Photomontage: A new task to change speaking into talking classrooms

Abstract

This study introduces photomontage as a task to facilitate talking in English as a Foreign Language classrooms. Thirty-three undergraduate English major students studying at the University of Guilan were assigned to design a composite photographic image by combining images from separate photographic sources, and use it as the stimulus to initiate talking in class. Students’ talks based on their own self-generated photomontage and their peers’ feedback in all classes were video recorded and the transcription was coded for words, syllables, T-units, content and function words. This transcription was used as the basis for analyzing the change in learners’ oral repertoire prior to, and after the experiment with regard to three areas of complexity, accuracy and fluency (CAF triad). The results of repeated measures analysis of variance and also paired samples t-test showed that photomontage had appealing potentials for fluency and accuracy but not for complexity.

Keywords:

• photomontage
• CAF triad
• complexity
• accuracy
• fluency

Public Interest Statement

This study introduces photomontage as a new task which combines art, and technology to trigger real talking in the English as a Foreign Language classrooms where speaking is a priority. To this end, thirty-three undergraduate students were required to find pictures from separate photographic sources or photographs and create a composite photographic image, either manually or by the help of one of the photo editing software available to them, and then use that as the stimulus to initiate discussion. The video into text transcription of all students’ talks and their peers’ feedback was analyzed for the change in learners’ oral repertoire with regard to three areas of complexity, accuracy and fluency (CAF triad). The results of repeated measures analysis of variance and also paired samples t-test showed that the incorporation of photomontage into speaking classrooms not only engaged the participants in more authentic interaction, but also enhanced their fluency and accuracy, though not the complexity of their utterances.

1. Introduction

Speaking is the ultimate goal of many language learners because “for most people, the ability to speak a language is synonymous with knowing that language” (Celce-Murcia, 2001, p. 32). Hence, teaching speaking and obtaining the best results has always attracted teachers as well as the researchers. However, history of English Language Teaching (ELT) is replete with methods relying on polling and input for teaching speaking, methods which have received numerous critics and entailed users’ dissatisfaction.

To address this deficiency, Thornbury (2005, p. 131) calls for alternatives to change language classrooms into talking classrooms where speaking is a priority and Long and Doughty (2009) suggest art as a possibility to make that happen (p. 75). In the same line, Petrina (2007) underscores the value of pictures because they make use of a massive range of cortical skills: color, dimension, form, line, text, visual rhythm, and specially imagination. They are evocative, more precise and potent in triggering wide array of associations, and therefore creating something that is original and valuable (Petrina, 2007).

To add to the earlier proposals, this study, anchored in the literature describing Vygotsky’s theory of social constructivism, introduces “Photomontage”—the manipulated photograph—as a new task which combines art, technology, and real talking. The postulation is that as learning is thoroughly social, by counting every person’s idea as effective, photomontage would trigger new arrays of discussion and negotiation in the classroom, which, in turn, would leave no room for boredom or lack of ideas to express. In other words, the claim is that the incorporation of photomontage into speaking classrooms might lead students toward a more “dialogic task” (Pritchard, 2007), which is the focus of social constructivism that emphasizes interaction between learners and also others.

Photomontage is defined by the Penguin English Dictionary as: composite picture made from several photographs, art, or process of making this. The students in this study too, were required to find pictures or photographs and create a composite picture either manually or by the help of one of the photo editing software available to them. The presupposition was that by using photomontage as the visual stimuli to onset interaction (Mumford, 2000), or using pictures as triggers for discussion (Kayi, 2006; Petrina, 2007) students’ affective filter would lower, and they would engage in authentic interaction. In particular, the study addresses the general question of how this methodology would influence the students’ oral repertoire with regard to the areas of Complexity, Accuracy, and Fluency (CAF) triad.

