Patterns of Persian EFL Learners ’ Comprehension of Idiomatic Expressions : Reading Strategies and Cross-cultural Mappings in Focus

This paper primarily focused on the description of the results of a study conducted with sixty Iranian adult EFL learners to investigate how the reading strategies and pragmatic elements are likely to govern and characterize the comprehension and interpretation process of English idioms with and without contextualized reading. It also intended to determine the role of cultural mappings and the extent to which Iranian EFL learners' knowledge of cultural idioms is affected by their L1 when they try to construct their own meanings. The researchers came up with some interesting inferences about such theme-based patterns of idiomatic expressions through descriptive statistics and analysis of the participants' metacognitive comments in four phases of the study.


Introduction
Every language has phrases or sentences that cannot be understood literally, most of which have historical, philosophical, sociocultural, or even political origins.Even if we know the meanings of all words in a phrase and understand the grammar completely, the meaning of the phrase may still be confusing.As an important part of the language and culture, idioms reflect the transformation in conceptualization of the universe and the relationship between human beings and the universe.This colorful aspect of languages is used to communicate our thoughts and feelings, to give life and richness to the language by taking the existing words, combining them in a new sense, and creating new meanings, just like a work of art (Lennon, 1998).
There are many factors involved in translating or reproducing an idiom or a fixed expression into another language such as the availability of a similar meaning available in the target language, significance of the specific lexical items constituting the idiom, appropriateness of using the idiomatic language in a given register, etc.The bigger the gap between L1 and L2, the more difficult the transfer of meaning will be.That is why L2 learners often render a metaphorical expression in the L2 by using an analogous counterpart of their L1 due to their lack of awareness of metaphorical concepts and lexical strategies.
When it comes to differences between the language and culture of native speakers of Persian and those of English speakers, there are certain attitudes, ideas, culturally prescribed rules of behavior and certain ways of social interaction that seem to be accepted by most English speakers that are generally observed differently by Persian speakers.Even if those students of English as a Foreign Language (EFL) might show a high degree of verbal fluency in their discourse, they invariably seem to lack the conceptual appropriateness that typifies that of natives.They tend to speak or write with the formal structures of English but think in terms of their Persian conceptual system.In fact, perceiving and producing English idioms are among the most difficult areas for Iranian English learners and those learners/translators resort to different strategies to bridge up the gap.

Purpose of the Study
Having taught translation subjects, specifically the course, "The Application of Idioms and Metaphorical Expressions in Translation", for several years now, the researchers of the present study have experienced considerable difficulties in conveying figurative and cultural language concepts, especially the idiomatic expressions, to EFL learners.In the meantime, most reference materials on English idioms seem to be primarily intuition based, teacher-made and randomly selected.That is why they often include seldom-used idioms and incorrect descriptions of the meaning and use of some idioms, hence limiting their usefulness to non-native students.Also, despite the huge number of studies in different aspects of idioms, there have been a few studies of how cultural elements in idioms are dealt with.This area seems to have been the most neglected and underexplored aspect of EFL teaching particularly at Iranian language centers and universities.This research, therefore, addressed this gap and employed a hybrid experimental research design (reading strategies plus introspections and retrospections questions) in an effort to understand more fully the psychological plausibility of the participants' reading behavior; that is, the nature of the mental processing involved in the comprehension and interpretation of texts containing idioms.The study specifically investigated the processing of different types of idioms as matching idioms (Lexical-Level or LL Idioms) between L1 and L2 (idioms with complete overlap, i.e. the same metaphorical mapping as in English and/or Persian is present and the same or at least similar lexical items are used); partially-matching idioms (Semi-lexical Level or SLL Idioms) between L1 and L2 (idioms with partial overlap, i.e. the same mapping occurs but this is expressed using different lexical items) and non-matching idioms (Post-lexical Level or PLL Idioms) between L1 and L2 (idioms which show a different domain mapping altogether), under two broad categories of general and American vs.British culture-bound idioms.Therefore, four research questions were proposed as follows: Q1•Would context (or lack thereof) play any significant role in Iranian EFL learners' comprehension and interpretation of idioms?Q2•What kind of reading strategies do they employ in searching for an idiomatic phrase?Q3•What are some cultural or meaning constraints, if any at all, surrounding their comprehension and interpretation of English phrasal idioms when such idioms are contextualized?Q4•What is the role of transfer (either positive or negative) in perceiving or producing those idioms and which idiom subtypes are faster and easier to understand?

