Effects of manipulating writing task complexity on learners’ performance in completing vocabulary and syntactic tasks

Abstract

The present study investigated the potential of writing in English as a foreign language (EFL) for language learning by manipulating cognitive task complexity based on related models and hypotheses. English essays written by 59 Chinese postgraduate EFL students from different subject areas were analysed with reference to writing complexity, accuracy, and fluency (CAF). Results showed that task complexity had no significant effect on EFL learners’ lexical complexity but had an influence on their syntactic variation in EFL writing. Findings suggest that manipulating writing task complexity could be a feasible means to promoting and enhancing EFL learners’ language learning. Such findings might broaden our understanding of the relationship between EFL writing and language learning in an EFL learning context. The interplay of EFL writing and EFL learning is also pedagogically relevant to those who are interested in appropriately sequencing tasks for more effective language teaching.

Keywords

EFL learning EFL writing lexical learning syntactic learning task complexity

I Introduction

Writing is supposed to play a vital role in facilitating English as a second/foreign language learners’ linguistic, communicative and academic success due to the fact that writing per se requires the writer to show a higher degree of lexical, syntactic, and grammatical accuracy as well as logical coherence in expressing the intended meanings than what can be achieved in speaking. Johnson (2017) argues that writing requires ‘a more sustained focus on language and its form’ (p. 15). Considering that there is an interface between second language (L2) learning and L2 writing, scholars have generally recognized the potential role of writing as a productive skill in L2 development from a theoretical perspective (Bitchener & Ferris, 2012; Manchón, 2009, 2011a, 2011b; Xu et al. in press; Zhang, 2013). It is of great importance to broaden our understanding of the potential role of writing in language learning in EFL contexts for researchers, writing instructors, and even learners themselves.

In the last two decades, a number of L2 writing studies have explored the suitability of writing as a potential medium for promoting L2 learning through manipulating writing tasks (e.g. Frear & Bitchener, 2015; Johnson, Mercado & Acevedo, 2012; Kuiken & Vedder, 2007, 2008, 2012; Ong & Zhang, 2010, 2013; Rahimi & Zhang, 2018). Nevertheless, these studies have not produced consistent results, especially with regard to EFL learners’ writing performance in aspects such as writing complexity, linguistic accuracy, and writing fluency (Khezrlou, 2020). Also, few empirical studies have investigated EFL writing and language learning through the lens of reasoning-demand task complexity. This is because reasoning-demand task complexity, which is a significant index of the extent to which tasks require learners to provide justifications or explain causalities, is an important variable that affects learners’ writing performance in different ways (e.g. Rahimi, 2018). As such, the effect of cognitive task complexity, which naturally involves different degrees of reasoning demands, and task sequencing, on writing performance is important to our understanding of how language learners balance their cognitive resources to meet the demand of writing tasks (Kormos, 2011).

Considering the nexus between tasks and writing, Byrnes and Manchón (2014) foreground the possibility for broadening the understanding of language learning and teaching through the lens of writing, following what Hyland (2003) proposed. Hyland posited that designing and sequencing writing tasks is vitally important for writing curriculum design, as, in this process, due attention has to be paid to task characteristics, which could alter learners’ cognitive and linguistic demands. In order to understand the intricacy of such relationships, scholars have started research on the teaching of writing in the L2 classroom. Such studies have approached the issues from three perspectives. The learning-to-write approach concentrates on L2 learners learning how to write (Hyland, 2011). The writing-to-learn-content approach ‘investigates writing as a vehicle for the learning of disciplinary subject-matter in the content areas, mainly (although not solely) in ‘second’ instructional settings’ (Hirvela, Hyland, & Manchón, 2016, p. 45). The writing-to-learn-language perspective regards writing as a vehicle for language learning and development (Manchón, 2009; Manchón & Williams, 2016; Williams, 2012). Among a variety of teaching approaches, task-based language teaching (TBLT) employs authentic language and learners are required to do meaningful tasks through the use of the target language. Our search of the literature clearly indicates that much task-based research has focussed on L2 oral and written production in contexts where the target language is used as the medium of instruction and it is also the language of daily communication (see Ellis, 2009; Johnson, 2017), but relatively few studies have been reported in the available literature on how EFL learners would perform when writing in a foreign language in a typical foreign language context. In addition, most of the TBLT studies focus on L2 learners’ oral language performance instead of written output. Even with the few studies on EFL learners, their focus is also on oral language production. Such a situation is also true with regard to Chinese institutions of higher education, where TBLT was introduced in the 1980s. Understandably, task-based variables such as task complexity and task condition are vital factors that affect cognitive demands on EFL learners (Skehan, 1998, 2018), and there is a need to examine how they affect learners’ performance in writing. Unfortunately, it still remains unclear how cognitive task complexity alongside the reasoning demands play out in affecting EFL learners’ writing. This study was conducted to fill these lacunae.

II Review of the literature

1 Theoretical background

Task complexity is a pivotal construct, referring to ‘the result of the attentional, memory, reasoning, and other information processing demands imposed by the structure of the task to the language learner’ (Robinson, 2001b, p. 29). Task complexity, which plays a predominant role in TBLT, can affect the quality of writing production and facilitate L2 learning as shown in a number of studies (e.g. Ruiz-Funes, 2015; Yoon & Polio, 2017). Task-based variables are focal points in L2 writing research, but currently, the research concerning the potential of writing tasks in the field of EFL writing for EFL learning is still scarce. The influence of task complexity on L2 writing remains an issue, particularly for EFL learners who are most often exposed to English in classroom settings. EFL writing ability is an essential embodiment of learners’ language proficiency, so how to improve EFL learners’ writing is of great importance. Accordingly, we explored the issue by recruiting Chinese EFL student writers as the participants.

Two frameworks could be utilised to make predictions about the relationship between cognitive task complexity and students’ linguistic performance. They are Skehan and Foster’s (2001) Limited Attentional Capacity Model (also known as the Trade-Off Hypothesis) and Robinson’s (2001a, 2001b, 2005) Cognition Hypothesis (also known as the Triadic Componential Framework). The two models have hypothesised how task sequencing affects learning by imposing task-related variables on learners’ cognitive resources. Both models primarily concentrate on L2 oral production. However, because writing and speaking have similar attributes, as they are productive abilities triggering cognitive processes to convey precise and coherent messages, some researchers have already managed to operationalize these two models in L2 writing (e.g. Frear & Bitchener, 2015; Ishikawa, 2007). The potential of L2 writing for L2 learning is related to the degree of task complexity, as writing is a problem-solving process involving many factors, including working memory capacity (Cumming, 1990). Due to the problem-solving nature of writing, L2 writers resort to the function of working memory for coping with lexical and syntactic problems in the process of fulfilling writing tasks. As Manchón and Roca de Larios (2007) remark, L2 writing tasks elicit problem-solving behavior, as L2 writing propels learners to engage in meaningful and challenging L2 production. L2 learners devote their attention to different dimensions of written output. Working memory capacity plays a vital role in the process of writing when individuals retrieve cognitive resources from the central executive system (Kellogg, 1996). Writing elicits attentional resources to form meaning relations that enable learners to refine their language performance (Cumming, 1990).

