the situation model. According to Zwaan and Radvansky
| [41] | Zwaan, R., & Radvansky, G. (1998). Situation models in language comprehension and memory. Psychological Bulletin, 123(2), 162-185. |
[41]
, language comprehension involves constructing a mental representation of the described situation in a given text (which is then termed the situation model), which can be further investigated through the event-indexing model. In the event-indexing model, when people read a clause, they then tend to construct a model of the situation that is denoted by that clause; herein, each event depicted by that clause can be indexed on five dimensions, including time, space, causation, motivation/goal, and entities (which include both protagonists and objects). That is, a situation model incorporates the temporal, spatial, causal, motivational, and person- and object-related information of a given context
| [41] | Zwaan, R., & Radvansky, G. (1998). Situation models in language comprehension and memory. Psychological Bulletin, 123(2), 162-185. |
[41]
. Based on the chronology of event(s), a situation model formed in a given reader’s cognition can be categorized into three types: (1) the current model, which is the model currently under construction at a given Time (tn); (2) the integrated model of the situations at Times t1 through to tn-1; and (3) the complete model of the situations at Times t1 through to tx. The current model is constructed at Time tn as the person in question reads a particular clause or sentence, called cn. The integrated model is the global model that is constructed by integrating (one at a time) the models that were constructed at Times t1 to tn − 1 while the reader reads clauses c1 to cn − 1. Finally, after all the textual input has been processed the model is complete and stored in the reader’s memory
| [41] | Zwaan, R., & Radvansky, G. (1998). Situation models in language comprehension and memory. Psychological Bulletin, 123(2), 162-185. |
[41]
. However, the complete model is not the final one because readers often ruminate over the story and generate additional inferences or develop entirely new models later on
| [41] | Zwaan, R., & Radvansky, G. (1998). Situation models in language comprehension and memory. Psychological Bulletin, 123(2), 162-185. |
[41]
. Thus, when a person reads a text, the forming of the situation model in their cognition is both dynamic and is constantly updated.
Generally speaking, IA can occur at both the language and the situation model levels. The alignment that happens at one level can then lead to an alignment at the other
| [32] | Wang, C., & Wang, M. (2014). The effects of alignment in continuation tasks. Modern Foreign Languages, (4), 501-512+584. |
[32]
. An observable indicator of IA at the language level is Structural Priming, which means that people, during language output, tend to repeat language structures (e.g., vocabulary and grammatical structures) that they have been exposed to or that have been used by others
| [2] | Bock, J. K. (1986). Syntactic persistence in language production. Cognitive Psychology, 18(3), 355-387. |
[2]
.
CT, which were first proposed by Wang
| [22] | Wang, C. (2010). Interactive alignment and foreign language teaching. Foreign Language Teaching and Research, (4), 297-299. |
[22]
, initially referred to a kind of writing task in which reading and writing, imitation and creation, as well as language input and output are all combined. In CT design, the ending paragraph(s) of a text (usually a story) are not given, with only the first sentence of the these concluding sections being shown, meaning that students are then required to complete the text based on their own reading of it.
The emergence of CT has a close relationship to the increased exploration of second/foreign language writing by linguists. For example, Swain
| [17] | Swain, M. (2006). Languaging, agency and collaboration in advanced second language learning. In H. Byrnes (Ed.), Advanced Language Learning: The Contributions of Halliday and Vygotsky (pp. 95–108). London: Continuum. |
[17]
suggested that language is not only the cognitive object of learning, but also the most important medium of learning. When people use language to solve problems, whether it is about choosing words/sentences to express their own thoughts, interpreting experimental results, or explaining other people’s behavior, the occurrence of “languaging” is promoted. This view of languaging, from a sociocultural perspective, provides support for the validity of the Input Hypothesis and Output Hypothesis on the basis that media functions during both the input and output processes
| [16] | Swain, M. (2005). The output hypothesis: Theory and research. In E. Hinkel (Ed.), Handbook of Research in Second Language Teaching and Learning (pp. 471-483, Chapter 26, 1144 Pages). Mahwah, NJ: Routledge. |
| [17] | Swain, M. (2006). Languaging, agency and collaboration in advanced second language learning. In H. Byrnes (Ed.), Advanced Language Learning: The Contributions of Halliday and Vygotsky (pp. 95–108). London: Continuum. |
[16, 17]
; hence, this then theoretically corroborates the design and application of CT in EFL writing class
| [39] | Zhang, L., & Sun, Y. (2014). The practice of writing classroom reform for English majors from the perspective of sociocultural theory. Foreign Language World, (5), 2-10. |
[39]
.
