381 Studies in Second Language Learning and Teaching Department of English Studies, Faculty of Pedagogy and Fine Arts, Adam Mickiewicz University, Kalisz SSLLT 12 (3). 2022. 381-404 http://dx.doi.org/10.14746/ssllt.2022.12.3.3 http://pressto.amu.edu.pl/index.php/ssllt Research trends in task-based language teaching: A bibliometric analysis from 1985 to 2020 Jie Qin South China Agricultural University, Guangzhou, China https://orcid.org/0000-0002-1678-9195 qinjie@scau.edu.cn Lei Lei Shanghai International Studies University, Shanghai, China https://orcid.org/0000-0002-3366-1855 leileicn@126.com Abstract This study offers a bibliometric analysis of research trends in task-based lan- guage teaching (TBLT) from 1985 to 2020. The analysis covers research ques- tions related to the publication trends, venues for publication, productive au- thors, highly cited articles and references and, more importantly, the most frequently explored TBLT-related topics and their developmental patterns across the past 35 years. Results showed that TBLT was still mostly ap- proached from the traditional cognitive-interactionist and psycholinguistic perspectives with a focus on tasks, individuals (i.e., learners and teachers), task-related variables (e.g., task complexity and task repetition), task perfor- mance, and the resultant linguistic forms. While this field of research has wit- nessed a growing interest in learners’ individual differences and computer- mediated, technologies-assisted learning, a decreasing trend has been ob- served in topics related to error and recast. Implications for task-based re- search, pedagogy, and research methodologies are discussed. Keywords: task-based language teaching; bibliometric analysis; research trend Jie Qin, Lei Lei 382 1. Introduction Task-based language teaching (TBLT), also labeled as task-based language learn- ing and task-based instruction, constitutes an approach to language teaching and learning that prioritizes the use of authentic language to complete mean- ingful tasks in the target language. Pedagogically originating from communica- tive language teaching and solidly grounded in second language acquisition (SLA) theories and research, TBLT has been exerting a significant influence on the teaching and learning of a second or foreign language (L2/FL) since its initiation in the 1980s (Candlin, 1987; Long, 1985; Nunan, 1989; Prabhu, 1987). By focus- ing on authentic, communicative tasks, TBLT emphasizes learners’ incidental ac- quisition of and engagement with language as a meaning-making tool. Hence, it constitutes a radical departure from the traditional, structural approaches that consider language as an object to be systematically taught and intentionally learned (Ellis et al., 2020; Long, 2016; Van den Branden, 2016). TBLT is far from being a single, monolithic approach. Researchers distin- guish between the strong and weak versions of TBLT1 and acknowledge a variety of perspectives from which TBLT may be theoretically approached, such as the cognitive-interactionist, psycholinguistic, sociocultural, psychological, and edu- cational perspectives (Ellis et al., 2020).2 Additionally, TBLT is an important area of language teaching where pedagogy and research are complementary and closely intertwined (Ellis et al., 2020). That is, attention to TBLT comes from not only SLA researchers interested in the effects of features of task design and their implementation on learning results (e.g., Qin, 2019), but also teachers and edu- cational authorities concerned with designing and implementing effective pro- grams and language instruction (e.g., Robinson, 2011). As the present study aims to conduct a systematic review of TBLT as a whole and its developmental trend, it incorporates various versions and theoretical perspectives on TBLT and takes into consideration both task-based pedagogy and task-based research. 1 The two versions of TBLT hold different viewpoints regarding the role of tasks in language teaching. The strong version argues that tasks should be the unit of language teaching, while everything else should be subsidiary. In contrast, the weak version claims that although tasks are a vital part of language instruction, they may be preceded or followed by focused in- struction (see Skehan, 1996, for a more elaborate discussion). 2 Ellis et al. (2020) elaborate on five theoretical perspectives on task-based research. The cog- nitive-interactionist approach examines the relationship between tasks, interaction, and ac- quisition, while the psycholinguistic perspective delves into the cognitive processes involved in the production of L2 tasks. In the sociocultural perspective, a task is viewed as an artefact for mediating learning through interaction. The psychological perspective draws on the theory and research that addresses individual learner factors. Finally, the educational perspective fo- cuses on general educational theories and research that draws on educational accounts. Research trends in task-based language teaching: A bibliometric analysis from 1985 to 2020 383 Over the past three decades or so, TBLT has gained a well-respected status among SLA researchers and language teachers (Ellis, 2003; Robinson, 2001; Skehan, 1996; Willis, 1996). It has recently attracted more attention and a grow- ing number of works concerning TBLT have been published, such as monographs (e.g., Ellis et al., 2020; Long, 2015), edited volumes (e.g., Ahmadian & Long, 2021; Samuda et al., 2018), state-of-the-art articles (e.g., Ellis, 2017; Long, 2016; Van den Branden, 2016), and empirical studies (e.g., Michel et al., 2020; Qin & Zhang, 2022). Moreover, an international conference (The International Conference on Task-Based Language Teaching) and a newly launched journal (Journal on Task- Based Language Teaching and Learning) have been dedicated to the discussion of tasks and TBLT. A number of meta-analyses have been conducted by synthesizing the ef- fects of features of TBLT on various outcome measures. Some of them have fo- cused on the overall effects of task-based, interaction-related features of TBLT on learners’ acquisition of specific grammatical and lexical structures (e.g., Cobb, 2010; Keck et al., 2006; Mackey & Goo, 2007). Others have laid emphasis on particular issues of task features, in particular task complexity, within TBLT (e.g., Jackson & Suethanapornkul, 2013; Johnson, 2017; Sasayama et al., 2015). In comparison, few of the meta-analyses have targeted the implementation and evaluation of long-term TBLT programs (e.g., Bryfonski & McKay, 2019). These meta-analyses, along with other research syntheses (e.g., Plonsky & Kim, 2016), have contributed much to our understanding of task-based research (e.g., task- based interaction in Mackey & Goo, 2007). However, what is still lacking in this line of inquiry is a presentation of the research status and current trends of TBLT as a whole. Up to now, the field has still been unequipped with a systematic, quantitative overview of the most frequently explored TBLT-related topics and their developmental patterns that have come to the fore since its emergence. Bibliometric analysis is a technique that uses bibliographic information to explore the research trends in a specific area (e.g., Lin & Lei, 2020 ) or country (e.g., Lei & Liao, 2017). It has recently begun to attract attention in the field of applied linguistics, due to its convenience and robustness in data analyses (e.g., Zhang, 2020). In particular, it has been applied to survey research in certain ar- eas, such as multilingualism (Lin & Lei, 2020), English as a lingua franca (O’Neil, 2018), and cognitive processing of emotion words (Liu et al., 2020). Thus, in this study, we adopt the bibliometric perspective to explore the developmental tra- jectory of TBLT research since 1985 when Long initiated the proposal (Long, 1985). More specifically, the following questions are to be addressed: 1) What is the research status in the field of TBLT? 2) What are the research trends in TBLT? Jie Qin, Lei Lei 384 The status of TBLT will be addressed by looking at the overall publication trend in the examined years, the major publication venues, the most productive authors in the research area, and the most highly cited articles and references. The research trends will be explored by examining the most frequently explored TBLT-related topics across the examined years. It is hoped that such a biblio- metric exploration can help us better understand research concerning task- based language teaching and learning, and the resultant synthesis may be of much interest and significance to TBLT researchers, educational practitioners, syllabus designers, as well as language policy makers. In the sections below, we first describe the methods used for the data analy- sis, followed by the results and discussion with regard to the research questions, with a focus on the most frequently explored topics across time. Finally, implications for task-based research, pedagogy, and research methodologies are discussed. 