As limited working memory indicates, the Limited Attentional Capacity Model is derived from working memory theory (Carter, 1998). Due to the limited information processing capacity, learners will give priority to one aspect of written production over the other aspects, when limited attentional resources are reached. There are trade-off effects between complexity, fluency, and accuracy. Specifically, increasing cognitive task complexity will facilitate fluency of language production, while the complexity and accuracy will be deteriorated. Tasks, as a series of well-organized teaching activities, are cognitively demanding. In this sense, more demanding writing tasks require more attentional resources from learners in the writing process (Rahimi & Zhang, 2019), and learners give priority to meaning or content rather than focus on linguistic form.

With an attempt to explore the effects of task features on language learning and development, Robinson, based on information processing theories (Schmidt, 2001), put forward his Cognition Hypothesis (Robinson, 2001a, 2001b, 2005), which postulates that learners focus on multiple aspects of linguistic forms and meaning, as individuals have multiple resources. The processing theory suggests learners can manipulate multiple resources. Robinson (2005, 2011) posits that increasing task complexity simultaneously results in better language performance. As a result, tasks should be designed and sequenced in accordance with task features, in order to facilitate language learning and development (Robinson, 2005, 2011). Robinson (2005) categorised task complexity into resource-directing and resource-depleting dimensions. The resource-directing dimension is where task complexity makes conceptual demands on the learner, such as having +/–few elements, +/–here-and-now and +/–no reasoning demands; the resource-dispersing dimension depicts that task complexity poses procedural demands including +/–planning, +/–single task, and +/–prior knowledge. The two dimensions of task complexity impose influence on individuals’ attentional resources, and thus manipulating the cognitive complexity of a writing task along different dimensions will lead to varying writing performance. Increasing task complexity along the resource-directing dimension leads to the improvement of complexity and accuracy as learners meet the demands of complex tasks, while the fluency of language production will decrease (Robinson, 2011). Learners seem to allocate their attentional resources to meet the demands of writing tasks. Therefore, the importance of task sequence and implementation calls for further research to validate the two hypotheses, and this is particularly important for an EFL learning context.

2 Writing task complexity and L2 learning

The theoretical research on the potential of writing for L2 learning has been burgeoning in recent years (e.g. Byrnes & Manchón, 2014; Ortega, 2012; Williams, 2012; Zhang, 2013). Writing is believed to be able to promote L2 learning and development due to its problem-solving nature. In addition, the features of writing such as permanence record (William, 2012) and cyclical nature (Kormos, 2014), draw language learners’ attention to linguistic form and content. Therefore, comprehensible output, for example, writing, can be a suitable medium to improve learners’ language learning and development (Swain, 1985, 1995; Williams, 2012).

A variety of writing tasks were manipulated in the field of task-based research including letter writing (e.g. Frear & Bitchener, 2015; Kuiken & Vedder, 2007, 2008, 2011, 2012), narration (e.g. Kormos, 2011, 2014; Way, Joiner, & Seaman, 2000) and essay writing (e.g. Ong, 2014; Ong & Zhang, 2010, 2013; Ruiz-Funes, 2015). However, those studies seem to have yielded different results pertaining to the effect of task variables on L2 writing and learning. In view of the interplay between writing task complexity and L2 learning, most of the research has investigated cognitive task complexity on resource-directing dimension involving +/–few elements, +/–here-and-now (e.g. Kuiken & Vedder, 2007, 2008, 2011; Xu et al., in press). Few studies have concentrated task complexity along resource-directing dimension associated with +/–reasoning demands. Varying writing task types and measures of linguistic production were employed in previous research, and the results seem to lend partial support to the two frameworks with regard to different language proficiency learners. In the early study, Ishikawa (2007) found that manipulation of task complexity along the resource-directing dimension could increase the complexity, accuracy, and fluency of L2 writing production, which is partially in accordance with Cognition Hypothesis. On the contrary, some researchers maintained that increasing cognitive task complexity along the resource-directing dimension affected one aspect of writing production, which lends support to the Limited Attentional Capacity Model. For example, Kormos’ (2011) research showed that task complexity had a significant effect on the lexical complexity in L2 writing performance. Frear and Bitchener (2015) partially replicated Kuiken and Vedder’s (2012) study and found that task complexity only exerted influence on syntactic complexity. These findings indicate that learners’ writing performance was affected by cognitive task complexity, regardless of their language proficiency level. Rahimi and Zhang (2019) explored the effects of increasing the number of elements and the degree of reasoning demands on L2 writing. They found that there was a significant positive effect on syntactic complexity and a negative influence on accuracy. No consensus, however, has been reached due to the variety of linguistic complexity measures and task-based variables. Hence, the importance of multiple metrics cannot be ignored. To fill these lacunae, further research needs to investigate the potential of cognitive task complexity concerning reasoning demands on learners in language learning, especially in an EFL learning context such as China.

Writing is ‘highly variable and contingent on education, opportunities for learning, and needs for use’ (Cumming, 2013, p. 1). In the case of EFL writing, it is also very important; it is particularly so for postgraduate students, who need to write their reports for sharing their research with the academic community in English. Thus, many efforts have been made to improve students’ EFL writing competence using a variety of approaches, including task-based language teaching. Research on task variables in L2 writing and L2 learning has been brought to the fore (Skehan, 1998; Robinson, 2001a). The aforementioned research, however, did not pay enough attention to the effect of the level of reasoning demands of the writing tasks on EFL writing performance. Additionally, EFL learners’ English proficiency, especially lower-intermediate learners’ proficiency, as an important variable, has been overlooked in those studies on writing task complexity. In our study, task complexity refers to reasoning processing demands (+reasoning demands) imposed by the task along the resource-directing dimension. When EFL learners accomplish a writing task, cognitive resources will be allocated to different aspects of their written production involving language meaning and form. We set out to explore the effects of EFL learners’ writing on their English vocabulary and syntactic learning by addressing two research questions as follows:

How does EFL writing task complexity affect EFL learners’ lexical performance?

How does EFL writing task complexity affect EFL learners’ syntactic performance?

III Method

1 Participants

A total of 72 full-time postgraduate students as a convenience sample, aged between 21 and 25 (M = 23), were recruited from two parallel classes of an English writing course. Our decision to use convenience sampling as an appropriate strategy for data collection was based on an understanding that it is arguably the most useful way to collect quantitative data, as expounded by many leading researchers in the fields of education and applied linguistics (e.g. Creswell & Creswell, 2018; Dörnyei, 2007; Leown & Godfroid, 2020; Glass & Hopkins, 1996). Given that we followed the principle of non-coercive and voluntary response, participation was purely voluntary. We invited all the 123 students in two classes to participate in the study, but in the end 72 students completed the two compositions.