Wang pioneered the research of CT in EFL reading-to-writing, in which they empirically started with a one-semester-long experiment
| [33] | Wang, C., Niu, R., & Zheng, X. (2000). Improving English through writing. Foreign Language Teaching and Research, 32(3), 207-212. |
[33]
on improving Chinese EFL learners’ English by means of composition writing. This line of research has since been advanced into examining interaction in language learning from the perspectives of dynamic system theory, the IA model, and the construction of grammar, with the fact that IA plays a central role in language use and that priming has a strong connection with language learning overall
| [21] | Wang, C. (2009). A study of the interrelationship between L2 learning variables and L2 use. Foreign Languages in China, (5), 53-59. |
[21]
. Wang
| [21] | Wang, C. (2009). A study of the interrelationship between L2 learning variables and L2 use. Foreign Languages in China, (5), 53-59. |
[21]
clarified the pivotal role of IA in L2 learning, with them then formally advocating for the application of CT in L2 learning and teaching
| [22] | Wang, C. (2010). Interactive alignment and foreign language teaching. Foreign Language Teaching and Research, (4), 297-299. |
[22]
. He believed if reading and writing are combined herein, students will be forced to form a deeper understanding of the content of the text, as well as cooperate/align with it. Additionally, while interacting with the content, structural priming is then catalyzed. In the process of completing combined tasks like CT, students are tasked with constantly reviewing the original text, consciously or unconsciously imitating it, and with using some of its expressions so as to reduce the issues in their own writing. Therefore, CT facilitate the interaction between readers and the text, between readers and the content of the text, and between readers’ language structure and that of the original text. Its practical benefits have been identified and several principles and requirements have been proposed
| [23] | Wang, C. (2012). The continuation task: An effective way to facilitate L2 learning. Foreign Language World, (5), 2-7. |
[23]
. As an assessment task, the washback effect, feasibility, and potential of CT were explored and used to navigate its application in the provincial and national matriculation test
| [31] | Wang, C., & Qi, L. (2013). A study of the continuation task as a proficiency test component. Foreign Language Teaching and Research, (5), 707-718+800. |
[31]
.
The notion of language imitation and idea creation in L2 education has been proposed from the perspective of IA
| [24] | Wang, C. (2014). Idea creation and language imitation: An effective route to L2 learning and teaching. Foreign Language World, (2), 42-48. |
[24]
with this being empirically tested by prior research
| [32] | Wang, C., & Wang, M. (2014). The effects of alignment in continuation tasks. Modern Foreign Languages, (4), 501-512+584. |
[32]
, as well as its underlying mechanism
| [25] | Wang, C. (2015). Why does the continuation task facilitate L2 learning? Foreign Language Teaching and Research, 47(5), 753-762. |
[25]
. Wang
| [26] | Wang, C. (2016). Learning by extension. Modern Foreign Languages, 39(6), 784-793+873. |
[26]
further advocated for a hypothesis that language is successfully learned through extension through which he proposed the
Xu-argument, which maintains that via
Xu or Completion, Extension and Creation (CEC), language is learned in high efficiency. The
Xu-argument (also the CEC approach), is still being tested
| [28] | Wang, C. (2018). How to improve the intensity of interaction in continuation task. Foreign Language World, (5), 40-45. |
| [29] | Wang, C. (2019). Caveats for applying continuation tasks. Foreign Languages and Teaching, (3), 1-7. |
| [30] | Wang, C. (2020). A fundamental issue in L2 learning: How to map static linguistic knowledge properly onto dynamic ideas? Modern Foreign Languages, 43(5), 593-600. |
[28-30]
including in an experiment by Peng, Wang, and Lu
| [13] | Peng, J., Wang, C., & Lu, X. (2020). Effect of the linguistic complexity of the input text on alignment, writing fluency, and writing accuracy in the continuation task. Language Teaching Research, 24(3), 364-381. |
[13]
, which revealed that the linguistic complexity of the input text significantly affects L2 learners’ alignment, writing fluency, and writing accuracy in CTs. Thus, understandings generated by the current study are intended to help provide new critical insights into how the language learning potential of CT can be maximized in the EFL context.