Background of the Study
Idioms make up a large proportion of any discourse, and the comprehension and production of them are the main parts of the studies of idiomaticity in both first and the second language literature.A number of studies conducted in the 1990s (e.g., Cronk and Schweigert 1992;Colombo 1993;Botelho da Silva and Cultler 1993;McGlone, Glucksberg, and Cacciari 1994) focused on idiom comprehension.Some scholars such as Cronk and Schweigert identified familiarity and literalness as measurable indications for the computation and representation of idiomatic meaning in the mental lexicon; others such as Botelho da Silva and Cutler studied the role of ill-formedness in idiom processing while the case of ambiguity and the relationship between context and different types of idioms was the main interest of others.(McGlone et al. 1994) According to Cooper (1999) four theories try to explain how native English speakers comprehend idioms: the first called Idiom-list hypothesis (Bobrow & Bell, 1973) which states a native speaker who encounters an idiom first interprets it literally.If a literal meaning doesn't fit the context in which the expression is situated, then he searches for the idiom in question in a special mental idiom lexicon and then chooses the figurative meaning.In the second model called lexical representation hypothesis (Swinney & Cutler,1979), idioms are considered to be long words that are stored in the mental lexicon along with all other words and both the literal and the figurative meanings of the expressions are processed simultaneously, which results in a "horse race" in which the context determines the more fitting interpretation.Third model, the direct access hypothesis (Gibbs, 1980(Gibbs, , 1984;;Schweigert, 1986), is an extension of the lexical representation hypothesis, for it posits that a native speaker rarely considers the literal meaning of an idiomatic expression but instead retrieves the figurative meaning directly from the mental lexicon.The fourth idiom-processing model, composition model (Gibbs, 1994;Tabossi & Zardon,1995), supersedes the three models described above.Giora's (1997) graded salience hypothesis is another hypothesis which came out of the discussion on idiom representation, processing, and comprehension, according to which, salient meanings of words or expressions are processed initially (if their lexicalized meaning can be retrieved directly from the mental lexicon rather than from the context) before less salient meanings are activated.Giora believed that metaphor and literal interpretations do not involve equivalent processes.The salient meanings of familiar and novel instances of metaphors, idioms, and irony are always accessed, and always initially, regardless of context, even rich and supportive contexts; that is, metaphor, idiom, and irony interpretation involves processing the literal meaning (see also Giora, 1999;Giora & Fein, 1999).In the meantime, other studies (e.g., Arnold and Hornett 1990;Levorato andCacciari 1992, 1995;Nippold and Rudzinski 1993;Titone 1994) have looked for relationships between idioms and amount of exposure and use, cognitive strategies and idiom comprehension/production, and age and awareness of semantic links and cognitive abilities.For instance, Nippold and Rudzinski (1993) like Arnold and Hornett (1990) and Levorato andCacciari (1992, 1995), found that familiarity, idiom transparency, and idiom performance gradually improved as subjects' age increased.Similarly, Titone (1994), found familiarity, compositionality, predictability, and literality to be important dimensions in the processing of L1 idioms.It is unfortunate that this intensity of research into L1 idiom processing and comprehension has not been matched by an equal level of interest among SLA researchers.However, Irujo (1986) utilized recognition and production tests to demonstrate that advanced students of English rely on knowledge of their native Spanish in order to comprehend and produce L2 idioms.She further observed that learners find those idioms which are identical in both L1 and L2 the easiest to comprehend and produce.Idioms which are similar in L1 and L2 present learners with only somewhat more difficulty, although production tests reveal interference from Spanish.Those idioms which are completely different in L1 and L2, however, prove the most difficult for learners to comprehend and produce, with almost no positive or negative transfer between the two languages.Liontas (2001), attempting to obtain information about the ways in which L2 learners process, comprehend, and interpret idiomatic expressions both in and out of context, found that idiom comprehension performance in Modern Greek significantly improves if contextual information is present.One key finding that emerged from the Liontas (1997Liontas ( , 2001) ) studies is that knowledge and understanding of vocabulary is directly linked to idiom performance regardless of whether contextual support is provided to learners.Boers and Demecheleer (2001) also draws attention to the cultural aspects of teaching idioms.They claim that the possible impact of cross-cultural variation on learners' interpreting idioms invites language teachers to give extra attention to figurative expression in the target language that relate to metaphoric themes that are less salient in the native language.Furthermore, they believe that an approach to teaching idioms will benefit from a teacher's awareness of cross-cultural as well as cross-linguistics differences.
In fact, several researchers have suggested that L2 learners, unlike L1 learners, appear to have considerable difficulties comprehending and producing idioms accurately (Cooper, 1998;Irujo, 1986Irujo, , 1993)).Sadeghilar (1993) focused on the application of translation in the process of learning idioms and found that identical idioms in both English and Persian would show positive transfer since they are the easiest to be comprehended and produced correctly.Similar idioms would show negative transfer and their comprehension is as high as identical idioms, but their productions reflect interference from Persian.Different idioms would show neither positive transfer nor negative one.Their comprehension and production are lower than those of other two types.Sadeghi (1995) also suggested that lexico-semantic based approach in teaching English idioms would lead to much better performance and comprehension for the Iranian EFL learners.
Equally important to note here is that no empirical study to date has investigated per se the reading behavior of Persian L2 learners-a common practice in many L1 psycholinguistic studies on idiomaticity, detecting and understanding idioms embedded in natural texts.
Based on the findings of the above-mentioned studies and with regard to the research questions of the present study stated above, the following null-hypotheses were formulated to be tested in this research: Ho1.Context doesn't play any significant role in Iranian EFL learners' comprehension and interpretation of idioms.
Ho2. Iranian EFL learners don't employ any specific reading strategies in searching for an idiomatic phrase.
Ho3.There are no cultural or meaning constraints surrounding learners' comprehension and interpretation of English idioms.
Ho4. Transfer (either positive or negative) doesn't have any role in perceiving or producing the idioms.

