Vocabulary is a crucial component of any language. It is, therefore, very important for language instructors to enhance learners’ vocabulary knowledge. Second language vocabulary may be taught through meaning-focused and form-focused tasks. Ellis (2003) claims that form-focused tasks attract learners’ attention to form in order to elicit specific linguistic features, and Teklesellassie (2016) believes that meaning-focused tasks are useful for improving the ability to communicate in English. Similarly, Abrams and Byrd (2016) point out that meaning-focused tasks can improve grammatical accuracy and lexical richness.

Based on function, tasks can be divided into two categories of output-oriented and input-oriented tasks. The advocates of input-oriented tasks regret that much of the classroom practice is allocated to output-oriented tasks in such a way that fluency in production precedes input-based activities that encourage receptive comprehension. On the other hand, according to Renandya (2012), output-based tasks such as different types of drills, information-gap activities and oral communication games enable students to develop fluency in language use.

According to Laufer and Girsai (2008), form-focused instruction has its roots in grammar learning, and we can find a large number of studies done in the context of grammar. However, the concept of form-focused instruction can be extended to other parts of language like vocabulary (Zarei & Afrash Ab, 2013).

A quick look at the relevant literature suggests that many studies have focused on grammar tasks, and the number of studies on the other components of language, like vocabulary, is small (Saeidi, Zaferanieh, & Shatery, 2012). In addition, in Iranian context, English is taught as a foreign language (EFL). The literature on form and meaning-focused tasks shows that many studies have been carried out in ESL contexts. This further corroborates the need to conduct a study in an EFL context to find out whether such instruction is adaptable to the system of language education in EFL contexts. In response to this need, this study investigated the effect of form-focused and meaning-focused output-oriented versus input-oriented task-based instruction on Iranian EFL learners’ vocabulary learning. 

LITERATURE REVIEW

Almost everyone agrees that vocabulary learning has a fundamental role to play in second language learning (Pérez & Alvira, 2017). However, in spite of the almost unanimous agreement among experts as to the importance of vocabulary, there is little agreement among experts and researchers about what it means to fully know a word and what kind of knowledge this is. Earlier definitions consider vocabulary as the knowledge of word meaning and the level of one’s access to this knowledge, but such definitions ignore such aspects of word knowledge as pronunciation, spelling, or morphological and syntactic properties (Qian, 2002). On the other hand, Nation (1990) included these aspects of lexical knowledge in his definition of vocabulary knowledge. He argued that a person’s lexical knowledge involves both receptive and productive knowledge, and all aspects of what is involved in knowing a word such as forms, meaning and usage.

As to the ways of learning vocabulary, there are several conflicting viewpoints as to the relative superiority of the two common approaches to learning second language vocabulary: contextual learning of words versus learning words out of context. Oxford and Scarcella (1994), for example, contend that while decontextualized learning (word lists) may help students memorize vocabulary for tests, students are likely to rapidly forget words memorized from lists. However, few studies have actually supported context-dependent vocabulary learning (e.g., Hulstjin, 1992; Tudor & Hafiz, 1989), leading people like Nation (2003) to the conclusion that the above claim is largely wrong and goes against the findings of second language research. Nonetheless, since the advent of the so-called ‘Task-based Approach’, many attempts have been made to teach vocabulary through tasks.

The term ‘task’ has also been defined variously. Littlewood (2004) refers to several definitions offered for task according to the extent to which a task focuses on communicative purpose as an essential criterion. Littlewood considers two dimensions that he thinks are crucial to understanding tasks. The first dimension is a continuum from focus on form to focus on meaning. The second dimension has to do with the degree of learner involvement that a task elicits. Ellis (2000) believes that a task is a ‘work plan’ that involves (a) some input (i.e. information requiring to process and use); and (b) some instructions relating to what outcome the learners are supposed to achieve. Bygate (2000) suggested the term ‘pedagogic task’, which refers to the purposeful language processing activities that are framed in a definite structure and which learners undertake in order to learn. 

According to Renandya (2012), input-oriented tasks develop learners' underlying linguistic system, whereas output-based tasks develop learners' skillful use of language. He adds that a large amount of practice is required in order to develop automaticity in both comprehension (through input-oriented practice) and production (through output-oriented practice).