4. Findings and Discussions
4.1. Relationship Between Reading and Writing in the Continuation Task
Wang
| [25] | Wang, C. (2015). Why does the continuation task facilitate L2 learning? Foreign Language Teaching and Research, 47(5), 753-762. |
[25]
claimed that reading is the foundation of continuation writing, which is consistent with the results of this study. In CT, reading and writing adopt the same form of representation—the situation model
| [14] | Pickering, M. J., & Garrod, S. (2004). Toward a mechanistic psychology of dialogue. Behavioral and Brain Sciences, 27(2), 169-90; discussion 190-226. |
| [41] | Zwaan, R., & Radvansky, G. (1998). Situation models in language comprehension and memory. Psychological Bulletin, 123(2), 162-185. |
[14, 41]
. The situation models in a reader’s cognition are dynamic. New situation models in one’s cognition then provide them with new material for writing.
This study found, however, that the situation model in a reader’s cognition may not perfectly align with the scenario depicted by the given text, or with the continuation work actually written by the reader. On the one hand, if there is a discrepancy between the text and the reader’s comprehension outcome, the situation model constructed in their cognition will be inconsistent with the text. Then their continuation work, which is the linguistic reflection of their developed situation model, may not align with the provided text. For example, as our think-aloud data showed, when doing the CT of text A (CT A), P1 reported as follows:
Then, they were at the airport. Oh no, they were on the plane. So, other passengers must be curious. His shouts (Should “shout” be in the plural form) was [were] so loud that other passengers stared at them and confused.
In this case, after looking back to the text, P1 thought that the protagonists of the story were on the plane while in fact they were at the terminal of the airport according to the original text. Therefore, the situation model constructed by P1 contradicted the situation presented in the text. However, it must also be noted that, whether a reader’s situation model aligns with the reading text or not, when they are writing according to their unique model, their writing work may not fully align with their cognitive situation model either. This is related to the writer’s language comprehension and expression abilities, as well as the specific part of their situation model that they intend to present in their writing work. For example, P1 mistook “in the terminal” for “on the plane” meaning that the spatial information of her situation model contradicted that of the reading text. Nonetheless, what P1 actually wrote in her continuation work is “His shouts was [were] so loud that other passengers stared at them and [got] confused.” As such, P1 did not explicitly state “on the plane” in her continuation work, which would have reflected the spatial discrepancy between her situation model and the original text. Furthermore, the phrase “other passengers” who were “on the plane” in her situation model was not modified by “on the plane” in her actual writing work, meaning that P1’s actual written story still aligns with that of the reading text because “other passengers” could exist both “in the terminal” and “on the plane.” Considering this, it can be inferred that, in CT writing, a person’s produced text is not directly aligned with the provided reading text but is rather presented via a medium— their situation model.
In summary, the process of reading involves the reader constantly constructing and updating the situation models in their cognition, with the situation model formed in this reading process being the direct foundation for their writing. Reading, including reading back while writing, provides motivation and updated content for continuations in writing via constantly constructing, updating, and retrieving one’s situation models. The language structures (vocabulary and syntax) of the reading text are then the necessary materials for the transformation from cognitive construction (situation models) to the production of linguistic text (continuation writing work).