Methodology
In order to trace students' meaning-assigning processes within the framework of cognitive linguistics which suggests that language units reveal how the mind works, participants were asked to engage in some reading tasks ranging from detecting idiom to full-context interpretation (phases 2, 3, and 4).[Insert Table 1 here] The participants were also asked to describe as thoughtfully as possible their affective state during task performance.Prior to conducting these experiments, native speakers of English were asked to rate frequently used phrasal idioms.In this study, the term "concept" was used as a mental structure representing the knowledge about the particular traits of reality gained as a result of the process of cognition to describe the pragmatic or semantic structure of idioms.Also for the purposes of this study, the term "context" was defined as a paragraph-length narrative text (five to ten sentences long) or a two-person dialogue (two to seven interactional exchanges) written with the natural tone of the target language in mind; idioms were therefore presented in contexts that clarify their actual meanings and uses in everyday speech.

Participants
The subjects of this study consisted of 60 Iranian EFL learners studying English at the Islamic Azad University, Takestan Branch, during summer 2009.They were selected from among 135 participants.The entire class ranged between 20 and 23 years of age.

Instruments and Materials
Two testing instruments were first developed and used by the researchers in this study.At first, the Michigan Test of English Language proficiency (MTELP) was administered to determine students' level of proficiency.Then a standardized test (including 60 multiple-choice items and a translation task) on idioms as the pre-test to determine the idiomatic knowledge of the participants was given to the participants who had been selected.Some Iranian students' common translation/interference problems with English idioms and their equivalents were deliberately included in the test to determine the possible effects that students' first language might create in their development of idiomatic knowledge or competence.Two experts in TEFL (Ph.D. colleagues) were also consulted for the accuracy and appropriateness of the translated texts.
Then according to the mean performance and the standard deviation of the tests, subjects were assigned into high, mid and low groups.The students were homogeneously sorted so that the initial study groups represented three groups of 20s in three levels.Those students whose scores fell one standard deviation below and above the mean were assigned as the mid group.Those subjects whose scores were two standard deviations below and above the mean were classified as the low and high groups respectively.To protect all students' anonymity throughout the study, researchers applied S initial (which stands for every Student) and a number given to every participant indicating his/her order in each group as Low: S1-S20, Mid: S21-S40, High: S41-60.Some pre-selected works of contemporary English literature and texts of daily conversations were chosen as the base material for this experiment.