2. Methods In this section, we describe the methods that were used in the study for the data analysis. 2.1. Data The data that we used in the present study were the bibliometric information of journal articles downloaded from the Web of Science. In order to more accu- rately and exhaustively harvest the information of the articles on TBLT, we first consulted literature pertinent to TBLT (e.g., Bryfonski & McKay, 2019; Ellis et al., 2020; Long, 2016; Plonsky & Kim, 2016) and prepared a list of 41 TBLT-related search terms (see Appendix). Then, following previous bibliometric studies in applied linguistics such as Lei and Liu (2019b), we queried the terms in all SSCI- and A&HCI-indexed journals in (applied) linguistics and education in the Web of Science Core Collection on September 8, 2020. Similar to previous studies such as Zhang (2020), we retrieved the Web of Science Core Collection for our data since the database is considered as one of the most well-known and widely used multidisciplinary bibliometric databases, which indexes high-quality journals with their bibliometric information (Roemer & Borchardt, 2015). We set the span of the queried literature from 1985 to 2020 since, as previously indicated, TBLT research started from 1985 as Michael Long published his seminal work A role for instruction in second language acquisition: Task-based language teach- ing in that year (Long, 1985). We obtained the bibliometric information of 1,215 articles at this stage. Since some of the articles might have been irrelevant to the present study, we closely read the titles and abstracts of the articles and selected a total of 518 articles for the follow-up analysis based on the following Research trends in task-based language teaching: A bibliometric analysis from 1985 to 2020 385 criteria: 1) the article should focus on language teaching or learning, rather than research issues in other subjects such as nursery, physics, chemistry, and so on; and 2) the article should be pertinent to the implementation of either the strong or weak version of a TBLT program or part of the program (such as needs analysis, task-based assessment, etc.); or 3) the article should include at least one task- relevant factor (such as task type, task feature, task condition, etc.) as the inde- pendent or dependent variables. 2.2. Data processing First, we counted the number of publications each year for the analysis of pub- lication trend. It should be noted that 18 of the 518 publications were tagged as “early access,” and we considered them as published in 2020. Then, we counted the number of publications for each journal and for each author for the analyses of major publication venues and most productive researchers in the area. Second, we identified highly cited articles with both the normalized and raw citation counts. We used the normalized citation count because the raw citation count may be biased in favor of earlier publications since they have more chance to receive citations (Lei & Liao, 2017; Lei & Yan, 2016). The normalized citation count was calculated by dividing the raw citation of each article by the total cita- tion count that all articles published in the same year received. For example, the raw citation count of Skehan (2009) was 230 and the total citation count of all the 21 articles published in 2009 was 1202. Hence, the normalized citation count of Skehan (2009) was 0.1913 (230/1202 = 0.1913). It should be pointed out that we only considered the situation when more than one article was published in a cer- tain year. If there was only one article published in a certain year, due to the lim- ited size of the data, the normalized citation count of that article should be 1, which was skewed and meaningless. For the same reason, we also considered high raw citation counts in case some important articles were left out as a result of the normalized citation. To summarize, we combined and reported on the lists of the top 10 highly cited articles from both normalized and raw citation counts, which should paint a fuller picture of the highly cited articles in the area. Third, we extracted all referenced works in the 518 articles and calculated their occurrence, that is, the number of citations they received in the articles. These highly cited works are considered as highly cited references in the re- search area (Lei & Liu, 2019a). Last, we extracted and identified research topics on TBLT. We syntactically parsed and extracted noun phrases from the abstracts with a homemade Python script based on the package spaCy (Lei et al., 2020; the script will be provided upon request). We followed previous studies such as Zhang (2020) and used the Jie Qin, Lei Lei 386 abstracts to identify the topics since the author-provided keywords are very limited in number and important topics may be overlooked (Zhang, 2020). This also applies to the titles since they are short in length. For example, in the following sentence (Ex- ample 1), both simple noun phrases such as It, such joint raised performance, accu- racy, complexity, a function, and task difficulty and complicated noun phrases such as such joint raised performance between accuracy and complexity, accuracy and com- plexity, and a function of task difficulty were parsed and extracted. Example 1 It is argued that such joint raised performance between accuracy and complexity is not a function of task difficulty… (Skehan, 2009) We considered noun phrases as the candidate research topics for the rea- son that a research topic or theme is a lexical noun phrase of high frequency that occurs across a wide range of texts (Justeson & Katz, 1995; Lei et al., 2020). Due to the limited data size, we decided, after several rounds of experimenta- tion, that for a noun phrase to be considered as a candidate topic it should occur at least five times across at least five abstracts. At this stage, a total of 296 noun phrases met the foregoing criteria. Then, we closely read the 296 noun phrases and discussed if they could be considered as candidate research topics on TBLT. Noun phrases such as they, it, this study, and the results were left out, while 94 ones were filtered in. Since some of the noun phrases were fairly similar in meaning, the 94 noun phrases were then combined into 44 research topics. For example, task types and task type were combined as the topic “task type,” and fluency, accuracy, and complexity were combined as “CALF.” In order to identify the trend of the research topics, we categorized the examined years into three research phases (i.e., Phase 1: 1985-2009, Phase 2: 2010-2015, and Phase 3: 2016-2020) for two reasons. First, the categorization should strike a balance between time and data size. The data of the present