4.2. Interactive Alignment at the Content Level
Referring to the situation model theory, this study coded and analyzed IA in CT from five situational dimensions: entity (person and object) (Part 4.2.1), time (Part 4.2.2), space (Part 4.2.3), causation (Part 4.2.4) and motivation (Part 4.2.5).
4.2.1. Person- and Object-Related Alignment
In CT A, both P1’s and P2’s continuation writings mentioned characters (Dannis, the black-haired girl, and the police) from the original text (additionally, P1 mentioned “the blonde girl” from the reading text as well). The traits (e.g., gender, name, and appearance) of these characters in the participants’ writings were consistent with those in the reading text, thus showing the person-related alignment between their continuation writing and the original text.
For example, in the think-aloud data, P1 reported: “She seemed absorbed in a book” following which she wrote in her continuation work as follows:
“What are you reading?” Dannis asked. “The kite runner” “A great book, I like this book too.” Dannis said.
In this example, the material in P1’s writing work is consistent with that of the original text through their inclusion of the object “book,” which demonstrates the object-related alignment between her writing and the original text.
Also in CT A, P2’s think-aloud data was reported as follows:
Dannis pointed at his pocket, because the original text has mentioned that his wallet had been in his pocket, thence it should be that he pointed at his pocket.
Accordingly, P2’s continuation work in this section is: Dannis asked with anger, “What did you do when you were so near to me?” Dannis pointed at his pocket. In this example, P2’s writing echoes the object-spatio connection “wallet had been in the pocket” mentioned in the original text, indicating an object-related situational alignment. Furthermore, in the continuation task of text B (CT B), P2 circled various crucial objects mentioned in the original text, such as “yellow blouse,” “berries,” and “helicopter,” and reproduced these in her own writing work, thus improving the object-related situational alignment between her writing and the original text.
4.2.2. Temporal Alignment
The results of this study also reveal that the alignment on the temporal dimension between a writer’s continuation work and the original text is achieved through the use of tenses, specific time nouns/phrases (e.g., “another night passed”), and particular verbs/verb phrases (e.g., “rose her head,” and “watch them running” in the following examples).
In CT A, P1 reported in the think-aloud process that: “she seemed absorbed in a book,” and wrote the following in her continuation work:
“Hello miss.” Dannis said to the black-hair woman and smile [smiled]. The woman rose [raised] her head and smile [smiled]: “Hello Mr.” “What are you reading?” Dannis asked…
In this example, the three events— 1) the girl reading, 2) Dannis walked to the girl and saying hello, and 3) the girl raising her head from her book— are a series of successive actions/activities. Furthermore, Dannis used the present continuous tense (“what are you reading?”) when talking to the girl, thereby keeping the situation in the same temporal domain as “the girl had been reading a book” that was constructed within the original text as “she seemed absorbed in a book.” All of these indicate an accurate temporal alignment between the original text and the writer’s continuation work. Furthermore, P2 wrote the following in the second paragraph of her continuation work:
… “Catch her, she is the thief!” Dannis shouted to the crowd. The girl shouted, “Catch him, he is the person who has always followed me! I don’t even know him!” The crowd kept watch [watching] them running.
The first sentence of the second continuation paragraph in the original text is “The girl ran and Dannis ran after her.” Thus, the sentence “The crowd kept watch [watching] them running” showcases that the actions/events “Dennis shouted...,” “The girl shouted…,” and “The crowd kept watch [watching]…” all occur in the same temporal domain as the action “ran” within the quote “The girl ran and Dannis ran after her.,” Therefore, IA has been achieved within the temporal dimension.