2. Review of literature

The fundamental premise of almost all current SLA (Second Language Acquisition) theories and approaches such as Lantolf’s sociocultural theory, Gass’s interactional perspective, Long’s Interaction hypothesis, Vygotsky’s Sociocultural Theory, and also Swain’s Output Hypothesis is that people learn a language by speaking or interaction (Adhikari, 2010). Yet, many researchers, such as Richards and Renandya (2002), state teaching speaking is difficult for many reasons. For one thing, effective communication requires the ability to use language appropriately in social interaction, but classroom talk is frequently limited and is used to check comprehension rather than develop thinking (Fisher, Frey, & Rothenberg, 2008). Besides, many learners are reluctant to talk in classes, partly due to shyness or partly because the class activities bore them (Hamzah & Ting, 2009; Lawtie, 2004). Besides, it is the teacher who talks most of the time or engages students in the game of “what’s in my head” or attaches to initiation-response-feedback loop which leaves no space for students’ free talk and creative responses, and limits their answers and kills creativity. However, as Swain (1985) says people learn to speak by speaking, therefore the class should provide ample opportunities for learners to practice what they have learnt. This is in line with interaction hypothesis that states learners learn faster through interacting or active use of language. (Bailey, 2004, chap. 5)

2.1. Photomontage within communicative language teaching framework

Communicative Language teaching (CLT) requires teachers to create a classroom environment where students have real-life communication, authentic activities, and meaningful tasks that promote oral language. In addition, some activities, the most common type of which are role plays and discussion, task-completion, information-gathering, opinion-sharing, information-transfer, and reasoning-gap activities are suggested to enhance communication.

This study claims that photomontage has the potentials to involve students in an interactive, social, and contextualized communicative event, and result in simultaneous interaction under time constraints because first of all it requires natural language use, as students’ talk revolves around the piece of artwork that they themselves have created. Second, in cases of miscommunication, the speaker is expected to try his/her best to resolve the problem, using communication strategies. Third, the language is simply used to communicate meaning as is the case in natural communication which happens spontaneously and under time pressure, and fourth, the topic of discussion revolves around the speakers’ interest. Furthermore, the language resulted from using Photomontage can be authentic, because it is commonly observed that people see a piece of artwork and start talking to the artist, seeking some information, or even defending ideas upon which they have been produced, or revealing inner thoughts to what it may or may not be. As these functions mirror the assumptions of current CLT as well (Fisher et al., 2008; Richards, 2006), photomontage can be practiced within this framework of language teaching.

2.2. Photomontage in socio-cognitive framework

Learning is both a social and a cognitive process (Richards, 2003), therefore, social and peer interactions would help extend understanding (Fisher et al., 2008). From the socio-cognitive approaches, students need to be given maximum opportunity for authentic social interaction to give them practice in the kinds of communication they will later engage in the community outside the classroom through giving them authentic tasks or projects (Jones, 2013). As for student-generated Photomontage, learners are required to engage in social, authentic interaction that would be a first-hand experience for what they would encounter in the real-world situations. This study claims that photomontage has the potential to give students a novel and creative task to do in order to improve their oral repertoire. There is controversy, however, over the word “creative” (Goodwyn, 2004). Creativity postulates fostering students’ ability way beyond their basic literacy. Although teachers’ universal wish that influences every instructional decision they make, and their every move in the classroom, is to maximize their students’ learning (Hagger, Klingner, & Aceves, 2010), trying to develop new perspectives of teaching is never easy, and so trying to cultivate ownership and initiative among students is a still greater aspiration that is infrequently realized (Kereks & King, 2010).

2.3. Research on teaching speaking

As stated earlier, a common strategy in language teaching is using pictures. Picture flash cards have been utilized for language teaching, for instance for teaching the language of clothes or appearance, for teaching vocabulary, verb tenses, or even in information gap activities (Mumford, 2000). Pictures are useful in evoking mental images for recalling a word or concept, and reinforcing literal, critical, and creative thinking, so they can also be used to teach content (Wood & Tinajero, 2002). Krueger (2012) discusses use of different kinds of picture dictation in the classroom. Basically in picture-based teaching, the visual stimuli or pictures are provided by the teacher and handful activities are suggested for classroom use.