study were distributed unevenly with many more works published in more re- cent years (see the section on results). We performed several rounds of experi- ments and decided on the present categorization, which seemed to be the most acceptable option, with 96 abstracts for Phase 1 and 211 abstracts for both Phases 2 and 3. Second, we considered 2009 and 2015 as the dividing years since they witnessed important publications such as Skehan (2009), Ellis (2009), and Long (2015) (see the section on result), which may in part justify their roles of the turning points in the development of TBLT research. We then calculated the raw frequency of the topics at each phase and the normalized frequency with the following formula (i.e., the relative frequency per 200 abstracts since both Phases 2 and 3 contained approximately 200 abstracts). Research trends in task-based language teaching: A bibliometric analysis from 1985 to 2020 387 Normalized frequency= Raw frequency Number of abstracts at the phase *200 Finally, we performed Chi-squared tests of the normalized frequencies of each topic and determined the trend of each topic based on the standardized or Pearson residual value of its Chi-squared test. The standardized residual re- fers to the strength of difference between the observed and expected values, or which cell functions on the significance of the Chi-squared test (Sharpe, 2015). If the absolute value of the standardized residual was close to or larger than 2, the topic was considered as experiencing an increasing or decreasing trend; oth- erwise, it was stable across the examined span (Agresti, 2007). 3. Results and discussion In this section, we describe and discuss the findings with regard to the research questions, that is, the publication trend, important publication venues, the most productive authors, the most highly cited articles and highly cited references, and, last but not least, the most frequently explored topics across time. 3.1. Publication trend The number of publications by year is presented in Table 1 and the publication trend is illustrated in Figure 1. It can be observed that during the first two decades (i.e., from 1987 to 2006) since TBLT’s first appearance in the 1980s, barely a few works were published in journals, with an annual figure of only 2 or 3 in general or 6 at most. However, this does not mean that TBLT did not attract researchers’ and educators’ attention at that time, as evidenced by a number of influential works on task-based teaching and learning in terms of monographs (e.g., Ellis, 2003; Nunan, 1989; Prabhu, 1987; Skehan, 1998a; Willis, 1996), edited volumes (e.g., Bygate et al., 2001; Leaver & Willis, 2004; Van den Branden, 2006), and book chapters (e.g., Candlin, 1987; Long, 1985), in addition to journal articles. Beginning from 2007, the number of journal publications related to TBLT began to show a discernibly upward trend, reaching a peak of 53 in 2015, ap- proximately 30 years after the launch of the field in 1985. From then on, the figure remained relatively constant, with a range of 45 to 55 annual publications (except 36 in 2018), making TBLT one of the hot topics in applied linguistics re- search. Such an observation is confirmed by the result of the simple linear re- gression that showed that the number of TBLT publications across the examined years had significantly increased (F(1, 29) = 90.330, p < .001) with a large effect size (Multiple R2 = .757, Adjusted R2 = .749). Jie Qin, Lei Lei 388 Table 1 Number of publications by year Year Number of publications Year Number of publications 1987 1 2006 1 1991 3 2007 9 1992 2 2008 14 1993 1 2009 21 1994 1 2010 20 1995 3 2011 23 1996 4 2012 34 1997 3 2013 37 1998 1 2014 26 1999 5 2015 53 2000 4 2016 45 2001 6 2017 55 2002 5 2018 36 2003 6 2019 46 2004 4 2020 47 2005 2 Total 518 Figure 1 Number of publications by year 3.2. Publication venues The top 10 journals in terms of the number of publications on TBLT are pre- sented in Table 2. All of them are high-impact journals in the discipline of applied linguistics (high-impact in terms of their impact factor since they all rank amongst the top 15 journals out of a total of more than 180 SSCI-indexed lin- guistics journals) and are to a great extent committed to language teaching and learning. The finding seemingly implies that TBLT has been widely accepted as a pedagogical approach to the teaching and learning of an L2/FL. It is of particular interest to note that three of the top 10 journals, that is, Language Learning & Tech- nology, Computer Assisted Language Learning, and ReCALL, are related to technol- ogy and computer-assisted learning. This demonstrates that researchers in TBLT are Research trends in task-based language teaching: A bibliometric analysis from 1985 to 2020 389 also interested in investigating the use of technology and computer techniques to design, implement, and manipulate task-based teaching and learning. Table 2 Top 10 publication venues Journals Number of articles published Language Teaching Research 60 System 41 Modern Language Journal 24 Language Learning & Technology 22 Applied Linguistics 20 Language Learning 19 TESOL Quarterly 18 Journal of Second Language Writing 17 Computer Assisted Language Learning 15 ReCALL 15 3.3. Most productive authors The authors with six or more publications on TBLT are listed in Table 3. Out of the nine authors, Peter Skehan and Rod Ellis, whose articles and books are often cited by TBLT researchers (see Tables 4 and 5), are highly productive as well, which reflects their leading roles in the field. Also, both YouJin Kim and Andrea Révész are remarkably productive in the field of TBLT, each with more than 10 published articles. Other au- thors listed in the table are also well-established scholars in the research area: Judit Kormos, Zsuzsanna Abrams, Laura Gurzynski-Weiss, Craig Lambert, and Caroline Payant. It should also be noted that Michael Long, though not listed as a productive author, contributed papers that are both highly cited and immensely influential (Long, 2015; Long & Crookes, 1992), which, together with his pioneering work (Long, 1985), have secured his position in the field of task-based learning and teaching. Table 3 Productive authors with six or more publications Authors Number of articles published Kim, YouJin 15 Révész, Andrea 11 Kormos, Judit 7 Skehan, Peter 7 Abrams, Zsuzsanna 6 Ellis, Rod 6 Gurzynski-Weiss, Laura 6 Lambert, Craig 6 Payant, Caroline 6 Jie Qin, Lei Lei 390 3.4. Most highly cited articles and highly cited references The most highly cited articles based on both raw and normalized citation counts are reported in Table 4. As can be seen from the table, topping the list are state-of-the- art articles on task-based syllabus design (Long & Crookes, 1992), instruction (Skehan, 1996; Swan, 2005), and performance (Skehan, 2009), which set up conceptual and methodological frameworks for more in-depth examination of specific issues con- cerning task-based learning and teaching. Moreover, many listed articles are theo- retical reviews and empirical studies on task complexity (Bishop et al., 1991; Robin- son, 1995, 2001; Webster & Ryan, 1991) and pre-task planning (Ellis, 2009; Skehan & Foster, 1999; Yuan & Ellis, 2003), two task variables that have generated heated discussions in the field. In addition, one highly cited article addresses task-based in- novation from teachers’ perspective (Carless, 2004) and two others investigate feed- back in task-based interaction (Mackey et al., 2003) and children’s development of phonological sensitivity (Anthony et al., 2003). Table 5 lists the highly cited references extracted from the references of 518 articles. These include some iconic books (Ellis, 2003; Long, 2015; Samuda & Bygate, 2008; Skehan, 1998a; Willis, 1996) and book chapters (Robinson, 2012; Swain, 1995) on tasks and task-based language teaching and