Procedures
The researchers went through four phases to conduct the present study.

Phase 1
Phase1 included a selection of a wide corpus of English idiomatic expressions by the native speakers of Modern English which is believed to be frequently used in everyday communication.One thousand and one hundred English idiomatic phrases were taken from authentic texts and materials.Three hundred items of the total idioms were focused just on the recognition and frequency rate of American and British Culture-Bound idioms (150 items for each).In fact, the meaning which lies behind this kind of expressions is always strongly linked to the specific cultural context where the text originates or with the cultural context it aims to re-create.
Those idioms were submitted during two online sessions to 20 English native speakers (10 American and 10 British -five males and five females in each) who live and study in Washington and London respectively, with the request to perform a Familiarity and Frequency Rating.The ages of the participants ranged from 18 to 35 years, with 15 (75%) between the ages of 18 and 24 years.The average age for the entire language sample was 23.25 years.The majority of the participants (65%) classified the family in which they grew up as middle class.25% of participants came from lower middle class families, and 10% from upper middle class families.35% of the sample held sophomore standing at the university, with the remaining participants divided into 20% seniors and 10% juniors.Only 15% of participants held freshman standing and the remaining 20% held graduate standing.
They rated their familiarity with the idioms using a scale that ranged from 5 ("very frequently used") to 1 ("don't know the idiom") as follows: 3) I may have heard this expression, but I don't really know what it means.
2) I may have heard this expression, but I do not know what it means.1) I have never heard this expression before.
The respondents were asked then to estimate the frequency use/rate of a given idiom in English language and culture by responding to the following statement:

I use this expression very often (5), often (4), sometimes (3), rarely (2), and never (1).
This rank-ordered procedure yielded for each idiom an average "subjective frequency" or "familiarity" rating obtained from all participants.The ratings provided a number of details about the frequency of an idiom in the English language and culture, which were then used as the basis for the following three experiments.Only those 48 idioms (i.e., sixteen percent of 300 culture-bound idioms and almost 4.36 % of the total base material) and 85 idioms (i.e., almost eleven percent of 800 idioms and about 7.73% of the total base material) that obtained a "very frequently used" rating of 5 were included for use in the subsequent analysis.To make the job easier, the former was called "cultural idioms" and the latter as "general idioms" respectively.[Insert Tables 2& 3here] Although the remaining eight hundred items had also some cultural elements, there was no significant difference of American or British English intended therein and the philosophy behind those questions was simply to rate the (non culture-bound) idioms used by all native speakers.That is because some idioms can be understood without the common cultural background, experience, reference or specific knowledge.Those idioms are widespread in other English varieties and even other languages and thus can be interpreted and understood more easily with the process of translation.

Phase 2
Phase2 called Idiom Detection Task or IDT, investigated how L2 learners identify and understand idioms in authentic texts of literature and what contextual reading cues and communication principles they used in searching for an idiomatic phrase.
Six paragraphs and two dialogs plus twenty multiple choice questions containing 45 idiomatic expressions and 20 culture-bound (10 American and 10 British) idioms, chosen at random from the list of 85 idioms and 48 culture-bound idioms rated '5' would constitute the material of this experiment and were printed on a 6-page booklet with appropriate space for participants to provide an answer.The length of the materials varied from five to eight sentences in length and from two to three interactional/dialog exchanges to ensure randomization of the material tested in a 70-minute session.Each participant was presented with a practice trial containing 5 items not used in the actual experiment, followed by the experimental items.Participants reported feeling comfortable with the identification procedure after this much practice.Following the practice session, each participant was instructed to answer the questions and underline within the body of the text the words or entire sentence that constituted the idiomatic expression in each text.Once the participants reached a decision, they were requested to explain why they thought the underlined words were an idiomatic phrase.[Insert Table 4 here]