Input-oriented tasks have not been regarded in the classroom as much as output-oriented tasks, which are popular among ELT professionals. This issue can be realized in many language programs as learners often ask how to use English for real communication situations early in the course. Therefore, most of the communicatively-based courses involve learners in classroom activities that encourage language production through role-plays, simulations and communication games. To Renandya (2012), fluency in production takes more time than input-oriented tasks that encourage receptive comprehension.

Form-focused instruction (FFI) has attracted considerable attention over the past decades. Initially, it was seen as a method, but later it was regarded as a kind of exposure other than natural exposure. Still later, it was considered as a set of classroom processes. Nowadays, it is usually seen as a set of psycholinguistically motivated teaching options (Ellis, 2001). Long (1988) is of the opinion that nothing may be gained from trying to teach isolated linguistic structures in a systematic way and in accordance with a structure syllabus, an approach he characterizes as ‘focus on forms’. In contrast, Ellis (2005) uses FFI as a general term to refer to any instructional activity, whether planned or incidental, that is intended to draw learners’ attention to linguistic forms. It, thus, includes both traditional approaches in which forms are taught based on the structural syllabus and more communicative methods, in which forms receive attention only in activities that are mostly meaning-focused.

In the 1970s and 1980s, emphasis shifted to language development that takes place while learners are engaged in meaning-focused activities. It gave birth to a new theoretical view of SLA which was termed Communicative Language Teaching (CLT). This method included using tasks as a stimulus for generating interaction among students (Swain & Lapkin, 1998). In a meaning-focused approach, “learners are usually not specifically taught the strategies, maxims and organizational principles that govern communicative language use but are expected to work these out for themselves through extensive task engagement” (Celce-Murcia, Dörnyei, & Thurrell, 1997, p. 141).

It is generally believed that if learners are gently pushed to produce output, their learning of vocabulary can be improved in several ways. That is why Swain and Lapkin (1998) argue that output tasks can help learners to notice their linguistic shortcomings, leading them to modify their output. Nation (2002) refers to four major strands for vocabulary instruction programs, namely meaning-focused input, meaning-focused output, fluency development and language-focused instruction. The proponents of form-focused instruction, such as Laufer (2005) and Laufer and Girsai (2008), believe that focus on meaning alone is insufficient for vocabulary learning.  They claim that lack of attention to word forms sometimes causes trouble for learners. Different empirical studies such as Watanabe (1997), Ellis and He (1999) and De la Fuente (2002) have shown the effectiveness of form-focused tasks in comparison with meaning-focused tasks.

According to Schmitt (2008), the first step in vocabulary learning process is establishing form-meaning links.  There is a consensus among scholars that effective vocabulary instruction should go beyond stereotyped strategies such as helping learners to look up unknown words in dictionary (Feldman & Kinsella, 1996; Ooi & Kim-Seoh, 1996).  Many studies have experimented with different kinds of strategies and techniques for vocabulary teaching (e.g., Allen, 1983; Thornbury, 2002). Regarding these issues, Schmitt (2008) concluded that there is no right or best way to teach vocabulary; it all depends on the situation and factors such as the type of student, the target words, the school system and curriculum, and many other factors.

As to the effect of form-focused instruction on second/foreign language learning, several studies have been carried out. Montgomery and Eisenstein (1985) studied the effect of form-focused instruction on the improvement of adult ESL learners’ language proficiency. The experimental group, in addition to regular form-focused classes were also enrolled in a special oral communicative program (involving field trips to sites where they routinely needed to communicate in English), and the control group, took only the required grammar classes. The comparison of the learners’ performance on pretest and posttest in an oral interview indicated that both groups improved in grammar, vocabulary, comprehension and pronunciation. However, Ellis (1994) commented that one group received more overall instruction and, therefore, the improvement might have been the result of the total amount of instruction rather than the type of instruction.

Spada (1997) investigated the second language development of 48 intermediate level adults in three classes which varied in terms of the proportion of time spent on explicit grammar instruction. Group A received basically form-focused instruction in their speaking activities (e.g., grammar exercises); group B received both form-focused and meaning-focused instruction, and group C received meaning-focused instruction in their speaking activities (e.g., role play). Spada noted that learners who received more form-focused instruction performed as well or better on grammatical structures and on conversational skills (including grammar, vocabulary, pronunciation and fluency) than those who received less form-focused instruction. In a similar study, Lan and Wu (2013) compared form-focused and meaning-focused teaching in terms of their effectiveness on EFL learners’ pronunciation. They concluded that form-focused instruction was more effective than meaning-focused instruction.