However, although participants were consciously keeping the tense of their writing consistent with that of the original text throughout the task, due to constraints in their language proficiency, they would sometimes forget to keep the tense consistent or were not able to express the time in the correct language forms. For instance, in CT A, P1 mentioned numerous times in her think-aloud that she should use the past tense in her writing; nevertheless, in her actual writing, she spelt “grab” wrong as “grap” and did not use the past tense therein. As a result, her continuation writing and the original text did not align well in terms of the temporal domain. P2 had similar experiences when completing CT B. In CT B, P1 reported the following in her think-aloud session:
Then, maybe she was too scared or too tired that she just passed out in her husband’s arms. How about this? Otherwise, it will be difficult to write the second paragraph.
In this case, P1 inferred from the first sentence “It was daybreak when Jane woke up” of the second continuation paragraph given by the original text that the ending of her first continuation paragraph must mention that Jane passed out. Because “passed out” creates a temporal connection with “woke up,” this showcases a temporal alignment between P1’s situation model and the original text. However, because of her insufficient vocabulary, she was unable to express the meaning of “passed out,” meaning that she then used another expression and wrote “Jane remembered nothing” instead of “Jane passed out” at the end of her first continuation paragraph (She felt so tired that nothing can remember [be remembered] after that). In other words, the actual linguistic expressions used by P1 in her writing have a much weaker alignment with the original text than the possible alignment her situation model could have made. This is also in line with the results of Wang’s (2015) study which found that the IA between the continuation writing work and the original text is influenced by the reader/writer’s language ability.
At the spatial dimension of the content, the results suggest that the spatial alignment between a writer’s continuation work and the original text can be achieved through the use of location nouns/phrases (e.g., the use of “open area” in the following examples) and particular non-location nouns/phrases (e.g., “the crowd,” “other passengers,” and “leaves” in the following examples).
In CT B, P2 achieved spatial alignment between her writing and the original text by using the explicit spatial expression “open area” from the original text. Further P1 achieved a spatial alignment between her writing and the original text by creating a new object “leaves” that indirectly aligns with the space (in the forest) depicted within the original text. In another example, during CT A, P2 read back the original text and reported as follows:
Then, the girl ran and Dannis ran after her, and a dramatic scene emerged. In the terminal, this girl run and Dannis ran after her. Dannis might turn to the crowd and ask the crowd to catch the girl… let me read back, they were sitting in the terminal. Yes, this girl might be very smart, so she might also shout to the crowd that…
In this example, P2 knew that the event in which “Dannis ran after the girl” was happening “in the terminal” wherein there might be other people, expressed as “the crowd” in P2’s words. Thus, in her continuation writing, Dannis asked the crowd for help. In this case, despite the fact that “the crowd” is a newly-constructed person-related model, it still manages to suggest some spatial information, thereby demonstrating the spatial alignment between P2’s writing and the original text. Similar results were found in CT A of P1.
However, as discussed previously, P1’s think-aloud data suggested that, although her continuation writing did align with the original text within the spatial dimension, her actual comprehension outcome (i.e., her situation model) is inconsistent with the original text in terms of this dimension. This discrepancy from P1 was confirmed in the interviews.
This study also found that numerous IA phenomena occurred during the think-aloud and continuation works of P1 and P2 in which they were trying to construct as many causal connections as possible between the original text and their written works. In the think-aloud data of CT A, P1 reported:
Then, what should Dannis do to get his wallet back? The original text says that later the girl ran and Dannis ran after her….
In this example, P1 encountered a consequence signaled by the original text (“The girl ran and Dannis ran after her”), meaning that she then attempted to construct a causal connection between the preceding and incoming events. Furthermore, considering that (a) the original text mentioned that Dannis thought it was the girl who had stolen his wallet and (b) P1 inferred the motivation of Dannis approaching the girl was to get his wallet back, P1 wrote her corresponding continuation work as follows:
But when he continue [continued] to say something, the woman stood up. Dannis realized that the woman maybe wanted to run away, so he stood up too and grap [grabbed] her left arm.
As a result, P1 constructed several casual connections between the original texts and her own continuation work, through which she made her writing consistent with the reading text in terms of causality. Another example was found in CT B and P2 reported the acts of careful reading and rereading, looking for the cause, inferring and constructing a consequence. These causal connections constructed suggest that an IA on causality did occur between P2’s continuation writing and the original text.