Nurhasan (2011) argues that colorful pictures related to students’ life are able to develop student speaking skill. In many exams such as Test of Spoken English, First Certificate in English, Certificate of Advanced English, etc pictures are used as triggers for narrating a story to assess the candidates’ speaking proficiency. Even some “pic tac toes” have been developed to enhance speaking but the problem is that in almost all cases, the pictures have been provided to learners by books, pamphlets, teachers, etc. The question which arises is would the results be different if the learners make the pictures themselves and use them to represent ideas?

Discussions, role-plays, and conversations are other frequently used strategies to teach speaking. Discussions are probably the most commonly used activity in the oral skills class (Celce-Murcia, 2001). A good discussion should be provocative to stimulate talk. Some instructors use “Free Discussion” and opt for topics in order to make the class involved in interaction (Kayi, 2006). However, Green, Christopher, and Lam (1997) shift the attention from teacher-selected to student-decided activities and maintain that students will participate more in discussions if they are allowed to select discussion topics themselves. Boonkit (2010) in his study found that freedom of topic selection encouraged the participant to feel comfortable and motivated to speak and maximized speaking confidence. The wide range of vocabulary related to the selected topics automatically increased and activated the English as a Foreign Language (EFL) learners’ English lexicon.

More recently, using technology, electronic tools, and software has come to help language teachers (Tysinger & McCoy, 2013). Shih (2010) in his study used video-based blogs to boost public speaking and found it totally effective and also satisfactory from students’ point of view. Lee and Liang (2012) combined visual materials to provoke public speaking through using video technology. Goodwyn (2004) has worked on the moving image and the considerable point about his work is giving students the chance to become producers in their own right in this technology era.

Today integrating technology into a course curriculum when appropriate is proving to be valuable for enhancing and extending the learners’ experience. Evens suggestions for using computer-mediated communication in teaching pronunciation and conversation are put forward to improve students’ oral skill (Minh Hong, 2006).

Yunus, Hashim, Embi, and Lubis (2010) used ICT “TELL ME MORE” as a teaching and learning tool to support the students’ learning of English language (Ghasemi & Hashemi, 2011). PowerPoint has also been used in classes as an educational tool for teaching and delivering materials (Hashemi, Azizinezhad, & Farokhi, 2012).

However, in all such cases, the software used is educational and mostly teachers leave no option for students to manipulate it, therefore students consider it as part of their curriculum and do not approach the task with sufficient enthusiasm.

In general, despite the diversity of the strategies employed in teaching speaking, some points were common. In most cases, teaching aids such as visuals (images, pictures) are used along with topics, technologies, or tasks which were considered appropriate for eliciting talk from the learners. However, none tried to persuade students use technology to compose a piece of visual art and use the product as a stimulus for class discussion, and negotiation. This study therefore is unique in its own way.

2.4. Assessing oral repertoire

Oral proficiency and performance are multi-componential constructs and their assessment should take this feature into consideration (Housen & Kuiken, 2009; Norris & Ortega, 2009). There are a range of approaches to account for performance on language learning tasks. Skehan (2009) suggests CAF triad for evaluating performance in written and oral tasks. He turns to Trade-off Hypothesis and Robinson’s Cognition Hypothesis to show the interrelationships between the elements of triad. He states that performance in each of these areas (CAF) requires attention and working memory involvement, then attending to one source may have negative impact on others. In other words, there is tension between form (complexity and accuracy) and fluency. This was verified by Tavakoli and Skehan (2005) who used picture series narrative and found higher complexity scores. They reasoned that making connection between picture elements increases language complexity. This study too used Skehan’s CAF tried to evaluate learners’ oral repertoire.