De la Fuente (2006) investigated the effects of three vocabulary lessons (one traditional and two task-based) on the learning of basic meanings, forms and morphological aspects of Spanish words. The results showed no particular effect of pedagogical approach on the immediate retention of the target words. However, the teaching method had a significant effect on the long-term retention of the target items. In addition, task-based lessons turned out to be more effective compared with the traditional lesson. It was also found that task-based lessons that had an explicit focus-on-forms component were considerably more effective than those that did not. Jahangard (2010) carried out a similar study in the Iranian context and concluded that form-focused vocabulary instruction can have a statistically significant effect on the end-of-the-course achievement of EFL learners.

Shintani (2012) examined the potential effects of task-repetition with a group of Japanese children. Fifteen novice learners of English were asked to complete a communicative listening task designed to introduce new words. In a period of about five weeks, the same task was repeated nine times. Based on the results, task repetition caused changes in both the teacher’s input and students’ output. Not only did the learners’ comprehension improve, but also they showed a high level of motivation to engage in the task throughout the study.

Sarani and Sahebi (2012) investigated the effectiveness of task-based ESP vocabulary teaching with two groups of Persian literature students in Iran. A technical vocabulary test was given to the participants as a pre-test. The comparison group participants received the words in a conventional way, whereas the participants of the experimental group were taught the technical words based on the task-based approach. The results were indicative of the effectiveness of the task-based approach in comparison with the conventional method.

Maftoon and Haratmeh (2013) sought to investigate the effectiveness of involvement load and task orientation in task effectiveness. They came to the conclusion that both task involvement load and task orientation play a role in task effectiveness.

Zarei and Afrash Ab (2013) compared output-oriented and input-oriented tasks in terms of their effect on L2 vocabulary comprehension and production. Eighty intermediate-level adult Iranian EFL learners were selected and divided into four groups. There were four types of task (i.e., multiple-choice cloze task, gap-filling task, word formation task, and code-mixing task. The first two tasks were input-oriented while the second two tasks were output-oriented. At the end of the experimental period, the researchers administered two posttests. The result showed that output-oriented tasks were more effective than input-oriented tasks on both L2 vocabulary comprehension and production.

Spada, Barkaoui, Peters, So, and Valeo (2009) compared two types of form-focused instruction with regard to their effect on the development of L2 knowledge. The two types of instruction were both pre-emptive in nature, that is, planned and teacher-generated. In integrated form-focused instruction, attention was drawn to form only within communicative practice; on the other hand, in isolated form-focused instruction, focus on form happened separately from communicative practice. Data obtained from a written grammar test and an oral communication task showed that the two groups did not make differential progress over time. However, integrated form-focused instruction appeared to have certain advantages in the oral production task, whereas isolated form-focused instruction turned out to be more effective on the written grammar.

As the above review suggests, although there have been studies on the different types of task and their potential effect on L2 vocabulary learning, there is a paucity of research on the comparative effects of these different types of task on L2 vocabulary learning. And the few studies that have compared the effect of different types of task on vocabulary learning have often come up with conflicting results.

PURPOSE OF THE STUDY

The above review establishes the necessary theoretical background for the purpose of this study, which attempts to answer the following research questions:

1. Are there any significant differences among the effects of meaning-focused and form-focused input-oriented and output-oriented tasks on Iranian elementary level EFL learners’ vocabulary comprehension?
2. Are there any significant differences among the effects of meaning- focused and form-focused input-oriented and output-oriented tasks on Iranian elementary level EFL learners’ vocabulary recall?

METHOD

Participants

The participants were 120 elementary-level EFL learners of a private school in Tehran. They were all male and native speakers of Persian. The participants’ age range was between 14 and 16. Their homogeneity was checked through a Key English Test (KET). Other than availability, the main reason for the selection of the participants was that at the time of the study, all of the participants had at least five semesters of language learning experience at institutes; therefore, they were expected to be familiar with reading tasks.