learning, in addi- tion to journal articles. For instance, Ellis’s (2003) Task-based language teaching and learning, using tasks as a means of data collection and a teaching tool, es- tablished bridges between SLA research and language pedagogy. Skehan’s (1998a) A cognitive approach to language learning discussed language learning from the perspectives of psycholinguistics, cognition, and individual differences, with a focus on the practical applications of these themes in task-based learning and language testing. Also among the most cited works by TBLT researchers but without a particular focus on tasks is Levelt’s (1989) Speaking: From intention to articulation, a comprehensive book addressing the psycholinguistic processes of speech production in learners’ first language. This testifies to TBLT researchers’ interest in borrowing psycholinguistic frameworks, in particular the speech pro- duction model, to conceptualize, analyze, and explain learners’ processing and production of oral tasks. Furthermore, three articles addressing the methodo- logical issues of SLA research, including the measurements of language complex- ity (Norris & Ortega, 2009), the analysis of speech unit as a measurement unit of spoken language (Foster et al., 2000), and the analysis of statistical power (Cohen, 1988), were also frequently cited by TBLT researchers. The researchers’ concerns on language measurements and statistical analysis are congruent with the growing number of empirical studies in this line of inquiry. Upon a closer look at Tables 4 and 5, it can be seen that five papers are both highly cited articles and highly cited references at the same time, demonstrating Research trends in task-based language teaching: A bibliometric analysis from 1985 to 2020 391 their significant role in the field. These are Skehan’s (1996) proposal for a com- prehensive framework for the implementation of task-based instruction and the methods by which the instruction may be put into practice, as well as two state- of-the-art articles on the cognition hypothesis (Robinson, 2001) versus the trade-off approach3 (Skehan, 2009), the two highly influential and frequently cited yet competing theories that have ignited much empirical research on TBLT. Also listed are Yuan and Ellis’s (2003) Skehan and Foster’s (1999) empirical ex- aminations of how task variables, such as task structure and planning conditions, may affect L2 learners’ speech production, especially in terms of fluency, com- plexity, and accuracy measurements. Table 4 Most highly cited articles Article Title Raw citation Normalized citation Long & Crookes (1992) Three approaches to task-based syllabus design 204 0.9577 Swan (2005) Legislation by hypothesis: The case of task-based instruction 111 0.8880 Skehan (1996) A framework for the implementation of task-based instruction 344 0.7765 Robinson (1995) Task complexity and second-language narrative discourse 93 0.7561 Robinson (2001) Task complexity, task difficulty, and task production: Exploring inter- actions in a componential framework 324 0.6365 Skehan & Foster (1999) The influence of task structure and processing conditions on narrative retellings 203 0.6324 Webster & Ryan (1991) Task complexity and manual reaction times in people who stutter 17 0.5000 Yuan & Ellis (2003) The effects of pre-task planning and on-line planning on fluency, com- plexity and accuracy in L2 monologic oral production 244 0.4171 Bishop et al. (1991) Age and task complexity variables in motor-performance of stuttering and nonstuttering children 14 0.4118 Carless (2004) Issues in teachers’ reinterpretation of a task-based innovation in pri- mary schools 85 0.4106 Anthony et al. (2003) Phonological sensitivity: A quasi-parallel progression of word struc- ture units and cognitive operations 158 0.2701 Mackey et al. (2003) Interactional input and the incorporation of feedback: An exploration of NS-NNS and NNS-NNS adult and child dyads 112 0.1915 Skehan (2009) Modelling second language performance: Integrating complexity, ac- curacy, fluency, and lexis 230 0.1913 Ellis (2009) The differential effects of three types of task planning on the fluency, complexity, and accuracy in L2 oral production 148 0.1231 3 Robinson’s (2010) cognition hypothesis was further developed into the SSARC (i.e., stabilize, sim- plify, automatize, restructure, and complexify) model, while Skehan’s (2009) trade-off approach was later reconceptualized as the limited attentional capacity approach. A recent book chapter offers a comprehensive synopsis and comparison of the two (Ellis et al., 2020, Chapter 3). Jie Qin, Lei Lei 392 Table 5 Most highly cited references Work Title Citation Ellis (2003) Task-based language teaching and learning 148 Skehan (1998a) A cognitive approach to language learning 133 Robinson (2001) Task complexity, task difficulty, and task production: Exploring interactions in a componential framework 90 Levelt (1989) Speaking: From intention to articulation 67 Foster & Skehan (1996) The influence of planning and task type on second language performance 64 Yuan & Ellis (2003) The effects of pre-task planning and on-line planning on fluency, complexity and accuracy in L2 monologic oral production 59 Robinson (2005) Cognitive complexity and task sequencing: Studies in a componential framework for second language task design 58 Samuda & Bygate (2008) Tasks in second language learning 57 Skehan (1996) A framework for the implementation of task-based instruction 54 Skehan & Foster (1997) Task type and task processing conditions as influences on foreign language performance 51 Long (2015) Second language acquisition and task-based language teaching 50 Robinson (2012) Task complexity, cognitive resources, and syllabus design: A triadic framework for examining task influences on SLA 48 Norris & Ortega (2009) Towards an organic approach to investigating CAF in instructed SLA: The case of complexity 46 Skehan (2009) Modelling second language performance: Integrating complexity, accuracy, flu- ency, and lexis 45 Swain (1995) Three functions of output in second language learning 45 Foster et al. (2000) Measuring spoken language: A unit for all reasons 43 Willis (1996) A framework for task-based learning 43 Skehan & Foster (1999) The influence of task structure and processing conditions on narrative retellings 42 Cohen (1988) Statistical power analysis for the behavioral sciences 42 Skehan (1998b) Task-based instruction 37 3.5. Most frequently explored topics across time The identified research trends for each topic, based on the above-described proce- dures of data processing and analyses, fall into six groups. A complete list of the 40 identified topics and their groupings, their normalized frequencies across the three phases (Phase 1: 1985-2009, Phase 2: 2010-2015, and Phase 3: 2016-2020), as well as their Chi-squared, p., and standardized residual values are presented in Table 6. Ten research topics, representing 25% of the total 40, have remained es- sentially constant in frequency over the three phases. Among the topics, the most frequent ones are task complexity (frequency: 27.08, 30.33, 35.07) and task difficulty/demand (frequency: 22.92, 17.06, 23.70), two variables related to how tasks may be sequenced in terms of less to more difficulty or complexity in pedagogical practices. This echoes the observation made in the previous sub- section that studies on task complexity and task difficulty/demand are listed among the most highly cited articles and references in this line of inquiry (see Tables 4 and 5). Ranking below them are meaning, testing/assessment, task implementation and pairs/dyads, all of which are widely explored topics in task-based research with an average frequency of 5 to 10. Each of the remaining topics, that is, strategy, content, and comprehension, has an average frequency of below 5. Research trends in task-based language