3. Method

This quasi-experimental research has examined the effect of one independent variable—photomontage on one dependent variable—oral repertoire assessed at three levels: Complexity, Accuracy, and Fluency. However, as randomization was not feasible, repeated measure was selected as the alternative by this study.

3.1. Participants

Thirty-three English major undergraduates participated in the experiment. The participants were from 18 to 30 years old. In particular, 94% of the participants were from 18 to 22 years old and the remaining 6% from 26 to 30. Prior to the recording, the learners gave their consent to go through the stages of the research, especially the video recording. All of the data—video recorded samples of the participants’ speaking performances—were gathered within some time intervals during the academic year of 2014–2015. The speaking samples were taken from the class called Language Lab, a four-unit course presented during the first two semesters of the BA (Bachelor of Arts) degree program. The course is mandatory for students majoring in English Literature, English Translation, and also English Teaching. The class meets twice a week for 90 minutes. The course is designated for the purpose of boosting learners’ listening and speaking ability.

3.2. Data collection

The participants were required to do two tasks for this class: (1) have discussion based on the input provided by the teacher, and (2) have discussion based on the Photomontage that they had made on their own. The input in task one was a video that was given to the students to watch in advance of every session and discuss it in class. All students had access to the DVD and its transcription at home. The other task was a Photomontage, a unified picture that the students had made by using several pictures to convey a particular message of their own. Every speaking event in either task was recorded for further analyses. The three components of CAF were measured as shown in Sections 3.2.1–3.2.3.

3.2.1. Complexity measures

Two basic measures were used for calculating complexity. One is the percentage of content words or type token ratio in which the ratio of content words to all words is multiplied by 100, called Lexical Density (in this study we will show it LD hereafter). In addition to the percentage of content words, the ratio of content to function words is also used. The calculation uses the same raw data needed for the calculation of the percentage of content words, establishes a ratio between the two categories of words, that is, content and function words (in this study we will present it with L/F). The following are the formulas used for each measure:

(1)	Percentage of content words = Number of content words/Total number of words × 100.
(2)	Ratio of content to function words = Number of content/Number of function words × 100.

Content words (also lexical words) enter into open sets which are infinitely extendable. Conversely grammatical words (function words) are words which operate in closed, finite systems in the language. The Content words are: nouns, adjectives, adverbs of time, place and manner, verb, multiword verbs, idioms and contraction of pronouns and main verbs. Function words (grammatical words) include: modals, auxiliary, determiners, pronouns, interrogative adverbs, negative adverbs, prepositions, conjunctions, discourse markers, sequencers, particles, lexicalized clauses, quantifier phrases, and reactive tokens (Mousavi, 1999).

3.2.2. Accuracy measures

Accuracy (correctness) means the degree of deviancy from a particular norm (Housen & Kuiken, 2009). Accuracy has been mainly measured by calculating the percentage of error-free verb forms (EFVF) and error-free T-units (Gilabert, 2004). By error, it is meant a linguistic form or combination of forms, which in the same context and under similar conditions of production would, in all likelihood, not be produced by the speakers’ native speaker counterparts (Lennon, 1996). The percentage of EFVF and error free T-units (EFTU) are used in this study. The following are the formulas for the calculation of these two measurements:

(1)	Percentage of error-free verb forms (EFVF) = Number of error-free verb forms/Number of verb forms × 100.
(2)	Percentage of error-free T-units (EFTU) = Number of error-free T-units/Number of T-units × 100.

3.2.3. Fluency measures

There has been a wide variety of approaches to measuring fluency. In this study, we utilized three of them, each with two ways to calculate, one is per minute of speech in the two tasks and the other in the whole speech they gave in each task because there is not time limitation for the tasks (Gilabert, 2004).