Instrumentation

The following materials and data collection instruments were used in this study:

The KET (Key English Test) was used to homogenize the participants in terms of proficiency level.  Since it was not possible to administer the oral section of the test, only paper one of the test on reading and writing sections was administered. The sub-test that was used for the purpose of the present study included 40 items in six parts. Part one included 5 items in matching format; part two had 5 items in multiple-choice (three choice) format in which the participants were asked to choose one of the given alternatives to fill in the blank of a statement; part three contained 10 items in a conversational context. In five of the items, the participants were expected to choose one of the given alternatives in response to a given cue. In the other five items, the participants’ were required to match two sets of statements given in two columns to recreate a conversation between two people. Part four included a passage followed by seven multiple-choice items. Part five included a cloze passage in which certain blanks were provided and numbered, and at the end of the passage three alternatives were provided for each blank. The last part (part 6) contained five productive items. The participants had to read the descriptions of some objects and write that object in the blanks provided. The initial letter of the target word was also given.  KET has been used extensively in many different EFL and ESL contexts and its psychometric characteristics are well-established. Nonetheless, since only a sub-test of KET was used here, to check the reliability of the sub-test in the context of the present study, the KR-21 formula was used, and the reliability index turned out to be .79.

The pretest consisted of 150 items given to the learners before the treatments to measure their prior vocabulary knowledge. The test contained 150 sentences in each of which one of the target words was bold-faced and underlined. The participants were expected to the Persian equivalents of the target items. The purpose of the pretest was to make sure that the students had no familiarity with the target words before the treatment. The posttest package consisted of a 50-item multiple-choice test (to measure comprehension) and a 50-item fill-in-the blanks test (recall test). The comprehension test was administered immediately after implementing the tasks to test the learners’ vocabulary comprehension. The recall test was administered seven days after administering the comprehension test. To ensure the reliability of the posttests, the reliability indices of the posttests were estimated using the KR-21 formula. They turned out to be .87 and .78 for the comprehension and recall tests, respectively.

The instructional materials included Family and Friends 2 series textbooks (by Naomi Simmons from Oxford Publication), which were used as their course book. At the end of each session (about 30 minutes), an appropriate task was used as their supplementary material to teach vocabulary. The treatment lasted 12 sessions: 2 sessions a week, and each session lasted one and a half hour. To start the project, it was very important for the researchers to have tasks that could be used for teaching vocabulary. They made use of the following tasks.

Meaning-Focused Input-Oriented Vocabulary Tasks

These were meaningful tasks in which the lexical items were controlled and the students needed to understand the lexical items in order to successfully produce a correct response. The tasks were inspired by Tajeddin and Daraee (2013). They consisted of cloze tests each containing 5 new vocabulary items. Each question on a different part of the text contained one target word. The learners did not have to do anything with the word. The purpose of the task was to only draw learners’ attention to the sentence in which the word was used.

Meaning-Focused Output-Oriented Vocabulary Tasks

In these tasks, the students had to read and understand the text containing five blanks and then reconstruct the text using the equivalent word in parentheses. The text was selected from the learners' textbook. Here, the learner tried to produce his/her own words.

Form-Focused Input-Oriented Vocabulary Tasks

In form-focused input-oriented tasks, the vocabulary items were introduced through unreal situations by highlighting explicit attention to lexical items. The vocabulary items were taught by attracting learners' attention to different forms of language use, such as sentences or bilingual dictionary lookup. The target words were presented to learners using word-lists in which lexical items were presented in isolation along with their Persian equivalents, which is considered as ‘focus on forms’ instruction. The words were also presented through passages for translation (in which linguistic information was provided about the meaning of sentences in the passage); this could be considered as a type of form-focused instruction.

The gap-filling task, inspired by Zarei and Afrash Ab (2013), was used for this purpose. This task contained a text and a glossary. The text contained five gaps, and the glossary contained the five target words plus five additional words that acted as distractors. These distractors were added to decrease the possibility of wild guessing. The L1 meaning of all 10 words was provided in the glossary on a separate page. Here, the learner was not supposed to produce language.

Form-Focused Output-Oriented Vocabulary Task

These were mechanical tasks or controlled practice activities where students were required to produce a response without having to understand the language they were using. This task was adapted from Zarei and Afrash Ab (2013), and consisted of five target words. The participants were required to write five sentences with those words. Small spelling errors and grammatical problems were not of primary importance; rather the inclusion of the five target words, conveyance of the general idea, and clarity of sentences were used as the primary criteria for scoring.