Phase 3
Phase3 tested the assumption that for second language learners' analysis of an idiomatic expression, when presented without a supporting context, is an obligatory, automatic process.If an idiomatic expression is recognized at an early point because of the one-to-one translation match of single lexemes between the Persian and English idiom, as the proposed Idiom-matching Hypothesis predicts, further analysis of the L2 idiomatic phrase would be, in Idiom-matching Hypothesis principle, no longer necessary in order to understand the metaphoric meaning of the phrase.Conversely, a non-matching phrasal idiom could be understood, if at all, only after a full linguistic analysis, i.e., a combined phonological, syntactic, and semantic analysis.If L1 idiomatic phrases were stored in memory as multiword lexical units, then the analysis of the phrasal idioms should not be necessary for comprehending the idiom.
Forty entries of general idioms plus ten culture-bound idiomatic expressions (5 American and 5 British in each) constituted the 50 phrasal idioms as the material of the present Zero Context Task (ZCT) phase.The researchers randomly selected ten phrasal idioms with their word-for-word counterparts in Persian and forty non-matching phrasal idioms from the remaining idioms rated '5' for this phase.The same original group of sixty students at the same university took part in this phase.All phrasal idioms were presented in their zero context (i.e., without supporting context) followed by three lines in which participants were asked to first define the meaning of each idioms before providing their equivalent Persian idioms.The 2-point scale as (1) correctly defined and (2) incorrectly defined were used for the evaluation of the idiomatic definitions given.[Insert Table 5 here]

Phase 4
Phase4, also called FCT dealing with understanding general and cultural idioms in full context, was focused on whether context would affect positively the way learners of English understand idiomatic expression when such idioms were presented with the context that supports their interpretation.It is hypothesized that when the context constrains the idiomatic meaning, a non-matching idiomatic expression is understood considerably better than when it is isolated from its surrounding context, but is still more difficult to interpret than a matching expression even in the presence of context with text cues supporting the idiomatic meaning.
Fifty phrasal idioms including 40 general idioms (10 matching, 30 non-matching phrasal idioms as indexed by the results of Phase 3 above) and 10 culture-bound idioms would constitute the material of the (FCT) phase.Each item containing the idiomatic expression was made up of either short paragraphs three to six sentences in length or short dialogs consisting of one to three interactional exchanges.Seven paragraphs and three dialogs containing the idioms reported in Phase3 were then randomly selected.The order of presentation of the experimental items was again varied to ensure randomization of material.Phase 4 used the same participants and phrasal idioms as in Phase 3 and was conducted immediately following the conclusion of Phase3.However, whereas in Phase 3 phrasal idioms were presented without supporting context, in Phase 4 all phrasal idioms were given in their full context, followed by three lines in which participants were asked to define a new meaning of each of the 40 phrasal idioms.Again, the new definitions offered were evaluated on a 2-point scale as (1) correctly defined or (2) incorrectly defined.The data was summarized in related tables and graphically in Figures.[Insert Table 6 here]

Phase1
The participants' selections were evaluated first on the basis of their correctness, which reached the total average of fifty four percent for general idioms (Low: 46.66, Mid: 55.55, High: 60) and about eight percent for the culture-bound idioms (Low: 4.25, Mid: 8.5, High: 10).The retrospective protocols were then classified on the basis of the recurring thematic units present in the responses of the participants, resulting in a number of conceptual categories created and then re-translated into English by the researchers (and tested for reliability and validity considerations by another native-speaker rater) while sorting the responses into units of information.Once their retrospective protocols were operationally defined, these units were analyzed, pooled, classified into general categories and given one single representative label and quantified in percentage terms of each category against all responses in a descending order.[Insert Table 7 here]

Phase 2
The number of idioms detected correctly ranged from a low of 13 to a high of 40 in general idioms and 1 and 7 for the cultural idioms respectively.The group average of all idioms detected was 24 out of 45 idioms (Low: 21, Mid: 25, High: 27) or about fifty three percent of all idiomatic items.Individually their performance ranged from a low of 29 percent to a high of 89 percent for the general and 5 to 35 for the cultural idioms.The second point worth noting based on the data is that while there is variability among learners of English in detecting idioms in authentic texts of English literature, the majority of them employed predominately the literal meaning of the idiom (19.1, 21.5 & 22 percent), their translation skills from L2 to L1 and vice versa (17.20, 16 & 17 percent), the syntactic and semantic arrangement of the lexical unit (14.5, 14 & 9 percent), their prior knowledge of the functions of idioms in context (11.15, 11 & 8 percent), and, finally, the context surrounding the phrasal idiom (4,7 & 7 percent) as their main guide in reaching a decision.Of greatest interest in the present context is the observation that these five reading strategies alone account for 66 percent of all reader activity (Low: 65.95, Mid: 69.5 & High: 63) than all other remaining strategies listed.Also, lack of a familiarity with vocabulary was a major obstacle in the accomplishment of the IDT.The fact that almost half of phrasal idioms were identified successfully (24 out of 45) suggests that learners of English are capable of using many kinds of reading strategies: forward inferencing; schema accommodation, assimilation, and adaptation; process of elimination; and contextual lexical, grammatical, and syntactic cues to name but the most important ones, in order to describe strategies such as improbability, literal translation, word arrangement and placement in text, context, and fantastic/metaphorical images.[Insert Figure1 here]