3.3. Data analysis

The study employed descriptive statistics, repeated measures analysis of variance (ANOVA), as well as paired samples t-tests for the comparison of the two task features based on the CAF triad. The subsets for the CAF triad were: Complexity (1) Percentage of Lexical Words (LD), Complexity (2) Ratio of Lexical to Function words (L/F), Accuracy (1) the percentage of error free verb forms (EFVF), Accuracy (2) the percentage of Error-free T-units (EFTU) and Fluency (1-a) average number of words per minute of each task (ANWPM), (1-b) average number of words in the whole task (ANWT), (2-a) average number of T-units per minute of each task (ANTUPM), (2-b) average number of T-units in the whole tasks (ANTUT), (3-a) average number of syllables per minute of each task (ANSPM), (3-b) average number of syllables in the whole tasks (ANST). All statistical analyses were carried out using SPSS 20 for Windows. Inter-rater reliability was calculated by means of percentage agreement in 10% of the samples.

3.4. The transcription and coding procedure

First of all the videos were watched through. For the second stage, the software Pot Player was used to control the recordings while in a Microsoft Word file; therefore simultaneous typing and access to the video file was possible. In this study, Thornbury (2005) guideline is used because it was more straightforward and easier to follow.

Table

=	Contiguous utterances, ones that runs without pause, despite interruptions from others
/	Overlapping utterances
//	Simultaneous utterances
()	A slip
[]	For laugh or unclear talk

All data were coded for T-units, words, and syllables in order to calculate fluency, also the content and function words were codified in order to make the counting easier for complexity measures, furthermore all kinds of errors in the text were highlighted, tagging different kinds of errors for accuracy measures, to ease operationalization of the procedures of the current study.

4. Research results

The objectives of the current study were twofold: (1) to examine the effect of photomontage—on students’ oral repertoire (2) to trace the changes in three areas of CAF. To meet these objectives, two hypotheses were adopted and tested. This section presents the findings.

4.1. Hypothesis 1

It was predicted that photomontage would have a positive impact on learners’ production. To trace the differences on the participants’ oral repertoire, all participants’ performance on the two tasks was compared. To check whether the differences in the means of students’ performances on the two tasks under investigation were equal or statistically different, repeated measures ANOVA was run. The results of descriptive statistics are displayed in Table 1.

Table 1. Descriptive statistics

	Mean	SD	N
Oral repertoire 1	1,114.8376	449.52710	33
Oral repertoire 2	3,316.3813	1,819.69263	33

To examine the significance of task, a total index for each students was calculated. For easy reference, the total score for teacher provided input was named Oral Repertoire 1, and the total for the photomontage task was called Oral Repertoire 2. Table 2 presents the contribution of the task type to any improvement in the students’ talking skill.

Table 2. Multivariate tests

Effect		Value	F	Hypothesis df	Error df	Sig.	Partial η^2
Task	Pillai’s Trace	.667	64.180^a	1.000	32.000	.000	.667
	Wilks’ Lambda	.333	64.180^a	1.000	32.000	.000	.667
	Hotelling’s Trace	2.006	64.180^a	1.000	32.000	.000	.667
	Roy’s Largest Root	2.006	64.180^a	1.000	32.000	.000	.667

^a Independent variable or factor.

As is evident from Table 2, there is significant effect of task type on oral repertoire [Wilk’s λ = .333, F (1, 32) = 64.180, p < .0005, multivariate partial η² = .667]. In addition, the multivariate partial η² = .667implies a very large effect size, according to Cohen (1988 as cited in Pallant, 2005).

In addition, to indicate if there was an overall significant difference between the means at the different conditions, Tests of Within-Subjects Effects were employed (Table 3).