teaching: A bibliometric analysis from 1985 to 2020 393 Fourteen research topics (35%) have witnessed a significant increase in fre- quency from Phase 1 to Phase 3. First, results display researchers’ growing interest in some essential topics in this field – task (frequency: 50.00, 46.45, 81.52), TBLT (fre- quency: 12.50, 36.97, 46.45), learner (frequency: 29.17, 38.86, 61.61) and teacher (frequency: 6.25, 21.80, 31.28), which is consistent with the accumulating status of TBLT as a whole among SLA researchers and language teachers. It should also be noted that four task-related variables, that is, task planning, task engagement, task repetition, and task modality, experienced a dramatic gain in frequency from almost 0 in Phase 1 to around 15 to 20 in Phase 3 (except task modality). Such results, to- gether with researchers’ sustained interest in task difficulty/demand and task com- plexity, demonstrate the field’s ongoing concerns with tasks and how implementation of factors related to tasks may affect learning results. An exceptional case here is task type (frequency: 2.08, 11.37, 7.58), which, in spite of the ascending tendency, gained less attention in Phase 3 than in Phase 2. Additionally, two measurements of task per- formance – CALF (complexity, accuracy, lexis, and fluency, frequency: 41.67, 48.34, 72.04) for monologues and language-related episode (frequency: 6.25, 14.22, 22.75) for interactions – have garnered growing attention over the past decades. Also note- worthy is researchers’ increasing passion for technology-related TBLT (frequency: 4.17, 19.91, 21.80), such as topics concerning technology, (synchronous) computer- mediated communication, and computer-assisted language learning. Such results echo the presence of three technology-related journals as top publication venues on the one hand (see Table 2) and the broader picture of SLA research in which technol- ogy-related topics have gained momentum on the other (Zhang, 2020). Another group of five topics (12.5%) have also manifested a noticeably upward trend, although their increase is not significant enough. Such topics are task comple- tion (frequency: 2.08, 3.79, 7.58) and task performance (frequency: 16.67, 19.91, 26.54), two typical dependent variables in task-based research, and task condi- tion/characteristics (frequency: 4.17, 3.79, 7.58), another task-related independent variable. Also, moderating variables (frequency: 14.58, 15.17, 21.80) including L2 pro- ficiency, working memory, anxiety, and motivation have gathered momentum. This indicates researchers’ rising interest in the psychological perspective of TBLT, that is, how individual differences may affect the learning results or task performance, in ad- dition to the traditional, cognitive-interactionist perspective (e.g., Révész, 2011). Two topics (5%), that is, error/monitoring (frequency: 12.50, 4.74, 2.84) and recast (frequency: 14.58, 13.27, 2.84), have become less popular within TBLT over the three phases. The downward trend of error/monitoring suggests TBLT researchers’ gradual loss of interest in learner errors or error analysis, which may be partly due to some scholars’ scepticism about native speakerism or stand- ard language ideologies (e.g., Ortega, 2019). The decreasing frequency of recast, together with the noticeable yet insignificant decrease in interaction (frequency: Jie Qin, Lei Lei 394 20.83, 23.70, 10.43) and feedback (frequency: 20.83, 20.85, 10.43), reflects that interactional feedback, which used to be widely explored among TBLT scholars in Phases 1 and 2, has received reduced concern over the past few years. However, this does not necessarily mean that task-based interaction has lost its momentum, as evidenced by researchers’ continuing and rising enthusiasm for pairs/dyads and language-related episode. Rather, it might indicate a potential shift in task- based interactional research from feedback and recast to task-related independ- ent variables (e.g., task complexity, task repetition) and measurements (e.g., lan- guage-related episode, testing/assessment). Two other topics showing the same noticeably but not significantly decreasing trend are form (frequency: 22.92, 15.17, 9.48) and language production (frequency: 8.33, 5.69, 3.79), which con- trasts with the stable popularity of meaning and comprehension respectively. The last group of topics (a total of five, 12.5%) features a sharp decline in frequency in Phase 2 yet regained growth in Phase 3. These topics include peda- gogy, task design, language development, and task-based instruction, each with a frequency of 6 to 13 in Phases 1 and 3, but of less than 4 in Phase 2. This reflects researchers’ fluctuating yet renewed interest in the pedagogical perspective of TBLT and the learners’ long-term development in the target language. Additionally, with regard to the learning contexts, topics related to both for- eign language (frequency: 4.17, 10.43, 21.80) and second language learning con- texts (frequency: 14.58, 18.96, 33.18) have attracted increasing attention across the three phases. As for the languages or countries involved in TBLT studies, interest in English or English-speaking countries remains predominant (frequency: 29.17, 36.97, 50.24), showing a noticeable yet insignificant increase from Phase 1 to Phase 3. Topics relevant to European countries or languages remain constant over the pe- riods (frequency: 14.58, 15.17, 13.27), while those concerning eastern countries or languages witness a significant decrease in Phase 2 (frequency: 12.50, 2.84, 10.43). Table 6 A complete list of topics and their trends in the three phases Topics P1 P2 P3 x_sq df p res. 1 res. 2 res. 3 Remained constant meaning 10.42 6.64 11.37 1.33 2 .52 0.31 -0.92 0.62 strategy 4.17 4.74 1.90 1.26 2 .53 0.30 0.60 -0.90 task difficulty/demand 22.92 17.06 23.70 1.24 2 .54 0.37 -0.90 0.54 content 2.08 2.84 4.74 1.16 2 .56 -0.63 -0.21 0.85 task implementation 8.33 5.69 4.74 1.11 2 .57 0.83 -0.23 -0.61 task complexity 27.08 30.33 35.07 1.05 2 .59 -0.67 -0.09 0.76 testing/assessment 8.33 9.48 12.32 0.84 2 .66 -0.54 -0.18 0.72 comprehension 4.17 2.84 2.84 0.36 2 .84 0.49 -0.24 -0.24 pairs/dyads 6.25 6.64 7.58 0.14 2 .93 -0.22 -0.07 0.29 languages/countries - Europe 14.58 15.17 13.27 0.13 2 .94 0.06 0.22 -0.28 Significantly increased task planning 2.08 14.22 17.06 11.38 2 .00 -2.71 0.93 1.78 task engagement 0.00 6.64 14.22 14.56 2 .00 -2.64 -0.12 2.76 Research trends in task-based language teaching: A bibliometric analysis from 1985 to 2020 395 task repetition 0.00 11.37 18.96 18.01 2 .00 -3.18 0.40 2.78 task 50.00 46.45 81.52 12.56 2 .00 -1.21 -1.67 2.88 TBLT 12.50 36.97 46.45 19.19 2 .00 -3.44 0.88 2.56 learner 29.17 38.86 61.61 12.84 2 .00 -2.14 -0.66 2.80 teacher 6.25 21.80 31.28 16.15 2 .00 -3.04 0.46 2.59 technology-related 4.17 19.91 21.80 12.26 2 .00 -2.84 1.18 1.66 FL learning context 4.17 10.43 21.80 13.18 2 .00 -2.29 -0.49 2.78 CALF 41.67 48.34 72.04 9.43 2 .01 -1.68 -0.77 2.45 language-related episode 6.25 14.22 22.75 9.45 2 .01 -2.15 -0.05 2.20 L2 learning context 14.58 18.96 33.18 8.50 2 .01 -1.62 -0.70 2.32 task type 2.08 11.37 7.58 6.22 2 .04 -1.86 1.65 0.22 task modality 0.00 1.90 5.69 6.64 2 .04 -1.59 -0.40 1.99 Noticeably but not significantly increased languages/countries-English 29.17 36.97 50.24 5.85 2 .05 -1.55 -0.29 1.84 task completion 2.08 3.79 7.58 3.53 2 .17 -1.13 -0.33 1.46 task performance 16.67 19.91 26.54 2.41 2 .30 -0.95 -0.25 1.20 moderating variable 14.58 15.17 21.80 1.87 2 .39 -0.63 -0.49 1.11 task condition/characteristics 4.17 3.79 7.58 1.69 2 .43 -0.45 -0.61 1.06 Significantly decreased error/monitoring 12.50 4.74 2.84 7.82 2 .02 2.24 -0.76 -1.49 recast 14.58 13.27 2.84 8.09 2 .02 1.36 0.95 -2.31 Noticeably but not significantly decreased form 22.92 15.17 9.48 5.74 2 .06 1.77 -0.17 -1.60 interaction 20.83 23.70 10.43 5.32 2 .07 0.59 1.26 -1.84 feedback 20.83 