4.2.5. Motivational Alignment
According to Zwaan and Radvansky
| [41] | Zwaan, R., & Radvansky, G. (1998). Situation models in language comprehension and memory. Psychological Bulletin, 123(2), 162-185. |
[41]
, readers can foreground information based on their general knowledge because of what they know about human motivations/goals and actions or about the narrative genre as a whole. In the process of reading, a person may find various motivations/goals outlined in the text, with some of these then being carried throughout the whole writing process. In this study, these are termed the dominant motivation(s). The dominant motivation/goal of story A is that Dannis wants to get his wallet back and the dominant motivation of story B is that Jane wants to either find her way back or to be found/saved by others.
The think-aloud data showed that, in CT A, P1 reported various times that “Dannis wanted to get his wallet back.” That is, P1 found the dominant motivation/intentionality of this story, based on which she then constructed certain plot points, such as: 1) Dannis started the conversation with the black-haired girl; 2) Dannis “grap [grabbed] her left arm” when “Dannis realized that the woman maybe wanted to run away;” 3) Dannis shouted, “My wallet! Stop!;” 4) Dannis called for the police and “then a police came and catched [caught] the woman.” These examples showcase an alignment on intentionality/motivation.
Further, story B mentioned that, when Jane was lost in the forest, “She wanted to hold him (her husband Tom) and tell him how much she loved him.” Keeping this intention/motivation in mind, P1 and P2 wrote the following in their continuation work::
Tom found her quickly, and they hugged tightly. “Oh Tom, I love you. I was so foolish to quarrel with you.” (P1)
Jane cried and said, “I will never quarrel with you, I love you.” (P2)
Both of these excerpts show that IA on the intentionality/motivation dimension did occur.
4.3. Interactive Alignment at the Language Level
In CT A, as shown by the think-aloud data, P1 reread the text numerous times for particular words/phrases and directly copied some of these from into her own writing, including “Dannis,” “black hair girl,” “blonde girl,” and “realized.” P2 circled various language structures (e.g., “so near to me,” “his pocket,” “shiny black hair”) and used them in her own continuation work. In CT B, P2 achieved a high degree of alignment between her writing and the original text through the use of most of the language structures from the original text, including vocabulary like “kept moving/walking,” “so angry,” “hungry and thirsty,” “see,” “flagged,” “yellow blouse,” “open area,” “took off,” “berries,” “strong” “quarrel,” and certain sentence patterns like “the farther she walked, the more…,” “If only…,” and “she walked to an open area.” In terms of P1’s writing work, an alignment at the language level was shown primarily through the copying of the names of the protagonists while there was little application of other expressions from the original text. It should be noted that, in CT B, P1 consciously imitated the “If only…” syntax of the original text, implying a tendency towards aligning the language of her own writing with that of the reading material. However, due to her language inefficiency, her use of language imitation and application failed due to a grammatical mistake. In this respect, we can conclude that, in CT, a writer’s language ability is critical, not only in content alignment, but also in terms of language alignment.
It is worth noting that, in CTA, P1 reported as follows:
(Give the wallet back to me. How can I express “give back” in English? How to express “give back to me”? “pay back”? no, that is used for money. Well, how about Dannis directly shouting “my wallet”?…)
The related writing work of this think-aloud section is:
“You thief” Dannis shout [shouted]
, “My wallet! Stop!” In this example, P1 wanted to use a wider variety of words and expressions but, due to her limited vocabulary, she was only able to use the expression “
my wallet” from the original text. This example suggests that, on one hand, CT does provide learners with certain language expressions that they can then imitate and apply in their own continuation writing for the purpose of creating new content. This is consistent with Wang’s findings
| [25] | Wang, C. (2015). Why does the continuation task facilitate L2 learning? Foreign Language Teaching and Research, 47(5), 753-762. |
[25]
.