Data Collection Procedure

Having selected an initial sample of 157 participants with the afore-mentioned characteristics through convenience sampling, the researchers administered the KET to ensure the homogeneity of the participants. Every correct answer was awarded one point and every incorrect answer was given 0 point. The total score of the test was 40. Those who scored between one standard deviation below and above the mean were selected as the participants of the study. 32 of the participants whose score was beyond this range or were absent on one or more of the exam sessions were excluded from all subsequent analyses. Five other participants were randomly excluded from statistical analysis to make the number of the participants in each group equal and to make the AVOVA design balanced. To check whether or not the participants were already familiar with the target words, the vocabulary pretest was administered. In the selection of the target words, care was taken subjectively to choose words which the participants were not expected to know. Still, the pretest results were used to consolidate the prediction. As a result of the pretest, those words that were unfamiliar to the participants were identified to be included in the posttests. Then, each group of participants was randomly assigned to one of the treatment conditions: meaning-focused input oriented group, meaning-focused output oriented group, form-focused input-oriented group and form-focused output-oriented group. Next, the treatment was given in 12 sessions.

The participants of the first group were instructed using meaning-focused input-oriented tasks. In this group, the participants were given a text with five numbered blanks. At the end of the text, there were four alternatives from which the participants chose one to fill each blank. In Group 2, the same words were presented through form-focused input-oriented tasks. The participants were provided with the same text that had blanks. This time, however, instead of choosing from among the given alternatives, the participants were provided with a word list in which the words were given along with their part of speech and meaning. This way, the participants’ attention was drawn to the lexical characteristics of the target items rather than the contextual meaning. Group three received the same instruction through meaning-focused output-oriented tasks. The participants read the same passage with the same blanks. This time, however, there was no selection involved. Instead, they had to fill the blanks with words of their own. Of course, in order not to allow for the possibility of the use of words that fitted the context without being the target words, the first letter of the target word and its Persian equivalent were given as cues in each blank. The last group received instruction through form-focused output-oriented tasks. The participants of this group were given a word list containing the target words along with information about each word such as part of speech and translation. They were asked to write a sentence with each word. An example of each type of task is given in the Appendix.

At the end of the treatment sessions, all the participants took the posttests. The obtained data were then summarized and submitted to statistical analysis.

Data Analysis

The collected data were analyzed using both descriptive and inferential statistics. Through descriptive statistics, the mean score and standard deviation of each group of participants were obtained. To answer each of the research questions, the scores of the participants on the tests of vocabulary comprehension and recall were compared using two separate two-way ANOVA procedures, having first made sure that the assumptions of the two-way ANOVA were met.

RESULTS

The First Research Question

The first research question sought to investigate the effects of meaning-focused and form-focused input-oriented and output-oriented tasks on vocabulary comprehension. To this end, the participants' scores on the vocabulary comprehension test were compared. Table 1 contains the result.

Table 1: Descriptive statistics for the two-way ANOVA on vocabulary comprehension

To see whether or not the observed differences reached statistical significance, a two-way ANOVA was used, the results of which are given in Table 2. Table 2 shows that task type has a significant effect on vocabulary comprehension, (F_{1, 116} = 97.67, p < .01), and that meaning-focused tasks are significantly more effective than form-focused tasks on vocabulary comprehension. At the same time, orientation is also a significant factor; input-oriented tasks are significantly more effective on vocabulary comprehension than output oriented tasks (F_{1, 116} = 35.00, p < .01).

However, the interaction effect between task type and task orientation is also significant, implying that the main effects are overshadowed. In other words, the main effects are somewhat washed away, and we cannot claim with certainty that the main factors are categorically effective. More specifically, the significant interaction effect implies that although output-oriented tasks are generally less effective than input-oriented ones on vocabulary comprehension, the difference between input-oriented and output-oriented tasks is far stronger in form-focused tasks. In other words, when learners’ attention is focused on form, productive activities are much less conducive to vocabulary comprehension than receptive activities. On the other hand, when learners are involved in meaningful communication, although input-oriented tasks are still more effective on comprehension than output-oriented tasks, the difference between the two is much less in comparison with form-focused tasks.