The participants majored in different disciplinary areas in a national university in Northeast China. We were interested in finding out how vocabulary and syntactic learning occurred in these postgraduates in relation to task complexity, with a clear understanding that they were being trained to be scientists, scholars or leaders in other disciplinary areas in the future and both their English proficiency and their English writing competence would play an important role in their future careers. Mandarin Chinese was these participants’ mother tongue. They had learned EFL for over 10 years, but none of them had any study abroad experiences. According to the placement test scores, all participants were at the lower-intermediate level of English proficiency, and, as required by the national syllabus for universities, they had to take a compulsory EFL writing course for satisfying one of the requirements for graduation.

2 Writing tasks

A narrative writing task and an argumentative writing task adapted from the national College English Tests (CET) bank were adopted in this study (see Appendix 1). CET is a nationally standardised English language proficiency test for non-English major university students in China. The validity and reliability of CET have been well documented. Students were familiar with topics and genres of the given task types as those were often found in English tests in China, such as the CET-Band 4 and CET-Band 6. Writing tasks entailing more reasoning demands are considered more cognitively complex in the field of L2 writing research (e.g. Rahimi, 2018). Given that research into the role of increasing cognitive task complexity along the resource-directing dimension (reasoning demands) in EFL learning contexts has not been given due attention, the two writing tasks were set up to investigate the relationship.

As briefly mentioned above, the reasons for choosing these two tasks were that the tasks of cognitive task complexity are said to be related to reasoning demands and genres (Ruiz-Funes, 2015). Although the theme of the two writing tasks was similar, the level of complexity was different due to varying degrees of reasoning demands. In general, research has already shown that an argumentative writing task needs more reasoning demands than a narrative task (see Polio, 2017). In order to increase the reliability of this study, 10 experienced English writing teachers were invited to judge the complexity of these two writing tasks. Each of them was invited to evaluate the complexity of the two writing tasks by rating them on a scale ranging from 10 to100 on the basis of their teaching experiences. The higher the score was, the more difficult the task was. Task 1 (narrative writing task) involved less reasoning demands; Task 2 (argumentative writing task) was a more complex task and it required students to expound their points of view, involving more reasoning demands. These two task types were adopted because they are typically used in English tests for EFL learners.

3 Data collection

A total of 72 students were invited to complete two writing tasks in two genres (argumentation and narration) within 60 minutes (each essay for 30 minutes) in a classroom setting, and they were not allowed to consult dictionaries, nor any other resources. Each participant was assigned two writing tasks with different levels of cognitive task complexity. The writing tasks were counterbalanced to reduce practice and fatigue effects, as was the case in a published study (Rahimi & Zhang, 2018).

4 Data analysis

a Measures of complexity, accuracy, and fluency (CAF)

Linguistic complexity depicts the extent to which language is elaborate and various (Lu, 2011; Norris & Ortega, 2009; Rahimi, 2018). Writing performance can be comprehensively captured by the triad of complexity, accuracy, and fluency constructs (Norris & Ortega, 2009; Skehan, 1998). However, the tripartite view of measuring writing performance in the field of L2 writing research has not reached any consensus. There is no acceptance criterion about CAF measures (e.g. Larsen-Freeman, 2009). Researchers have adopted different measures to evaluate learners’ linguistic performance; for instance, the complexity construct is polysemous and it is based on different arguments (Pallotti, 2015). Bearing in mind such a situation, we modified comprehensive metrics to evaluate linguistic performance because of the EFL learning context in which the study was located. The indices adopted to measure syntactic complexity were based on Lu’s (2010) L2 Syntactic Complexity Analyser by virtue of its sensitivity to linguistic complexity (also see Jin, Lu & Ni, 2020). Despite the fact that most of the researchers have measured syntactic complexity by adopting indices such as the number of T-units, several scholars (e.g. Kuiken & Vedder, 2012, 2017; Ruiz-Funes, 2015) argue that they may not be very useful to studies conducted in EFL contexts. A T-unit means ‘one main clause plus any subordinate clause or non-clausal structure that is attached to or embedded in it’ (Hunt, 1970, p. 4). It can be discerned that T-unit is not a suitable measure when we want to present a holistic picture of EFL learners’ syntactic complexity, as the construct of syntactic complexity is multidimensional. Wolfe-Quintero, Inagaki, and Kim (1998) have proposed that passives, articles, relative clauses and complex nominals are also indicative of linguistic performance. Moreover, the importance of multiple metrics of syntactic complexity has been underscored when evaluating L2 learners’ writing performance and writing development (Bulté & Housen, 2012; Norris & Ortega, 2009). Because of these reasons, we adopted comprehensive measures of syntactic complexity with the aim of investigating the potential role of task complexity in vocabulary and syntactic learning in EFL writing. We adopted Wolfe-Quintero et al.’s (1998) indices to measure lexical complexity, accuracy, and fluency in students’ EFL texts. Table 1 presents an overview of CAF measures adopted in this study.

Table 1.

Complexity, accuracy, and fluency (CAF) measures in the present study.

Construct		Code	Full form
Complexity	Lexical complexity	WT√2 W	The ratio of different word types to the total number of words
Complexity	Syntactic complexity	MLC; MLS; MLT; C/T; DC/C; DC/T; CP/C; CP/T; T/S; CN/C; CN/T; VP/T	mean length of clause; mean length of sentence; mean length of T-unit; clause per T-unit; dependent clauses per clause; dependent clauses per T-unit; coordinate phrases per clause; coordinate phrases per T-unit; T-units per sentence; complex nominals per clause; complex nominals per T-unit; verb phrases per T-unit
Accuracy		EFC/C	error-free clause ratios
Fluency		W/T	words per T-unit

By referring to existing studies, we adopted the index of lexical complexity, corrected type-token ratio (CTTR), i.e. WT√2 W, which means the ratio of different word types to the total number of words. L2 Lexical Complexity Analyser (Lu, 2012) was employed to analyse lexical complexity of English writing production automatically. Lexical complexity reflects the variety of lexical forms deployed by a speaker or writer (Wendel, 1997). The higher the ratio is, the more varied are the words that learners use in their essays. The traditional measure of lexical complexity, i.e. type-token ratio (TTR), is sensitive to sample size (Arnaud, 1992), which we did not think was suitable for measuring EFL learners’ writing production. Therefore, CTTR was employed to minimize the sample size effect in this study.

The L2 Syntactic Complexity Analyser (Lu, 2010) was used to robotize comprehensive syntactic complexity analysis of these collected essays. To ascertain syntactic complexity in the written texts, 12 indices were adopted to measure syntactic complexity of students’ essays. C/S and complex T-units per T-unit were not used, as research has shown that they are less effective in measuring syntactic complexity (e.g. Lu, 2011; Yoon & Polio, 2017). The L2 indices included four dimensions: Length of unit, subordination, coordination, and particular structures. Specifically, the length of unit includes mean length of clause (MLC), mean length of sentence (MLS) and mean length of T-unit (MLT). Subordination contains clause per T-unit (C/T), dependent clauses per clause (DC/C) and dependent clauses per T-unit (DC/T). Coordination involves coordinate phrases per clause (CP/C), coordinate phrases per T-unit (CN/T) and T-units per sentence (T/S). Particular structures can be categorised into complex nominals per clause (CN/C), complex nominals per T-unit (CN/T), and verb phrases per T-unit (VP/T). These 12 indices can depict a more comprehensive picture of EFL learners’ syntactic complexity performance.