Phase 3
Of the 40 general idioms here examined, 25 idioms (62.5 percent) on average (Low: 29, Mid: 24 & High: 22) were incorrectly defined.The number of phrasal idioms incorrectly defined ranged from a low of 17(42.5 percent) to a high of 35 (87.5 percent), whereas those that were correctly defined ranged from a low of 5 (12.5 percent) to a high of 25 (62.5 percent), resulting in an overall success rate of 37 percent (Low: 27,Mid: 40 & High: 45).The ZCT data therefore clearly suggest that idiom performance may be dependent on idiom subtype; that is, on the conceptual-semantic image distance (i.e., the degree of opacity) between Persian and English idioms.In addition, it further underscores the claim that idiom type affects the speed and ease of idiom comprehension and interpretation.But what these results most strongly indicate, however, is that the understanding of all idiom types, particularly PLL idioms, is significantly hampered by lack of context.One can realize at first look that the learner variability and the difference of correct definitions versus incorrect definitions is a negative one (Low:-46, Mid:-20& High:-10).However, upon closer scrutiny within the two idiom phrasal types-matching idioms and non-matching idioms-it emerges that matching idioms were correctly defined more than four times as often for Low, about three times for Mid, and about four times for High; while nearly all of the matching idioms were correctly defined (74.16 percent), out of the 10 non-matching idioms only 21.66 percent were found to be correct.Also, the number of phrasal idioms correctly defined in each idiom type ranged from a low of 3 to a high of 10 and 1 and 15 respectively.These findings suggest that non-matching phrasal idioms are processed much difficult than matching phrasal idioms.

Phase 4
Of the 40 phrasal idioms examined in context, only 18 idioms or 46 percent on average (Low:57.5,Mid: 42.5, High:38) were incorrectly defined or 21 percent difference in decrease of error interpretation from the previous ZCT.The number of phrasal idioms incorrectly defined ranged from a low of 9 (22.5 percent) to a high of 32 (80 percent), whereas those that were correctly defined ranged from a low of 8 (20 percent) to a high of 31 (77.5 percent), resulting in an overall success rate of 54 percent or 22 out of 40 phrasal idioms (Low: 42.5, Mid: 57.5, High: 62).Individually, participants' performance ranged from a low of 20 percent to a high of 77.5 percent.Upon closer inspection of the FCT idiom type, it emerges again that the participants correctly defined more matching idioms (90 percent or 9 out of 10 idioms-Low: 83.5, Mid: 91.5, High: 95.5) than non-matching idioms (42 percent or 13 out of 30 idioms-Low: 29, Mid: 45, High: 51), a result which is consistent with the Idiom-matching Hypothesis posited at the outset of this study.Without exception, all participants showed considerable progress from the previous ZCT experiment in the comprehension and interpretation of English phrasal idioms.The increase in idiom performance (i.e., comprehension gain), when compared to the gain achieved in the ZCT, led to an overall increase in group performance of 17 percent for the general idioms 29 percent for culture-bound idioms.This increase was more pronounced with the non-matching idioms (20 percent) than with the matching idioms (16 percent) given the fact that the participants had already achieved a high level of success with such idioms in the ZCT.Regarding the success rate of those correctly defined cultural idioms in phase 4, it was also observed that there was 29 percent increase than phase 3. [Insert Table 8 here]