Table 3. Tests of within-subjects effects

Measure: oral repertoire
Source		Type III sum of squares	df	Mean square	F	Sig.	Partial η^2
Task	Sphericity assumed	79,972,112.436	1	79,972,112.436	64.180	.000	.667
	Greenhouse–Geisser	79,972,112.436	1.000	79,972,112.436	64.180	.000	.667
	Huynh–Feldt	79,972,112.436	1.000	79,972,112.436	64.180	.000	.667
	Lower-bound	79,972,112.436	1.000	79,972,112.436	64.180	.000	.667
Error (Task)	Sphericity assumed	39,873,993.209	32	1,246,062.288
	Greenhouse–Geisser	39,873,993.209	32.000	1,246,062.288
	Huynh–Feldt	39,873,993.209	32.000	1,246,062.288
	Lower-bound	39,873,993.209	32.000	1,246,062.288

From Table 3, we discover that there is significant effect for the task [sig = .000, F = 64.18, p < .05]. The contrast in within-subject effect is presented in Table 4.

Table 4. Tests of within-subjects contrasts

Measure: oral repertoire
Source	Task	Type III sum of squares	df	Mean square	F	Sig.	Partial η^2
Task	Linear	79,972,112.436	1	79,972,112.436	64.180	.000	.667
Error (Task)	Linear	39,873,993.209	32	1,246,062.288

As Table 4 reveals there is significant effect for task [sig = .000, F = 64.18, p < .05]. Besides, to answer the question if the participants differ on their scores on either task type, the tests of between-subjects effects were employed (Table 5).

Table 5. Tests of between-subjects effects

Measure: oral repertoire
Transformed variable: average
Source	Type III sum of squares	df	Mean square	F	Sig.	Partial η^2
Intercept	323,989,063.955	1	323,989,063.955	142.897	.000	.817
Error	72,553,394.916	32	2,267,293.591

The results in Table 5 indicate significant difference in performance on two tasks [sig = .000, F = 142.89, p < 05].

To conclude, it can be stated that task type affects participants’ performance. From the tables above, it can be found out that student-generated photomontage task had more effect on improving students’ oral repertoire than the first input-based task.

4.2. Hypothesis 2

Hypothesis two predicted that the mean difference regarding the CAF Triad would be higher in the second task—student-generated photomontage. To test the hypothesis, paired samples t-tests were employed to compute the pairwise differences between values of CAF components for each case, and to test whether the average differs from 0 (Table 6).

Table 6. Descriptive statistics for paired samples t-tests, means and standard deviations

		Mean	N	SD	Std. error mean
Pair 1	Complexity1	1.3160	33	.19447	.03385
	Complexity2	1.2040	33	.24153	.04204
Pair 2	Accuracy1	1.3095	33	.26104	.04544
	Accuracy2	1.7834	33	.19646	.03420
Pair 3	Fluency1	1,112.2121	33	449.60363	78.26595
	Fluency2	3,313.3939	33	1,819.79355	316.78540

From Table 6, it can be understood that accuracy (X⁻ = 1.30 and SD = .26 in model-based, and X⁻ = 1.78 and SD = .19 in the photomontage task) and fluency (X⁻ = 1,112.21 and SD = 449.60 for the input-based task and X⁻ = 3,313.39 and SD = 1,819.79 for the photomontage task) have benefitted most from the second task, student-generated photomontage, whereas complexity (with the X⁻ = 1.31 and the SD = .19 for input-based lecture and X⁻ = 1.20 and SD = .24 for student generated photomontage task) benefitted least.

Tables 7 and 8 show the paired samples t-test results for the two groups, in the two tasks in three domains of Complexity, Accuracy and Fluency.