Error-free clause ratios (EFC/C) were employed to evaluate the accuracy of students’ written production based on Wolfe-Quintero et al.’s (1998) proposal that EFC/C is the best measure of accuracy for intermediate level learners, as it measures the percentage of error-free clauses per clause; while analysing error-free T-unit is more appropriate for capturing advanced learners’ errors. Error-free clauses are defined as clauses that have no errors in syntax, morphology, lexical choice or word order (Tavakoli & Skehan, 2005). Errors in capitalisation, spelling, and punctuation were not taken into account in our study. Accuracy is about grammatical and lexical correctness in students’ writing due to their effort to allocate attentional resources. The higher the ratio is, the more accurate their writing production is.

Fluency was measured by calculating the mean number of words per T-unit (W/T) in the present study, which means that fluency is an outcome of the total number of words divided by the total number of T-unit. The better the result is, the more fluent the writing production is. Wolfe-Quintero et al. (1998) posit that this measure is one of the most reliable measures for evaluating fluency. This is because fluency demonstrates that learners can access more words and structures in a limited time, whereas a lack of fluency means that they can access only a few words or structures (Wolfe-Quintero et al., 1998). The inter-rater reliability coefficients were above .85 for accuracy and fluency measures.

b Statistical analysis

In an attempt to find out the effects of these two writing tasks on EFL learners’ vocabulary and syntactic performance, we analysed 59 essays of each task and 118 essays in total, while 13 invalid essays were not included because students did not fulfill the word number requirement, or they digressed from the given topic. Paired-samples t-tests were conducted respectively to assess the lexical complexity, syntactic complexity, accuracy, and fluency in these essays. To measure the effect size, Cohen’s d was adopted. The inter-rater reliability was also calculated to ensure the reliability of manual coding measures.

IV Results

1 Effects on learners’ lexical complexity

The mean value for Task 1 was 5.02 and for Task 2 was 4.90, showing no statistically significant difference between cognitive task complexity and lexical complexity (see Table 2). The effect size, calculated using Cohen’s d, was .28, which is considered small. The findings showed that the complex task did not lead to EFL learners’ using complex vocabulary. In other words, task complexity had no statistically significant influence on learners’ lexical complexity in English essays.

Table 2.

Comparison of lexical complexity in Task 1 and Task 2.

	Task 1		Task 2		MD	p
	M	SD	M	SD	MD	p
WT√2 W	5.02	.39	4.90	.46	.12	.08

Notes. WT√2 W = The ratio of different word types to the total number of words. *p < .05.

2 Effects on learners’ syntactic complexity

Regarding the length of unit, the mean values of three indices, MLS, MLT and MLC in Task 1, were a bit lower than those in Task 2 (see Table 3), which meant that the more complex task elicited longer unit in the writing production. The results of paired-samples t-tests revealed a significant distinction between the two writing tasks as far as the length of unit was concerned (MLS, p = .04, d = .27; MLT, p = .00, d = .55; MLC, p = .00, d = .55). As a result, learners employed longer clauses in the complex writing task than in the simple writing task. The length of unit in the two writing tasks reached a significant level, and three indices of the length of unit presented an upward trend as task complexity increased.

Table 3.

Comparison of length of unit in Task 1 and Task 2.

	Task 1		Task 2		MD	p
	M	SD	M	SD	MD	p
MLS	14.02	3.79	15.01	3.52	-.99	.04
MLT	12.43	2.46	13.87	2.79	-1.44	.00
MLC	8.83	1.57	9.87	2.16	-1.04	.00

Notes. MLS = mean length of sentence. MLC = mean length of clause. MLT = mean length of T-unit. *p < .05.

As shown in Tables 4 and 5, increasing task complexity along the resource-directing dimension did not have significant influence on EFL learners’ subordination and coordination in writing performance. With regard to subordination, the mean values of C/T and DC/T were larger than those of the mean values in Task 2 (C/T, d = .11; DC/T, d = .06). As for DC/C, task complexity affected EFL learners’ subordination production (DC/C, d = .11). The mean value decreased when the task complexity increased. The mean values of CP/T were equal in two writing tasks, and CP/C presented an upward trend (CP/T, d = .00; CP/C, d = .07). Regarding coordination (see Table 5), students generally performed better in the complex writing task. Additionally, the mean value of T/S in Task 1 was a little larger than that of Task 2 (T/S, d = .30). Nevertheless, the difference concerning subordination as well as coordination between two different tasks was not significant.

Table 4.

Comparison of subordination in Task 1 and Task 2.

	Task 1		Task 2		MD	p
	M	SD	M	SD	MD	p
C/T	1.43	.24	1.46	.34	-.03	.57
DC/C	.29	.18	.28	.18	.01	.69
DC/T	.39	.18	.40	.29	-.01	.86

Notes. C/T = clause per T-unit. DC/C = dependent clauses per clause. DC/T = dependent clauses per T-unit. *p < .05.

Table 5.

Comparison of coordination in Task 1 and Task 2.

	Task 1		Task 2		MD	p
	M	SD	M	SD	MD	p
T/S	1.12	.15	1.08	.11	.04	.06
CP/T	.32	.17	.32	.21	.00	.97
CP/C	.22	.13	.23	.17	-.01	.76

Notes. T/S = T-units per sentence. CP/T = coordinate phrases per T-unit. CP/C = coordinate phrases per clause. *p < .05.

As Table 6 shows, all three indices of particular structures, namely, VP/T, CN/T, and CN/C displayed an upward trend. The mean values of VP/T, CN/T, and CN/C in the argumentative writing task were larger than those of the narrative writing task. Only CN/T and CN/C were significantly different due to manipulating different cognitive task complexity (CN/T, p = .00; d = .70; CN/C, p = .00, d = .67), compared with VP/T, which was non-significant (p = .22, d = .07). Task complexity had a significant influence on learners’ complex nominals use per clause and per T-unit. In order to perform the complex writing task better, EFL students generated more complex nominals in the sentences to express their views accurately and clearly. This is quite natural, because a complex nominal, which is an eligible index to exemplify the particular structures in EFL learners’ writing production, is usually useful when writers need to clarify views. However, task complexity did not affect verb phrases per T-unit significantly, and the number of particular structures in Task 2 was larger than that in Task 1.

Table 6.

Comparison of particular structures in Task 1 and Task 2.

	Task 1		Task 2		MD	p
	M	SD	M	SD	MD	p
VP/T	1.95	3.37	2.12	1.05	-.17	.22
CN/T	1.35	.39	1.66	.49	-.31	.00
CN/C	.96	.25	1.17	.37	-.21	.00

Notes. VP/T = verb phrases per T-unit. CN/T = complex nominals per T-unit. CN/C = complex nominals per clause. *p < .05.