Discussion
Based on the results of the study mentioned above, the data provided convincing evidence that lack of context exerts a strong negative effect upon the accuracy of idiom interpretation by L2 learners.SLL and PLL idioms do have stronger impact than LL idioms.Indeed, performance increased by 17 and 29 percent in the context condition for the general idioms and the culture-bound idioms respectively (i.e., from 37 in phase 3 to 54 percent in phase 4 for the general idioms and from 12 in phase 3 to 41 percent in phase 4 for the culture-bound idioms).Moreover, it was shown that idiom interpretation is seriously distorted if there is a lack of context for both matching and non-matching idioms.While idiom interpretation was considerably higher for highly matching idioms (i.e., identical idioms in L1 and L2), 90.16 percent, non-matching idioms (i.e., completely different idioms in L1 and L2) caused learners considerable difficulty.Performance in these idioms only reached 41.66 percent, clearly suggesting that lack of context impacts idiom understanding.However, with the introduction of context, even the non-matching idioms saw an increase of 20 percent in performance accuracy.[Insert Figure 2 here] The results of phase 2, coupled with the many valuable insights gained from the participants' retrospective accounts and post-task evaluations, provided ample evidence that L2 learners do compute literal and idiomatic meanings separately, yielding two alternative interpretations.Only after the literal interpretation has been considered and rejected, the "idiomatic" sense becomes available.Beyond that, the insightful answers participants offered in this experiment show the many challenges learners of English encounter when attempting to capture the meaning of a phrasal idiom as well as their affective state of learning during the process of idiom identification and understanding.Also, their retrospective protocols revealed that, in fact, they are much more sophisticated linguistically and strategically than surmised by the instructors.

Conclusion
The findings indicate that context plays an important role in the construction of idiomatic meaning and Identical and Different idioms in both context and non-context treatments, compared with those of similar idioms, enjoy the main effects in this regard.Also, different strategies are employed by language learners; while lack of tacit knowledge of idioms is a major cause for their failure in comprehending the idiomatic meanings.At the same time, the degree of opacity between target and domain idioms influences the speed and ease of idiom understanding and treating idiomatic expressions just as sets of separate haphazard lexical items, independent of human conceptual and cultural system would lead to negative transfer.In fact, all statistical evidence nullifies our Null Hypotheses for alternative hypotheses; therefore, the researchers can safely conclude that the null hypotheses are false.
In light of the findings of phase 2, one final general question of interest concerns whether the same pattern of idiom strategizing is also observed in participants from a variety of second languages before it can be concluded definitively that there is a universal modus operandi in identifying and understanding phrasal idioms.
Furthermore, since culture-bound expressions are unique to any language, they require an adequate cultural awareness of both source and target languages (e.g.American/British English and Persian in our study), and they cannot be understood just from the meaning of their individual words.At the same time, in most cases, High level participants performed proportionally better than Mid and Low levels and they all showed better results with American (than British) culture-bound idioms.
5) I have heard this expression, and I know what it means.4) I have heard this expression, and I have an idea of what it means.

Table 1 .
Sample Exercises of Some English Idioms

m all right, Jack. PLL 10 Strike while the iron is hot SLL 11 Sell the family silver SLL 11 Blessing in disguise SLL 12 The gnomes of Zurich PLL 12 Look over somebody's shoulders SLL 13 Live the life Riley PLL 13 Born with a silver spoon in one's mouth PLL 14 The corridors of power SLL 14 Put the cart before the horse SLL 15 Spit and sawdust PLL 15 Fall between two stools PLL 16 The jewel in the crown SLL 16 Not one's cup of tea PLL 17 Break the mould LL 17 In one's salad days PLL 18 Doom and gloom PLL 18 Get the show on the road PLL 19 Dear old Blighty PLL 19 Throw caution to the wind PLL 20 The ghost in the machine PLL 20 Paint the town red PLL 21 Everything but the kitchen sink PLL 21 Put someone in the picture SLL 22 (All) Lombard Street to a China orange PLL 22 Have an ax to grind (with someone) PLL 23 B i g g i r l ' s b l o u s e PLL 23 At sixes and sevens LL 24 Bob is your uncle PLL 24 Skate on thin ice SLL
Vol. 6, No. 8; August 2010Published Canadian Center of Science and Education 93

Table 7 .
Sample Idioms, Representative Strategies Used and Metacognitive Comments in the IDT paraphrasing the idiom without giving an interpretation, elimination process, the texts' surface codes, giving different feedbacks such as being frustrated or confused, or other strategies, etc)