Furthermore, we found that, in CT A, the language alignment was primarily demonstrated through the writers’ vocabulary, with there being no alignment in syntax. Conversely, in CT B, language alignment was observed, not only at the vocabulary level, but also at the syntax one. This could be explained via the learning effect that would have occurred after the participants had completed the first task. However, after a close comparison between the two reading materials, we found that text B has more abundant sentence patterns that are worth imitating (e.g., “the more…the more…” and “only if…”). In this regard, we found that, in CT, the specific reading material plays a crucial role in the process of students learning a foreign/second language. The more abundant the language structures (including syntax, phrases, etc.) that a reading material contains, the more likely that language learners would be able imitate these and apply them to their own writing.
4.4. Attitudes Towards the Continuation Task
First, the two participants affirmed the effect of CT in promoting foreign/second language learning, similar to the findings of Wang’s
| [25] | Wang, C. (2015). Why does the continuation task facilitate L2 learning? Foreign Language Teaching and Research, 47(5), 753-762. |
[25]
study, and said that CT was more interesting than the other writing tasks they had engaged in and that it gave them more room for imagination and improved their motivation to write. Second, this study also found that CT forces language learners to go beyond their familiar content and linguistic expressions, expands their topic choices and vocabulary, and strengthens their language expression ability. In the follow-up interview, P1 said:
“My strongest feeling after the continuation task is that I should memorize more words. Anyway, I should at least consolidate the previous words. Otherwise, like in the tasks, I just wanted to use a word, but I couldn’t remember it even though I had thought of it for a long time, or I just misspelled it. More practice is needed in English...”
P2 mentioned:
“There is a high requirement for an individual’s comprehensive English ability in continuation task. Most of the continuation tasks are narratives, so we need to think about what proper words should be chosen, to think about making sentences, and how to express it in a coherent way to avoid inconsistency. And all these are flexible assessments of English writing ability.”
Third, Wang’s
| [25] | Wang, C. (2015). Why does the continuation task facilitate L2 learning? Foreign Language Teaching and Research, 47(5), 753-762. |
[25]
study outlined the fact that how familiar a writer is with the topic has a significant influence on their writing, with it being more difficult for them to write if the topic of the text is unfamiliar to them. Similarly, in the follow-up interview, P1 reported that it was difficult for her to finish CT B because the story took place in the forest and is related to wilderness survival, whereas she was not familiar with this topic and could not imagine the framework of that kind of space. Moreover, she did not know about any wilderness survival skills so could not think of how Jane’s husband would have saved her. Nevertheless, when compared with CT B, she was familiar with the topic of text A “
losing things/thievery” and thus felt much more confident when writing on this. Therefore, it becomes clear that, the more familiar students are with the topic and spatial framework of the story, the easier it is for them to write about it.
Fourth, we found that, if writers incorporate continuation writing in their comfort writing zone, they tend to express ideas using their own familiar vocabulary rather than imitating and applying the language structures of the reading text, wherein the IA effect is weakened. This result was reflected in the CT of P1. For example, in CT A, she reported the following:
(the black-haired girl was reading a book… “what are you reading?” Dannis asked… and what is the name of this book? How to be a thief? Haha, no, that is impossible. But I only remember one book’s English name, so I will write this book. The kite runner. This is easy to write.)
Furthermore, the think-aloud data indicated that, in P1’s reading and writing process, they showed a lack of attention paid to the specific objects within the original text (especially in CT B). In contrast, P2 reread the text numerous times in order to identify specific objects therein and, every time she read it, she would mark each phrase (e.g., yellow blouse and helicopter) and would then reproduce these objects in her own writing. Therefore, when compared with P2’ s writing works, despite the fact that P1’s writing has logical consistency with the original text, they tended to create new entities/details in their writing rather than make use of or echo the entities/details from the reading text; thus, her continuation writing works demonstrated a weaker alignment than P2’s at both the content and language levels.
Share This Article
Twitter
Linked In
Facebook
Copy
Download