3 Effects on learners’ writing accuracy

Task complexity had no significant effect on learners’ writing accuracy. As Table 7 shows, the mean value of EFC/C is .82 in the simple writing task and .80 in the complex task (d = .17). The result yielded no evidence of a significant difference between task complexity and accuracy, although accuracy was affected positively by increasing the cognitive demands of task complexity.

Table 7.

Comparison of accuracy in Task 1 and Task 2.

	Task 1		Task 2		MD	p
	M	SD	M	SD	MD	p
EFC/C	.82	.10	.80	.13	.02	.57

Notes. EFC/C = error-free clause ratios. *p < .05.

4 Effects on learners’ writing fluency

The mean value of W/T was 12.34 words in Task 1 while the mean value about fluency was 13.87 words in Task 2, and task complexity had influence on the fluency of learners’ writing production, as the number of words increased when the task complexity increased (p = .00, d = .57). Increasing task complexity had a significant adverse impact on learners’ writing fluency. Table 8 displays descriptive statistics of writing fluency in Task 1 and Task 2.

Table 8.

Comparison of fluency in Task 1 and Task 2.

	Task 1		Task 2		MD	p
	M	SD	M	SD	MD	p
W/T	12.34	2.57	13.87	2.79	-1.53	.00

Notes. W/T = words per T-unit. *p < .05.

V Discussion

1 Effects of task complexity on EFL writing lexical complexity

Increasing cognitive task complexity did not influence lexical complexity statistically significantly. The finding supports the predications of the Limited Attentional Capacity Model, though it is inconsistent with Cognition Hypothesis. Such a result might be due to the manipulation of task complexity along the resource-directing dimension, which could have influenced form and meaning in language production in spite of the individual’s multi-dimensional attentional resources. EFL learners may acquire more vocabulary in a more complex task, as suggested by the hypothesis. However, the lower-intermediate level EFL participants generated rather complex vocabulary when fulfilling the simple writing task but not a complex task. The word types decreased in number with increasing task complexity along the resource-directing dimension (+/–reasoning demands). This result is in contrast with Kormos’ (2011) study, and such a result, we think, is due to different task types used in our study and that of Kormos’.

Generally speaking, more complex writing tasks trigger students’ use of more complex vocabulary because simple words cannot fulfill the complex task and that these advanced or high-intermediate level EFL learners are in good command of English vocabulary and the complex writing task afforded then an opportunity to try these words in writing. Nonetheless, it was found that lower-intermediate level EFL learners used passive voice rather than active voice. It appears that increasing writing task complexity might have made them feel intimidated so much that they were not confident about those complex words, of which they only knew the meaning but were not adept at using them in writing. The finding implies that EFL writing instructors should be aware that complex writing task is not of much help for low-intermediate level learners to improve their lexical learning. The instructors and curriculum designers are supposed to design writing tasks by taking the learners’ English proficiency into consideration. Increasing the complexity of writing tasks may be a double-edged sword. If the task works well, it is beneficial for EFL learners; otherwise, it may frustrate them. A complex writing task may not help much in lexical development for lower-intermediate EFL learners.

2 Effects of task complexity on EFL writing syntactic complexity

The results concerning the length of unit are consistent with some previous studies (e.g. Ishikawa, 2007; Ruiz-Funes, 2015). These findings also lend support to Cognition Hypothesis. As task complexity increased, low-intermediate EFL learners produced longer sentences to express and clarify their views. Results also indicated that cognitive task complexity had no significant effect on learners’ writing production concerning subordination and coordination in spite of some insignificant changes. These findings partially support some previous studies regarding subordination (e.g. Ellis & Yuan, 2004; Fear & Bitchener, 2015; Rahimi, 2018). In order to perform the complex writing task better, EFL learners generated more complex nominals in the sentences to express their views accurately and clearly. This is quite natural, as complex nominal, which is an eligible index to exemplify the particular structures in EFL learners’ writing production, is usually useful when writers need to clarify views. However, task complexity did not affect verb phrases per T-unit significantly, and the number of particular structures in Task 2 was larger than that in Task 1.

In line with findings of some previous studies (e.g. Rahimi, 2018; Rahimi & Zhang, 2018, 2019; Ruiz-Funes, 2015; Xu et al. in press), the present study also revealed that increasing task complexity generally elicited more complex syntactic use. But we found task complexity had a significant effect on syntactic complexity to some extent, which differed from Ruiz-Funes’ (2015) study, where task complexity was not found to have any significant impact on learners’ syntactic complexity. The reason for this difference is that in our study we adopted multiple measures for assessing syntactic complexity, which also explains why increasing task complexity affected students’ writing output in the case of syntactic development. Specifically, the length of unit, complex nominals per T-unit and verb phrases per T-unit structures were positively affected by increasing task complexity. The students generated more of those particular structures and the length of unit also increased in the more complex writing task. The results did not yield significant differences between task complexity and syntactic complexity in the case of subordination and coordination use. Such a finding perhaps might be due to the fact that attentional resources are limited for these lower-intermediate level EFL learners. It is not an easy job for them to retrieve and extract their prior knowledge when writing in a foreign language, a skill that is well-recognised as the most challenging among all the language skills.

In general, syntactic complexity in terms of the length of unit and particular structures was affected positively by means of increasing task complexity, and the overall results regarding syntactic complexity partially lend support to the two models (Skehan & Foster, 1999; Robinson, 2011). Moreover, these findings confirm that tasks have a great effect on language production (Khezrlou, 2020; Ong & Zhang, 2013), and multidimensional syntactic complexity metrics should be adopted to capture L2 writing development (Norris & Ortega, 2009). Nonetheless, the above findings do not corroborate the results of some previous studies (e.g. Frear & Bitchener, 2015), perhaps due to the different metrics of syntactic complexity used in these studies. We adopted Lu’s (2010) measures, which is more comprehensive than those used in some previous studies (e.g. Frear & Bitchener, 2015; Norris & Ortega, 2009). Normally, in these earlier studies, subordination was regarded as an indication of syntactic complexity, but it may not capture complex features of syntactic complexity and therefore might not be suitable for measuring EFL learners’ writing performance. In order to present a holistic picture of the findings reported above, we summarise the effects of increasing task complexity on syntactic complexity in Figure 1.

Figure 1.

Summary of effects of increasing task complexity on syntactic complexity.

These findings imply that the instructors should design the writing tasks according to their teaching objectives. If their focus is to enhance the learners’ ability to use more complex sentences or sentence structures, they may design more complex writing tasks. The students’ academic background may also need to be taken into consideration, as previous research has revealed that students from humanities and social sciences tend to use more complex sentence structures while students from science and engineering prefer using simple sentence structures in their writing (e.g. Jiang, 2016). It is worth further exploring the issue.