Table 7. Paired samples correlations

		N	Correlation	Sig.
Pair 1	Complexity1 and Complexity2	33	.242	.175
Pair 2	Accuracy1 and Accuracy2	33	.303	.086
Pair 3	Fluency1 and Fluency2	33	.624	.000

Table 8. Paired samples test

		Paired differences					t	df	Sig (2-tailed)
		Mean	SD	Std. error mean	95% Confidence interval of the difference
					Lower	Upper
Pair 1	Complexity1 − Complexity2	.11203	.27095	.04717	.01596	.20810	2.375	32	.024
Pair 2	Accuracy1 − Accuracy2	−.47392	.27505	.04788	−.57144	−.37639	−9.898	32	.000
Pair 3	Fluency1 − Fluency2	−2,201.18182	1,578.68785	274.81428	−2,760.96019	−1,641.40345	−8.010	32	.000

As can be seen in Tables 7 and 8, the results of the paired samples t-test indicate a statistically significant increase in accuracy (M = .47, SD = .27) [with t (32) = 9.89, p < .05] and Fluency (M = 2,201.18, SD = 1,578.68) [with t (32) = 8.01, p < .05].

To sum up, hypothesis two which stated that students’ self-generated photomontage will reveal more change in different areas of the CAF triad is supported. Based on the mean score and paired samples t-test results (p = .05), the student-generated photomontage task shows statistically significant change in fluency and accuracy measures and outperforms the input-based one. Therefore, it can be concluded that the student-generated photomontage task had significant effect on learners’ oral repertoire in accuracy and fluency domains and positively affected the quality of their talking in these two areas.

5. Discussion

This study examined the effect of photomontage on EFL learners’ oral repertoire. Hours of videos from 33 undergraduates in two tasks were transcribed and studied with reference to three areas of complexity, accuracy and fluency, the CAF triad. The findings confirmed that student generated photomontage task had more effect on improving students’ oral repertoire. Considering the definition of task as something that learners do or carry out using their existing language resources (Nunan, 2004; Richards, 2006; Richards & Rodgers, 2002), the findings of this research call upon adding another component to the definition: task as something learners choose to do rather than do what they are told. This is not new if we consider that similar need was observed by other researchers such as Brown (2007) who suggested giving students opportunities to initiate oral communication. The same observation was made in the speaking behavior of the students on the task chosen by this research Photomontage. In fact, it was confirmed that while photomontage can fit the notion of task properly (Hunt, 2002), it can have the added value that students decide about the task and their understanding of their peer-generated pictures, however the responses being divergent. This is compatible with the definition of speaking that is wanting to say something. So any task which reinforces the authentic need to speak is likely to be more effective than the most common types of output activities in EFL speaking classrooms. In other words, what is needed to be done for the students in EFL oral classes is to encourage them to move from interest to involvement to commitment.

In addition, the statistical analysis for the second hypothesis indicated that although implementing photomontage does not show any significant change in learners’ complexity domain—lexical density and lexical complexity—in comparison with their input-based lecture (sig = .024), it does reveal a significant change in learners’ accuracy (sig = .000), and fluency (sig = .000).

This observation—improvement in accuracy and fluency, but not in complexity—supports the discontinuity of development process and suggests that distinct and separate stages with different kinds of behavior occur in each stage of the learners’ developmental process, although there is no exact time at which everybody manifests an ability. The findings contribute to the literature on developmental process, because they tackle one line of future research—dealing with the discontinuity of contextual and cultural models.

Finally, the improvement in two of the three components of the CAF triad is partly in line with Skehan (2009) that shows the interrelationships between the elements of triad, but states that there is tension between form (complexity and accuracy) and fluency; though the tension observed here was between communicative components (accuracy/fluency) and formal component (complexity). All of these imply that the concept “task” still stands as a robust and testable construct for teaching and syllabus design.

Additional information

Funding

Funding. The authors received no direct funding for this research.

Notes on contributors

Jaleh Hassaskhah

Jaleh Hassaskhah is an associate professor in the ELT Department at the University of Guilan, Iran. Her areas of interest are curriculum development, language assessment and approaches to language teaching, all with particular interest in the learners’ diversity.

Shohreh Rahimizadeh Asli

Shohreh Rahimizadeh Asli holds an MA in Teaching English as a Foreign Language (TEFL). She is currently a language teacher. Her areas of interest are approaches to language teaching and educational technology.