Table 2: Two-way ANOVA results on vocabulary comprehension

a. R squared = .57 (adjusted R squared = .49)

The Second Research Question

The second research question sought to compare the effects of meaning-focused and form-focused input-oriented and output-oriented tasks vocabulary recall. To this end, the participants' scores on the recall posttest were compared. Tables 3 and 4 contain the results.

 Table 4 shows that task type has a significant effect on vocabulary recall, (F_1,116 = 23.354, p < .01) and that meaning-focused tasks are significantly more effective than form-focused tasks. On the other hand, orientation is not a significant factor in L2 vocabulary recall (F_{1, 116} = .34, p > .05). Meanwhile, the interaction effect between task type and task orientation is statistically significant.

Table 3: Descriptive statistics for the two-way ANOVA on vocabulary recall

This means that among meaning-focused tasks, output-oriented tasks are more effective on vocabulary production than input-oriented tasks. However, when tasks are form-focused, input-oriented tasks are significantly more effective on vocabulary production than output-oriented ones.

Table 4: Two-way ANOVA results on vocabulary recall

a. R Squared = .37 (adjusted R Squared = .31)

DISCUSSION

The results of the present study showed that teaching vocabulary items through meaning-focused tasks would lead to better comprehension and recall. Despite the fact that form-focused tasks were also innovative for the participants of the study, meaning-focused ones provided better opportunities for successful vocabulary learning. This is in accord with that of Rashtchi and Keyvanfar (2012), who found that form-focused instruction in Iran is not fully effective on English language teaching. At the same time, this finding of the present study is in contrast with that of Tajeddin and Daraee (2013), who investigated pre-emptive and reactive FFI on intermediate Iranian EFL learners’ vocabulary learning. The results of their study indicated that although the differences between the two experimental groups were not statistically significant, FF techniques turned out to be slightly more effective.

The results of this study about vocabulary recall are in contrast with those of Khonamri and Roostaee (2014), who reported that there was no significant difference between form-focused and meaning-focused tasks in terms of their effect on EFL learners’ vocabulary recall. Nonetheless, these results lend support to those obtained by Saeidi, Zaferanieh, and Shatery (2012), who reported that learners’ scores in the focus-on-meaning group were significantly higher than the focus-on-forms group.

From a general perspective, unlike a number of previous studies, such as Watanabe (1997), Ellis and He (1999) and De la Fuente (2002), which have shown that form-focused tasks are more effective than meaning-focused tasks in promoting learners' language achievement, the results of this study showed that meaning-focused tasks were more effective than form-focused tasks on EFL learners' vocabulary comprehension and recall. This finding of the present study contradicts those of the above studies. This finding is also incongruent with that of Pishghadam, Shapoori and Shayesteh (2011), who showed that the form-focused instruction group (dictogloss task) significantly outperformed the other two groups on a collocation test. The finding is also incompatible with those of Tajeddin and Daraee (2013), who found that the retention of unfamiliar words was higher in form‐focused groups.

On the other hand, this study showed that task orientation is not a determining factor in task effectiveness. This is in contrast to Maftoon and Haratmeh (2013), who found that input/output-orientation of tasks is more effective than their involvement load.

The reason behind the superiority of meaning-focused tasks over form-focused tasks might be their nature. Meaning-focused tasks can be utilized to make EFL learners take part in deeper cognitive information processing and, therefore, expand their learning and deepen vocabulary retention. In fact, meaning-focused tasks are used to produce a cognitive structure for new information and to establish connections between learners’ prior knowledge and new information. Therefore, when this ideal condition is provided for learners, language learning happens, and these cognitive activities lead to better retention.

Another finding of this study that needs to be noted is the interaction affect. A significant interaction effect between task type and task orientation in vocabulary comprehension implies that when a task demands that learners attend to form, receptive tasks are much more conducive to vocabulary comprehension than productive ones. Such an interaction is somehow natural because the nature of input-oriented tasks is closely related to comprehension. When a task is input-oriented, it gives learners practice in processing input; the natural outcome of this processing is comprehension.