3 Effects of task complexity on EFL writing accuracy

It appeared to be true that more complex tasks would induce more errors in writing, as reported in Ruiz-Funes (2015). Our finding runs counter to those from some previous studies (e.g. Ishikawa, 2007; Kuiken & Vedder, 2011), but corroborates the Limited Attentional Capacity Model, as accuracy in writing increased in the complex task. Although the mean value of accuracy decreased when task complexity increased, the difference between task complexity and accuracy was not statistically significant. Learners apparently paid attention to linguistic forms with a loss of attentional resources on content in their writing. If they decided to give priority to the content of their writing, they would necessarily devote their cognitive resources to the content, and, therefore, errors in language form would become evident.

4 Effects of task complexity on EFL writing fluency

The difference in fluency was statistically significant between the two writing tasks with different cognitive task complexity; that is, EFL learners’ writing was more fluent in the complex writing task. The result contradicts some existing research (e.g. Rahimi & Zhang, 2018; Ruiz-Funes, 2015) that increasing task complexity had a significant adverse influence on learners’ writing fluency. This might be due to participants’ English proficiency and the different operationalisations of fluency measures in these studies. This finding is in line with Cognition Hypothesis. It might be assumed that increasing cognitive writing task complexity contributes to writing fluency more than accuracy. Some prior studies have also produced similar results (e.g. Ong & Zhang, 2010). The finding shows that more complex task triggers EFL learners to write more fluently. It seems to be in contrast with the previous research result that the fluency of writing production reduces, while the complexity and accuracy increase simultaneously (Robinson, 2001b). Perhaps the different result might be explained by a related study of L2 development (e.g. Housen & Pierrard, 2005), which categorised sequential macro-processes, namely, knowledge internalisation, knowledge modification, and knowledge consolidation. More complex writing task, like Task 2, needs learners’ higher reasoning demands, which enables them to improve their syntactic use, but cannot focus on accuracy at the same time due to the influence of limited attentional resources. Accordingly, increasing the cognitive complexity of writing tasks may influence the writing output and potentially lead to language development.

5 Effects of cognitive task complexity on EFL learning

In light of the major findings mentioned above, increasing task complexity can result in a trade-off effect between linguistic form and language content in writing for lower-intermediate English learners, who may focus on form at the expense of content, because individuals have limited attentional capacity on account of working memory capacity. Notably, there exists a trade-off effect on form and content manifested as the interplay of the CAF triad in language production (Skehan, 1998). These findings of this study corroborate Skehan and Foster’s (2001) conviction that increasing cognitive task complexity along the resource-directing dimension generally results in a trade-off effect in linguistic performance, as is stated in their Limited Attentional Capacity Model. There is a tension between complexity and accuracy (Skehan & Foster, 2001). Specifically, learners possess a limited processing capacity due to single and limited resources, and thus manipulation of task complexity affects writing production in terms of CAF.

These findings imply that writing is a feasible means to facilitate language learning in the EFL learning context, which is related to reasoning demands of task complexity. As task demands along the resource-directing dimension increase in the writing task, there exists a trade-off effect among the three aspects (complexity, accuracy, and fluency) of learners’ writing production. Tasks, in accordance with their structure that requires a great deal of information processes, can facilitate learners’ diverse reasoning demands. Given that Skehan and Foster (1997) proposed that individuals’ processing capacity is limited, it might be the case that EFL learners with lower-intermediate proficiency allocated their limited attentional resources to one aspect of language production at the expense of the other two aspects of the writing process.

As for the problem-solving nature of writing, task complexity can elicit learners to use cognitive processes in language acquisition (Cumming, 1990). Task complexity along the resource-directing dimension (+/–reasoning demands) successfully reduced the cognitive load of writing tasks on learners’ working memory resources. Moreover, in order to perform a writing task, EFL writers have to synchronize the cognitive process, which demands the crucial components of working memory. A writing task involving the coordination of cognitive demands and the retrieval of the writer’s linguistic and genre knowledge can overload the processing capacity of learners’ working memory (Ong & Zhang, 2010). Therefore, learners gave priority to complexity and fluency in their writing. As far as lower-intermediate Chinese EFL learners are concerned, this study depicts a general picture of learners’ writing output with reference to the CAF triad. Linguistic complexity and accuracy may not be acquired at the same time. Specifically, lower-intermediate EFL learners wrote more complex sentences at the cost of accuracy in their writing.

VI Conclusions

The present study analysed the essays written by EFL learners to investigate the relationship between EFL writing and task-based language learning, that is, how cognitive task complexity affected language learning in an EFL learning context. Our findings show that increasing reasoning demands of cognitive task complexity generally improved EFL learners’ writing complexity and fluency, and these findings, in a way, help us to have a better understanding of the effects of increasing reasoning demands of cognitive task complexity on EFL learners’ writing and language learning. The understanding of the interplay between EFL writing and language learning provides new insights that writing is a feasible means to promoting and enhancing language learning for lower-intermediate language learners in EFL contexts. These findings might have theoretical, methodological, and pedagogical implications.

Theoretically, the results of the present study are partially supportive of Skehan and Foster’s (2001) Limited Attentional Capacity Model, which corroborates that individuals’ attentional resources are limited and there exists a trade-off effect among complexity, accuracy, and fluency. This study enriches the theories of EFL writing and EFL learning. Increasing cognitive task complexity promotes syntactic complexity and fluency of EFL writing. As far as lower-intermediate EFL learners are concerned, this study depicts a general picture of learners’ writing output regarding complexity, accuracy, and fluency. The competition between complexity and accuracy is evident. Specifically, EFL learners write more complexly at the cost of accuracy in their writing. Methodologically, this study highlights the significance of adopting multidimensional operationalisation of syntactic complexity to capture EFL learners’ linguistic complexity so that the results are more comprehensive and accurate.

Pedagogically, EFL instructors and syllabus designers might need to take task complexity into account when designing syllabuses, assigning writing tasks and creating writing tests. Tasks should be designed and sequenced in light of task characteristics (Robinson, 2005). To improve students’ writing ability, writing tasks are arranged properly from simple to complex in accordance with task complexity. It is recommended that the task sequence in assigning writing tasks should be considered carefully. Cognitively demanding tasks are likely to result in trade-off effects between form and content, and it is difficult for EFL learners to attend to form and content simultaneously. More cognitive demands might overweigh learners’ cognitive load, and in some cases, avert learners’ limited attentional resources to fluency. Instructors, therefore, might need to take into consideration their students’ English proficiency, cognitive and emotional development as well as the effects of the cognitive demand that writing tasks impose on their students.

The present study also sheds further light on the important role of EFL writing in EFL learning. In the process of writing, a cognitively complex task directs learners’ attentional resources to fluency and syntactic complexity rather than accuracy and lexical complexity. Syntactic complexity and fluency are positively affected. Yet, task complexity affects learners’ lexical complexity and accuracy negatively. For lower-intermediate EFL learners, trade-off effects among CAF measures seem to be prominent. According to Swain’s (1985) Output Hypothesis, pushed output (writing) can play a positive role in L2 acquisition, which means pushed output may facilitate L2 development. For instance, syntactic complexity induced by task complexity enables L2 learners to engage in the process of L2 learning.