20.85 10.43 4.16 2 .12 0.83 0.84 -1.67 language production 8.33 5.69 3.79 1.75 2 .41 0.98 -0.10 -0.88 Significantly or noticeably decreased in period 2 languages/countries-eastern 12.50 2.84 10.43 6.02 2 .05 1.33 -1.96 0.63 pedagogy 8.33 1.90 8.53 4.56 2 .10 0.83 -1.74 0.91 task design 8.33 1.90 8.53 4.56 2 .10 0.83 -1.74 0.91 language development 8.33 3.79 11.37 3.72 2 .16 0.18 -1.44 1.27 task-based instruction 6.25 2.84 6.64 1.66 2 .44 0.44 -1.05 0.61 To sum up, it seems that TBLT researchers have retained and enhanced their interest in tasks, individuals involved in tasks (i.e., learners, teachers), a range of variables related to task conditions and task implementation (e.g., task complexity, task repetition), and how task-related variables may affect learners’ activities and performances (e.g., CALF, task engagement). In other words, re- search in TBLT has been and is still typically approached from the traditional cognitive-interactionist (e.g., Mackey et al., 2003) and psycholinguistic perspec- tives (e.g., Robinson, 2012), with the emergence of new task-related independ- ent (e.g., task planning) and dependent variables (e.g., language-related epi- sode). Also, TBLT research has been conducted in both L2/FL learning contexts addressing the teaching and learning of a variety of target languages. Besides that, two conspicuous patterns are observed from the results. One is that the psychological perspective of TBLT research, that is, how individual differences may bring about divergent performances of tasks or alter the relationships be- tween task variables and learning effects, has represented a rising trend. An- other pertains to a growing concern with technology-related, task-based teach- ing and learning of languages. In contrast, the field has witnessed a diminished Jie Qin, Lei Lei 396 interest in topics related to errors, recast, and feedback. It should also be pointed out that only a few topics are related to the pedagogical perspective of TBLT (i.e., pedagogy, task-based instruction) or to learners’ long-term develop- ment in the target language (i.e., language development), while no topics have been concerned with the sociocultural perspective. 4. Conclusion and implications This bibliometric study on TBLT has provided a bird’s-eye view of important, val- uable information on the publication trend, venues for publication, productive authors, highly cited articles and highly cited references. More importantly, we have identified the most frequently explored TBLT-related topics and analyzed the developmental patterns of those topics across the past decades, which may help us gain a more profound understanding of key issues related to tasks, task performance, task-based teaching, and so on. Such a synthesis of research brings with it significant implications for task-based research and pedagogy and in terms of methodological innovation. Apart from the traditional cognitive-interactionist and psycholinguistic ap- proaches to TBLT (e.g., Mackey et al., 2003; Robinson, 2012) which are very pop- ular today and will probably remain so in the future, the findings seem to sug- gest that this research area may be further expanded in other ways. One is the emphasis on the individuals involved in task performance, including the consid- eration of learners’ individual differences and the investigation of factors related to teachers, such as teacher training. Another implication concerns the applica- tion of technologies to task-based teaching and learning, which will enhance our understanding of the conceptualization, design, and evaluation of tasks (Gonzá- lez-Lloret & Ortega, 2014). Moreover, research on TBLT should be conducted with a greater variety of learners, with more divergent target languages, and in more diversified learning contexts, in order to gain a more holistic picture of this teaching approach. A further implication, which, however, does not find direct support in the present findings, may be an integration of TBLT with other theo- retical approaches in applied linguistics. For instance, in response to some schol- ars’ call to move the field of SLA, including that of TBLT, forward to its meaning- oriented perspective, a few conceptual articles and empirical studies have en- deavored to integrate systemic functional linguistics with TBLT (Byrnes, 2019; Ortega, 2015; Qin, 2022; Ryshina-Pankova, 2015). The present study also has important implications for pedagogical prac- tices. First, as task-based pedagogy and task-based research are complementary and intertwined with each other, such important data on task-based research as presented in this study will surely inform and guide the teaching and learning Research trends in task-based language teaching: A bibliometric analysis from 1985 to 2020 397 practices, which may in return provide valuable feedback and impetus for fur- ther disciplinary development. Second, the topics that TBLT researchers are most interested in may overlap with issues that language educators are con- cerned with to some extent. For instance, the rising popularity of topics related to teachers partly reflects educationists’ realization of the important role teach- ers play in task-based teaching, in addition to the attention already paid to learn- ers and tasks. Third, findings from this research synthesis also provide potential guidelines for language teaching, such as the sequencing of tasks informed by research results concerning task difficulty and complexity, the recognition that tasks do not work for every individual in the same way, and the active use of various technologies that may mediate teaching and learning. There are also some methodological considerations as to how studies like this should be conducted. For example, one possible methodological implication of this study is the use of dependency-based method for the extraction of re- search topic candidates. Previous bibliometric studies used either the N-grams- based method (Lei & Liu, 2019a, 2019b) or the topic modeling technique (Li & Lei, 2021) to extract topic candidates. Although such methods seem to work, they were challenged due to their manual judgment and interpretability issues (Lei et al., 2020). The present study adopted the dependency-based method and the re- sults revealed it as an effective and efficient approach in the linguistics area with room for improvement. For example, future research may explore other measures than frequency and range as well as more sophisticated methods such as machine learning algorithms for the improvement of the newly proposed method. Finally, it should be pointed out that the present study only used abstracts of research articles to explore TBLT-related topics. Future studies may consider full texts of not only research articles but also texts of other genres such as mon- ographs and book chapters in order to paint a fuller picture of the research in the area. In addition, we used a list of TBLT-related search terms for the retrieval of the research data. Although such a method helped us more accurately and exhaustively harvest the data, the retrieval based on its application, may miss some emerging topics since any list of search terms may not be fully exhaustive. Future research may employ umbrella terms such as “task” to search the data. Acknowledgements We would like to express our gratitude to editors Mirosław Pawlak and Chengchen Li and two anonymous reviewers for their invaluable comments and suggestions. This work was supported by a grant from the 2022 Social Sciences Planning Pro- ject of Guangdong Province, China (project title: An Empirical Study on Develop- ment and Enhancement of Foreign Language Learners’ Creativity). Jie Qin, Lei Lei 398 References Agresti, A. (2007). An introduction to categorical data analysis (2nd ed.). John Wiley & Sons. Ahmadian, M. J., & Long, M. H. (Eds.) (2021). The Cambridge handbook