A more interesting interaction effect was observed in vocabulary recall. When it comes to meaning-focused tasks, output-oriented tasks are more effective on vocabulary recall than input-oriented tasks. However, when tasks are form-focused, input-oriented tasks are significantly more effective than output-oriented ones. The implication of this interaction is that in traditional contexts where learners are encouraged to focus on form, production relies heavily on the input that learners receive. That is because the input gives learners an explicit understanding of the nature of the target language item, in our case vocabulary. Form-focused output-oriented tasks (like the mechanical, decontextualized sentence construction) may give learners opportunities to produce words. However, since the practice is not meaning-focused, the result may simply be limited performance without much competence. In simpler terms, there may be no guarantee that transfer of training occurs here. On the other hand, when tasks are meaning-based, no explicit information is presented to learners about words. Words are simply used in meaningful communicative contexts. Since the focus is also on the meaningful communication, there is the danger that learners may not even notice how some words are used. Consequently, their productive knowledge of words will not improve. Output-oriented tasks, on the other hand, demand that learners use words in meaningful contexts. Therefore, even if they do not know how a word is used, the demand of the situation forces them to notice the gap in their knowledge and eventually learn to produce words appropriately.

As to the differences between the findings of this and other studies, there may be a number of factors. One of the factors that might have contributed to such differences could be the characteristics of the participants. This study was carried out with only male teenage elementary-level learners of English at private language institutes in an EFL context like Iran. As it was reviewed earlier, other studies were conducted with participants having characteristics that were different from those of this study (e.g., in terms of age, sex, level of proficiency, the context of learning and so on).

Another factor that might account for part of the differences between the results of this and other studies could be the study habits of the participants. In this study, attempt was made to engage the learners in each group only with a pre-selected type of task. However, we have to admit that there was no way of controlling what the learners did outside of the classroom context. Some of them might have practiced other tasks at home than those they were practiced inside the classroom. This is somewhat due to the educational culture of the research setting. In the context of Iran, teaching has traditionally been teacher-centred, and teachers have been almost the sole speaker in language classes. In such classes, student activity is not very welcome, and teachers even take their time to silence learners. As a result, learners are almost used to input-oriented tasks and may not feel comfortable with output-oriented tasks, especially when the focus of the task is on form rather than meaning. This explains why input-oriented tasks were more conducive to vocabulary comprehension with both form-focused and meaning-focused tasks, and more effective on vocabulary production with form-focused tasks.

CONCLUSION AND IMPLICATIONS

From the findings of the present study, it may be concluded that a carefully-designed instructional program in which vocabulary items are presented using the suitable type of instruction will be more effective than one in which a haphazard combination of tasks are used. Moreover, since there was a significant interaction effect between task type (meaning-focused versus form-focused) and task orientation (input-oriented versus output-oriented), one may conclude that one must be careful in generalizing the results. In other words, the significant effect of task orientation on vocabulary learning should not mislead us to conclude that one type is categorically better than the other type, and that the more effective type has to be generally preferred over the other one. In other words, the significant interaction effect leads one to the conclusion that the differential effectiveness of input-oriented and output-oriented tasks on receptive and productive vocabulary knowledge is moderated by whether the tasks are meaning-focused or form-focused.

It was also observed that, although input-oriented tasks were more effective than output-oriented ones in three out of four conditions, output-oriented tasks were more effective on vocabulary recall when they were meaning-based. From this, it may be concluded that while these findings support the role of input in vocabulary learning, they also lend support to the output hypothesis in meaningful, communicative and productive situations.

These findings may have theoretical as well as pedagogical implications for researchers, teachers, syllabus designers, and learners. On the theoretical front, these findings lend support to both input hypothesis and output hypothesis, suggesting that there are germs of truth in both hypotheses, and that blind adherence to either of these hypotheses may not be wise.  For EFL teachers, the results of this study can provide valuable opportunities to engage learners in vocabulary learning. In EFL contexts, since there is little exposure to language out of classroom, it seems necessary to provide conditions for learners to take maximum advantage of the class time. The knowledge of how each task type and task orientation affect vocabulary comprehension and recall can help learners make more informed decisions about their choice of the learning tasks. Such knowledge can also help learners to resist the temptation to stick to the security of routines and to use a preselected set of tasks simply because they are used to them. 

It may also be concluded that the knowledge of how task type can influence vocabulary learning may encourage syllabus designers to move toward more communicative, meaning-focused activities and course content and by so doing act as agents of change to encourage teachers and learners to engage more actively in meaning-based activities.