The limitation of this study lies in the small sample size of participants, which cannot represent all EFL learners’ writing performance. Due to participants’ low-intermediate English proficiency in our study, the results of this study cannot be generalized to other English learners with different language proficiency. Furthermore, this study examined the effects of task complexity along the resource-directing dimension, but the interaction of the resource-directing dimension and the resources-dispersing dimension was not given attention. It can be speculated that these two dimensions may influence the writing production in interaction, and, therefore, further research is needed to take into consideration other dimensions of task complexity in relation to the interaction of task complexity and English proficiency.

Footnotes

Appendix 1 Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was supported by a grant from Jilin University’s Social Sciences Citation Index (SSCI) Journal Publication Fund (JLU2014) and a New Zealand China Doctoral Research Scholarship (PLC3478).

ORCID iDs

Ju Zhan

Lawrence Jun Zhang

References

Arnaud

P.J.

(1992). Objective lexical and grammatical characteristics of L2 written compositions and the validity of separate-component tests. In Arnaud

P.J.L.

Béjoint

(Eds.), Vocabulary and applied linguistics (pp. 133–145). London: Palgrave Macmillan.

Bitchener

Ferris

D.R.

(2012). Written corrective feedback in second language acquisition and writing. New York: Routledge.

Bulté

Housen

(2012). Defining and operationalising L2 complexity. Philadelphia, PA: Benjamins.

Byrnes

Manchón

(2014). In Byrnes

Manchón

(Eds.), Task-based language learning: Insights from and for L2 writing (pp. 1–23). Philadelphia, PA: John Benjamins.

Carter

(1998). Mapping the mind. Berkeley, CA: University of California Press.

Creswell

J.W.

Creswell

J.D.

(2018). Research design: Qualitative, quantitative, and mixed methods approaches. 5th edition. Thousand Oaks, CA: Sage Publications.

Cumming

(1990). Metalinguistic and ideational thinking in second language composing. Written Communication, 7, 482–511.

Cumming

(2013). Writing development in second language acquisition. In Chapelle

(Ed.), The encyclopedia of applied linguistics. Malden, MA: Wiley-Blackwell.

Dörnyei

(2007). Research methods in applied linguistics: Quantitative, qualitative, and mixed methodologies. Oxford, England: Oxford University Press.

10.

Ellis

(2009). The differential effects of three types of task planning on the fluency, complexity, and accuracy in L2 oral production. Applied linguistics, 30, 474–509.

11.

Ellis

Yuan

(2004). The effects of planning on fluency, complexity, and accuracy in second language narrative writing. Studies in Second Language Acquisition, 26, 59–84.

12.

Frear

M.W.

Bitchener

(2015). The effects of cognitive task complexity on writing complexity. Journal of Second Language Writing, 30, 45–57.

13.

Glass

Hopkins

(1996). Statistical methods in education and psychology. 3rd edition. Boston, MA: Allyn and Bacon.

14.

Hirvela

Hyland

Manchón

R.M.

(2016). Dimensions in L2 writing theory and research: Learning to write and writing to learn. In Manchón

R.M.

Matsuda

P.K.

(Eds.), Handbook of second and foreign language writing (pp. 45–64). Boston, MA / Berlin: De Gruyter Mouton.

15.

Housen

Pierrard

(2005). Instructed second language acquisition: Introduction. In Housen

Pierrard

(Eds.), Investigations in instructed second language acquisition (pp. 1–27). Berlin: de Gruyter.

16.

Hunt

K.W.

(1970). Syntactic maturity in schoolchildren and adults. Monographs of the Society for Research in Child Development, 35(1).

17.

Hyland

(2003). Second language writing. Cambridge: Cambridge University Press.

18.

Hyland

(2011). Learning to write: Issues in theory, research and pedagogy. In Manchón

R.M.

(Ed.), Learning-to-write and writing-to-learn in an additional language (pp. 17–35). Philadelphia, PA: Benjamins.

19.

Ishikawa

(2007). The effect of manipulating task complexity along the [+/−Here-and-Now] dimension on L2 written narrative discourse. In Garcia Mayo

M.P.

(Ed.), Investigating tasks in formal language learning (pp. 136–156). Clevedon: Multilingual Matters.

20.

Jiang

(2016). Douglas Biber and Bethany Gray: Grammatical complexity in academic English: Linguistic change in writing. Applied Linguistics, 37, 881–884.

21.

Jin

(2020). Syntactic complexity in adapted teaching materials: Differences among grade levels and implications for benchmarking. Modern Language Journal, 104, 192–208.

22.

Johnson

M.D.

(2017). Cognitive task complexity and L2 written syntactic complexity, accuracy, lexical complexity, and fluency: A research synthesis and meta-analysis. Journal of Second Language Writing, 37, 13–38.

23.

Johnson

M.D.

Mercado

Acevedo

(2012). The effect of planning sub-processes on L2 writing fluency, grammatical complexity, and lexical complexity. Journal of Second Language Writing, 21, 264–282.

24.

Kellogg

R.T.

(1996). A model of working memory in writing. In Levy

C.M.

Ransdell

(Eds.), The science of writing: Theories, methods, individual differences and applications (pp. 57–71). Mahwah, NJ: Lawrence Erlbaum.

25.

Khezrlou

(2020). The role of task repetition with direct written corrective feedback in L2 writing complexity, accuracy and fluency. Journal of Second Language Studies, 3(1), 31–54.

26.

Kormos

(2011). Task complexity and linguistic and discourse features of narrative writing performance. Journal of Second Language Writing, 20, 148–161.

27.

Kormos

(2014). Differences across modalities of performance: An investigation of linguistic and discourse complexity in narrative tasks. In Byrnes

Manchón

R.M.

(Eds.), Task-based language learning: Insights from and for L2 writing (pp. 193–216). Philadelphia, PA: Benjamins.

28.

Kuiken

Vedder

(2007). Task complexity and measures of linguistic performance in L2 writing. International Review of Applied Linguistics in Language Teaching, 45, 261–284.

29.

Kuiken

Vedder

(2008). Cognitive task complexity and written output in Italian and French as a foreign language. Journal of Second Language Writing, 17, 48–60.

30.

Kuiken

Vedder

(2011). Task complexity and linguistic performance in L2 writing and speaking: The effect of mode. In Robinson

(Ed.), Second language task complexity: Researching the cognition hypothesis of language learning and performance (pp. 91–104). Philadelphia, PA: Benjamins.

31.

Kuiken

Vedder

(2012). Syntactic complexity, lexical variation and accuracy as a function of task complexity and proficiency level in L2 writing and speaking. In Housen

Kuiken

Vedder

(Eds.), Dimensions of L2 performance and proficiency: Complexity, accuracy and fluency in SLA (pp. 143–170). Philadelphia, PA: Benjamins.

32.

Kuiken

Vedder

(2017). Functional adequacy in L2 writing: Towards a new rating scale. Language Testing, 34, 321–336.