of task- based language teaching. Cambridge University Press. Anthony, J. L., Lonigan, C. J., Driscoll, K., Philips, B. M., & Burgess, S. R. (2003). Phonological sensitivity: A quasi-parallel progression of word structure units and cognitive operations. Reading Research Quarterly, 38(4), 470- 487. https://doi.org/10.1598/rrq.38.4.3 Bishop, J. H., Williams, H. G., & Cooper, W. A. (1991). Age and task complexity variables in motor performance of stuttering and nonstuttering children. Journal of Flu- ency Disorders, 16(4), 207-217. https://doi.org/10.1016/0094-730x(91)90003-u Bryfonski, L., & McKay, T. H. (2019). TBLT implementation and evaluation: A meta- analysis. Language Teaching Research, 23(5), 603-632. https://doi.org/ 10.1177/1362168817744389 Bygate, M., Skehan, P., & Swain, M. (Eds.). (2001). Researching pedagogic tasks: Second language learning, teaching and testing. Longman. Byrnes, H. (2019). Applying SFL for understanding and fostering instructed sec- ond language development. In G. Thompson, W. Bowcher, L. Fontaine, & D. Schönthal (Eds.), The Cambridge handbook of Systemic Functional Lin- guistics (pp. 512-536). Cambridge University Press. Candlin, C. (1987). Towards task-based language learning. In C. Candlin & D. Murphy (Eds.), Language learning tasks (pp. 5-22). Prentice Hall. Carless, D. (2004). Issues in teachers’ reinterpretation of a task-based innovation in pri- mary schools. TESOL Quarterly, 38(4), 639-662. https://doi.org/10.2307/3588283 Cobb, M. (2010). Meta-analysis of the effectiveness of task-based interaction in form-focused instruction of adult learners in foreign and second language teaching [Doctoral dissertation, University of San Francisco]. ProQuest Dissertations and Theses (No. 3442086). Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Routledge. Ellis, R. (2003). Task-based language learning and teaching. Oxford University Press. Ellis, R. (2009). The differential effects of three types of task planning on the fluency, complexity, and accuracy in L2 oral production. Applied Linguistics, 30(4), 474-509. https://doi.org/10.1093/applin/amp042 Ellis, R. (2017). Position paper: Moving task-based language teaching forward. Lan- guage Teaching, 50(4), 507-526. https://doi.org/10.1017/S0261444817000179 Ellis, R., Skehan, P., Li, S., Shintani, N., & Lambert, C. (2020). Task-based language teaching: Theory and practice. Cambridge University Press. Research trends in task-based language teaching: A bibliometric analysis from 1985 to 2020 399 Foster, P., & Skehan, P. (1996). The influence of planning and task type on second language performance. Studies in Second Language Acquisition, 18, 299- 324. https://doi.org/10.1017/S0272263100015047 Foster, P., Tonkyn, A., & Wigglesworth, G. (2000). Measuring spoken language: A unit for all reasons. Applied Linguistics, 21(3), 354-375. https://doi.org/10. 1093/applin/21.3.354 González-Lloret, M., & Ortega, L. (Eds.) (2014). Technology-mediated TBLT: Re- searching technology and tasks. John Benjamins. Jackson, D. O., & Suethanapornkul, S. (2013). The cognition hypothesis: A syn- thesis and meta-analysis of research on second language task complexity. Language Learning, 63, 330-367. https://doi.org/10.1111/lang.12008 Johnson, M. D. (2017). Cognitive task complexity and L2 written syntactic com- plexity, accuracy, lexical complexity, and fluency: A research synthesis and meta-analysis. Journal of Second Language Writing, 37, 13-38. https://do i.org/10.1016/j.jslw.2017.06.001 Justeson, J. S., & Katz, S. M. (1995). Technical terminology: Some linguistic prop- erties and an algorithm for identification in text. Natural Language Engi- neering, 1(1), 9-27. https://doi.org/10.1017/s1351324900000048 Keck, C. M., Iberri-Shea, G., Tracy-Ventura, N., & Wa-Mbaleka, S. (2006). Investi- gating the empirical link between interaction and acquisition: A quantita- tive meta-analysis. In L. Ortega & J. Norris (Eds.), Synthesizing research on language learning and teaching (pp. 91-131). John Benjamins. Leaver, B., & Willis, J. R. (Eds.). (2004). Task-based instruction in foreign language education. Georgetown University Press. Lei, L., & Liao, S. (2017). Publications in linguistics journals from Mainland China, Hong Kong, Taiwan, and Macau (2003-2012): A bibliometric analysis. Journal of Quantitative Linguistics, 24(1), 54-64. https://doi.org/10.1080/0929617 4.2016.1260274 Lei, L., & Liu, D. (2019a). Research trends in applied linguistics from 2005-2016: A bibliometric analysis and its implications. Applied Linguistics, 40(3), 540- 561. https://doi.org/10.1093/applin/amy003 Lei, L., & Liu, D. (2019b). The research trends and contributions of System’s pub- lications over the past four decades (1973-2017): A bibliometric analysis. System, 80, 1-13. https://doi.org/10.1016/j.system.2018.10.003 Lei, L., & Yan, S. (2016). Readability and citations in information science: Evidence from abstracts and articles of four journals (2003-2012). Scientometrics, 108(3), 1155-1169. https://doi.org/10.1007/s11192-016-2036-9 Lei, L., Deng, Y., & Liu, D. (2020). Examining research topics with a dependency- based noun phrase extraction method: A case in accounting. Library Hi Tech. 1-13. https://doi.org/10.1108/LHT-12-2019-0247. Jie Qin, Lei Lei 400 Levelt, W. J. M. (1989). Speaking: From intention to articulation. MIT Press. Li, X., & Lei, L. (2021). A bibliometric analysis of topic modelling studies (2000- 2017). Journal of Information Science, 47(2), 161-175. https://doi.org/10. 1177/0165551519877049 Lin, Z., & Lei, L. (2020). The research trends of multilingualism in applied linguis- tics and education (2000-2019): A bibliometric analysis. Sustainability, 12, 1-14. https://doi.org/10.3390/su12156058 Liu, J. Fan, L., & Yin, H. (2020). A bibliometric analysis on cognitive processing of emotional words. Applied Linguistics, 35, 353-365. https://doi.org/10.1093/llc/fqz025 Long, M. H. (1985). A role for instruction in second language acquisition: Task based language teaching. In K. Hyltenstam & M. Pienemann (Eds.), Modelling and assessing second language development (pp. 77-99). Multilingual Matters. Long, M. H. (2015). Second language acquisition and task-based language teaching. Wiley Blackwell. Long, M. H. (2016). In defense of tasks and TBLT: Nonissues and real issues. An- nual Review of Applied Linguistics, 36, 5-33. https://doi.org/10.1017/S0 267190515000057 Long, M. H., & Crookes, G. (1992). Three approaches to task-based syllabus de- sign. TESOL Quarterly, 26(1), 27-56. https://doi.org/10.2307/3587368 Mackey, A., & Goo, J. (2007). Interaction research in SLA: A meta-analysis and research synthesis. In A. Mackey (Ed.), Conversational interaction in SLA: A collection of empirical studies (pp. 408-452). Oxford University Press. Mackey, A., Oliver, R., & Leeman, J. (2003). Interactional input and the incorporation of feedback: An exploration of NS-NNS and NNS-NNS adult and child dyads. Language Learning, 53(1), 35-66. https://doi.org/10.1111/1467-9922.00210 Michel, M., Révész, A., Lu, X., Kourtali, N.