33.

Larsen-Freeman

(2009). Adjusting expectations: The study of complexity, accuracy, and fluency in second language acquisition. Applied Linguistics, 30, 579–589.

34.

Leown

Godfroid

(2020). Advancing quantitative research methods. In McKinley

Rose

(Eds.), The Routledge handbook of research methods in applied linguistics (pp. 98–107). London: Routledge.

35.

(2010). Automatic analysis of syntactic complexity in second language writing. International Journal of Corpus Linguistics, 15, 474–496.

36.

(2011). A corpus-based evaluation of syntactic complexity measures as indices of college-level ESL writers’ language development. TESOL Quarterly, 45, 36–62.

37.

(2012). The relationship of lexical richness to the quality of ESL learners’ oral narratives. Modern Language Journal, 96, 190–208.

38.

Manchón

R.M.

(2009). Writing in foreign language contexts. Learning, teaching, and research. Clevedon: Multilingual Matters.

39.

Manchón

R.M.

(2011a). The language learning potential of writing in foreign language contexts: Lessons from research. In Cimasko

Reichelt

(Eds.), Foreign language writing instruction: Principles and practices (pp. 44–64). Anderson, SC: Parlor.

40.

Manchón

R.M.

(Ed.). (2011b). Learning-to-write and writing-to-learn in an additional language. Philadelphia, PA: Benjamins.

41.

Manchón

R.M.

Roca de Larios

(2007). Writing-to-learn in instructed language learning contexts. In Alcón

Safont

(Eds.), Intercultural language use and language learning (pp. 101–121). Dordrecht: Springer.

42.

Manchón

R.M.

Williams

(2016). L2 writing and SLA studies. In Manchón

R.M.

Matsuda

P.K.

(Eds.), Handbook of second and foreign language writing, (pp. 567–586). Boston, MA: de Gruyter.

43.

Norris

J.M.

Ortega

(2009). Towards an organic approach to investigating CAF in instructed SLA: The case of complexity. Applied Linguistics, 30, 555–578.

44.

Ong

(2014). How do planning time and task conditions affect metacognitive processes of L2 writers. Journal of Second Language Writing, 23, 17–30.

45.

Ong

Zhang

L.J.

(2010). Effects of task complexity on the fluency and lexical complexity in EFL students’ argumentative writing. Journal of Second Language Writing, 19, 218–233.

46.

Ong

Zhang

L.J.

(2013). Effects of the manipulation of cognitive processes on EFL writers’ text quality. TESOL Quarterly, 47, 375–398.

47.

Ortega

(2012). Epilogue: Exploring L2 writing–SLA interfaces. Journal of Second Language Writing, 21, 404–415.

48.

Pallotti

(2015). A simple view of linguistic complexity. Second Language Research, 31, 117–134.

49.

Polio

(2017). Second language writing development: A research agenda. Language Teaching, 50, 261–275.

50.

Rahimi

(2018). Effects of increasing the degree of reasoning and the number of elements on L2 argumentative writing. Language Teaching Research, 23, 633–654.

51.

Rahimi

Zhang

L.J.

(2018). Effects of task complexity and planning conditions on L2 argumentative writing production. Discourse Processes, 55, 726–742.

52.

Rahimi

Zhang

L.J.

(2019). Writing task complexity, students’ motivational beliefs, anxiety and their writing production in English as a second language. Reading and Writing, 32, 761–786.

53.

Robinson

(2001a). Task complexity, cognitive resources, and syllabus design: A triadic framework for examining task influences on SLA. In Robinson

(Ed.), Cognition and second language instruction (pp. 287–318). Cambridge: Cambridge University Press.

54.

Robinson

(2001b). Task complexity, task difficulty, and task production: Exploring interactions in a componential framework. Applied Linguistics, 22, 27–57.

55.

Robinson

(2005). Cognitive complexity and task sequencing: Studies in a componential framework for second language task design. International Review of Applied Linguistics in Language Teaching, 43, 1–32.

56.

Robinson

(2011). Second language task complexity, the Cognition Hypothesis, language learning, and performance. In Robinson

(Ed.), Second language task complexity: Researching the cognition hypothesis of language learning and performance (pp. 3–38). Amsterdam: Benjamins.

57.

Ruiz-Funes

(2015). Exploring the potential of second/foreign language writing for language learning: The effects of task factors and learner variables. Journal of Second Language Writing, 28, 1–19.

58.

Schmidt

(2001). Attention. In Robinson

(Ed.), Cognition and second language instruction (pp. 3–32). Cambridge: Cambridge University Press.

59.

Skehan

(1998). A Cognitive approach to language learning. Oxford: Oxford University Press.

60.

Skehan

(2018). Second language task-based performance: Theory, research, assessment. London: Routledge.

61.

Skehan

Foster

(1997). Task type and task processing conditions as influences on foreign language performance. Language Teaching Research, 1(3), 185–211.

62.

Skehan

Foster

(1999). The influence of task structure and processing conditions on narrative retelling. Language Learning, 49, 93–120.

63.

Skehan

Foster

(2001). Cognition and tasks. In Robinson

(Ed.), Cognition and second language instruction (pp. 183–205). Cambridge: Cambridge University Press.

64.

Swain

(1985). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In Gass

Madden

(Eds.), Input in second language acquisition (pp. 235–253). Rowley, MA: Newbury House.

65.

Swain

(1995). Three functions of output in second language learning. In Cook

Seidlhofer

(Eds.), Principle and practice in applied linguistics: Studies in honour of H.G. Widdowson (pp. 125–144). Oxford: Oxford University Press.

66.

Tavakoli

Skehan

(2005). Strategic planning, task structure, and performance testing. In Ellis

(Ed.), Planning and task performance in a second language (pp. 239–273). Philadelphia, PA: Benjamins.

67.

Way

D.P.

Joiner

E.G.

Seaman

M.A.

(2000). Writing in the secondary foreign language classroom: The effects of prompts and tasks on novice learners of French. Modern Language Journal, 84, 171–184.

68.

Wendel

J.N.

(1997). Planning and second language narrative production. Unpublished PhD thesis, Temple University, Japan.

69.

Williams

(2012). The potential role(s) of writing in second language development. Journal of Second Language Writing, 21, 321–331.

70.

Wolfe-Quintero

Inagaki

Kim

H.Y.

(1998). Second language development in writing: measures of fluency, accuracy, and complexity. Honolulu, HI: University of Hawaii, Second Language Teaching and Curriculum Center.

71.

T.S.

Zhang

L. J.

Gaffney

J.S.

(in press). Examining the relative effectiveness of task complexity and cognitive demands on students’ writing in a second language. Studies in Second Language Acquisition.

72.

Yoon

H.J.

Polio

(2017). The linguistic development of students of English as a second language in two written genres. TESOL Quarterly, 51, 275–301.

73.

Zhang

L.J.

(2013). Second language writing as and for second language learning. Journal of Second Language Writing, 22, 446–447.