-E., Lee, M., & Borges, L. (2020). Inves- tigating L2 writing processes across independent and integrated tasks: A mixed-methods study. Second Language Research, 36(3), 307-334. https:// doi.org/10.1177/0267658320915501 Norris, J. M., & Ortega, L. (2009). Towards an organic approach to investigating CAF in instructed SLA: The case of complexity. Applied Linguistics, 30(4), 555-578. https://doi.org/10.1093/applin/amp044 Nunan, D. (1989). Designing tasks for the communicative classroom. Cambridge University Press. O’Neil, D. (2018). English as the lingua franca of international publishing. World Englishes, 37(2), 146-165. https://doi.org/10.1111/weng.12293 Ortega, L. (2015). Syntactic complexity in L2 writing: Progress and expansion. Journal of Second Language Writing, 29, 82-94. https://doi.org/10.1016/ j.jslw.2015.06.008 Research trends in task-based language teaching: A bibliometric analysis from 1985 to 2020 401 Ortega, L. (2019). Orchestrating second language learning in classrooms: Nudg- ing for a sea-change. Paper presented as a plenary talk at the 29th EU- ROSLA Conference, University of Lund, Sweden, August 28-31. Plonsky, L., & Kim, Y. (2016). Task-based learner production: A substantive and methodological review. Annual Review of Applied Linguistics, 36, 73-97. https://doi.org/10.1017/S0267190516000015 Prabhu, N. S. (1987). Second language pedagogy. Oxford University Press. Qin, J. (2019). Effects of repeated practice in pre-task planning on the acquisition of English personal pronouns by Chinese EFL learners. System, 81(2), 100- 109. https://doi.org/10.1016/j.system.2019.01.010 Qin, J. (2022). Potential contribution of SFL to task-based research: An examina- tion of planning effects using genre-based theme analysis. System, 104, 102695. https://doi.org/10.1016/j.system.2021.102695 Qin, J., & Zhang, Y. (2022). Pre-task planning and discourse cohesion: Analysis of Chinese EFL learners’ referential use in oral narratives. Language Teaching Research, 26(1), 60-78. https://doi.org/10.1177/1362168819883896 Révész, A. (2011). Task complexity, focus on L2 constructions, and individual differ- ences: A classroom-based study. Modern Language Journal, 95 (Supplemen- tary Issue), 162-181. https://doi.org/10.1111/j.1540-4781.2011.01241.x Robinson, P. (1995). Task complexity and second language narrative discourse. Language Learning, 45(1), 99-140. https://doi.org/10.1111/j.1467-1770. 1995.tb00964.x Robinson, P. (2001). Task complexity, task difficulty, and task production: Explor- ing interactions in a componential framework. Applied Linguistics, 22(1), 27-57. https://doi.org/10.1093/applin/22.1.27 Robinson, P. (2005). Cognitive complexity and task sequencing: Studies in a com- ponential framework for second language task design. International Re- view of Applied Linguistics in Language Teaching, 43(1), 1-32. https://doi.org/ 10.1515/iral.2005.43.1.1 Robinson, P. (2010). Situating and distributing cognition across task demands: The SSARC model of pedagogic task sequencing. In M. Putz & L. Sicola (Eds.), Cognitive processing in second language acquisition: Inside the learner’s mind (pp. 243-268). John Benjamins. Robinson, P. (2011). Task-based language learning: A review of issues. Language Learn- ing, 61(Suppl. 1), 1-36. https://doi.org/10.1111/j.1467-9922.2011.00641.x Robinson, P. (2012). Task complexity, cognitive resources, and syllabus design: A triadic framework for examining task influences on SLA. In P. Robinson (Ed.), Cognition and second language instruction (pp. 287-318). Cambridge University Press. Roemer, R. C., & Borchardt, R. (2015). Meaningful metrics: A 21st-century librarian’s guide to bibliometrics, altmetrics, and research impact. American Library Association. Jie Qin, Lei Lei 402 Ryshina-Pankova, M. (2015). A meaning-based approach to the study of com- plexity in L2 writing: The case of grammatical metaphor. Journal of Second Language Writing, 29, 51-63. https://doi.org/10.1016/j.jslw.2015.06.005 Samuda, V., & Bygate, M. (2008). Tasks in second language learning. Palgrave Macmillan. Samuda, V., Van den Branden, K., & Bygate, M. (Eds.). (2018). TBLT as a researched pedagogy. John Benjamins. Sasayama, S., Malicka, A., & Norris, J. (2015). Primary challenges in cognitive task complexity research: Results of a comprehensive research synthesis. Paper presented at the colloquium “An international collaborative research net- work (CRN) on task complexity,” Sixth International Conference on Task- Based Language Teaching (TBLT), Katolieke Universiteit Leuven, Belgium. Sharpe, D. (2015). Chi-square test is statistically significant: Now what? Practical Assess- ment, Research, and Evaluation, 20(8), 1-10. https://doi.org/10.7275/tbfa-x148 Skehan, P. (1996). A framework for the implementation of task-based instruction. Applied Linguistics, 17(1), 38-62. https://doi.org/10.1093/applin/17.1.38 Skehan, P. (1998a). A cognitive approach to language learning. Oxford University Press. Skehan, P. (1998b). Task-based instruction. Annual Review of Applied Linguistics, 18, 268-286. https://doi.org/10.1017/S0267190500003585 Skehan, P. (2009). Modelling second language performance: Integrating com- plexity, accuracy, fluency, and lexis. Applied Linguistics, 30(4), 510-532. https://doi.org/10.1093/applin/amp047 Skehan, P., & Foster, P. (1997). Task type and task processing conditions as influ- ences on foreign language performance. Language Teaching Research, 1(3), 185-211. https://doi.org/10.1177/136216889700100302 Skehan, P., & Foster, P. (1999). The influence of task structure and processing conditions on narrative retellings. Language Learning, 49(1), 93-120. https://doi.org/10.1111/1467-9922.00071 Swain, M. (1995). Three functions of output in second language learning. In H. G. Widdowson, G. Cook, & B. Seidlhofer (Eds.), Principle and practice in applied linguistics: Studies in honour of H. G. Widdowson (pp. 125-144). Oxford University Press. Swan, M. (2005). Legislation by hypothesis: The case of task-based instruction. Ap- plied Linguistics, 26(3), 376-401. https://doi.org/10.1093/applin/ami013 Van den Branden, K. (Ed.) (2006). Task-based language education: From theory to practice. Cambridge University Press. Van den Branden, K. (2016). Task-based language teaching. In G. Hall (Ed.), The Routledge handbook of English language teaching. Routledge. Webster, W. G., & Ryan, C. R. (1991). Task complexity and manual reaction times in people who stutter. Journal of Speech and Hearing Research, 34(4), 708- 714. https://doi.org/10.1044/jshr.3404.708 Research trends in task-based language teaching: A bibliometric analysis from 1985 to 2020 403 Willis, J. R. (1996). A framework for task-based learning. Longman. Yuan, F., & Ellis, R. (2003). The effects of pre-task planning and on-line planning on fluency, complexity and accuracy in L2 monologic oral production. Ap- plied Linguistics, 24(1), 1-27. https://doi.org/10.1093/applin/24.1.1 Zhang, X. (2020). A bibliometric analysis of second language acquisition between 1997 and 2018. Studies in Second Language Acquisition, 42(1), 199-222. https://doi.org/10.1017/s0272263119000573 Jie Qin, Lei Lei 404 APPENDIX Search terms used for the Web of Science query* We categorize the TBLT-related search terms into three types. The first type corresponds to some general terms including TBLT, TSLT, TBL, task-based language teaching, task-supported language teaching, task-based learning, task-based learning research, task-based instruc- tion, task-based interaction, task-based assessment, task-based evaluation, task-based im- plementation, task-based innovation, and task-based performance. Additionally, as the field distinguishes the strong and weak versions of TBLT (a distinc- tion can be found in Skehan, 1996), we also categorize the search terms into the two types. Search terms concerned with the strong version of TBLT include task-based needs analysis, target task, pedagogic task, task syllabus, methodological principle, pedagogic procedure, and task-based performance test. Those relevant to the weak version of TBLT cover such terms as task type, task complexity, task difficulty, task repetition, task sequencing, task planning, task familiarity, task implementation, task feature, task characteristics, task mo- dality, task design, task condition, task-based program evaluation, task grading, and task structure. The inclusion of these search terms was based on two important monographs of the field (Ellis et al., 2020; Long, 2016) and some meta-analyses (Bryfonski & McKay, 2019; Plonsky & Kim, 2016). * For terms with a hyphen such as task-based and task-supported, we searched variants with and without a hyphen for more accurate results (i.e., task-based, task based, task-supported, and task supported).