Considering pragmatic competence as a key to communication competence (Bachman, 1990) reflects the significance of formulaic sequences(FSs) in pragmatics and second language learning (Wood, 2015).FSs consist of any sequence of words that are supposed to be stored as a whole in the long-term memory and mostly retrieved automatically or without much processing at the time of application (Wray, 2000;2002).FSs for their processing advantages play a significant role in first and second language acquisition and learning (Richards & Schmidt, 1985; Wood, 2002; 2010). FSs play an important role in social contacts and the development of L2 pragmatic competence, and also accomplish specific pragmatic goals (Bardovi-Harlig, 2012). FSs have four functions in language use, namely: (a) functional use, (b) social interaction, (c) discourse organization, and (d) precise information transfer (Schmitt, 2006). 

Bardovi-Harlig (2006) claimed that the term formula describes at least two different types of formulas, namely: acquisitional and social so that the first one is acquired at the early stages of acquisition and usually cannot be analyzed by the interlanguage grammar (Bardovi-Harlig, 2012; Schmidt, 1983); whilst, social formulas are applied in a particular social context (Bardovi-Harlig, 2009).

According to Coulmas (2011), “in embodying societal knowledge FSs are essential in the handling of day-to-day situations (p.4)”. FSs also referred to as automatic speech “localized in both the right and left hemispheres” (Krashen, 1981, p. 84). They are often retained after left hemispherectomy which is a very rare surgical procedure. In this surgery, half of the brain is removed to treat a variety of seizure disorders as in the case of non-fluent syntactic aphasia in which the person is not able to make a syntactically correct sentence (Jones, 1986).

FSs are often applied to meet recurrent communication requirements (Conklin & Schmitt, 2008). These recurrent communicative requirements usually have conventionalized language expressions attached to them such as, “I'm sorry to hear about…to” and “I'd be happy to…..” for expressing sympathy and complying with a request respectively (Nattinger & DeCarrico, 1992, pp.62- 3). 

For the comprehension and production of FSs, in addition to pragmalinguistic and sociopragmatic knowledge, a learner should be aware of the cultural aspects of these sequences because some types of FSs are culturally specific (Taguchi, 2009). Furthermore, because there is an intrinsic connection between native-like competence and native culture, it is difficult for L2 learners to realize idiomatic expressions and thus they may not make the native-like selection and native-like performance to prevent communication failure ((Pawley &Syder, 1983; Wood, 2015; Wray, 2002). Based on the discussion above, the researchers of the current study believe that more investigations are required into this neglected issue in language studies from different aspects because FSs constitute a considerable part of the adult native speaker’s pragmatic competence and acquiring a great deal of these sequences to efficiently cope with developing social situations and discourse requirements seems necessary (Coulmas, 2011).

There is a consensus that despite the communicational value of FSs, they are difficult for L2 learners to be proficient in (Bardovi-Harlig & Vellenga, 2012; Moon, 1992; Scarcella, 1979; Yorio, 1989). Research has shown that L2 learners of high linguistic and grammatical proficiency may not necessarily have pragmatic proficiency as one of the elements involved in communicational interactions (Kecskes, 2000).Previous research revealed that many factors are involved in the acquisition and appropriate use of FSs as well as pragmatic units of the target language. These factors include L1 culture transfer, individual choice, false generalization, motivation, language aptitude, sociocultural adaptation (Dӧrnyeiet al., 2004; Kecskes, 2000), level of proficiency, and study-abroad experience (Staples et al., 2013; Taguchi, 2013). Furthermore, L2 learners do not choose those pragmatic units which are culture-specific (Kecskes, 2000).

Research on all types of FSs has been both cross-sectional and developmental. Some studies have considered the processes and factors involved in the production and comprehension of FSs (Faerch, 1989; Faerch & Kasper, 1984; Kasper, 1984; Kecskes, 2000; 2012; Rafieyan et al., 2013; 2015; 2018; Taguchi, 2005; Taguchi, 2007a; 2007b).Other studies have investigated the acquisition and developmental processes of pragmatics (Kasper & Schmidt, 1996; Schmidt, 1992), and some of the studies have investigated on FSs from a clinical perspective (Sidtis et al., 2009). Some scholars have investigated the role of mental processing onthe production of discourse and pragmatics (Schmidt, 1990), the effect of L1, and the level of instruction or other factors on the use of FSs (Bardovi-Harlig, 2008; Bardovi-harilig&Vellenga, 2012; Bardovi-Harlig et al., 2008; Dӧrnyei et al., 2004; Matsumura, 2003; Rafieyan et al., 2014a; 2014b; Salemi et al., 2012; Staples et al., 2013; Taguchi, 2012; 2013; Tajeddin & Momenian, 2012), and cultural distances (Kecskes,2012).

However, based on the literature reviewed, research on the role of cultural intelligence and the level of proficiency concerning the use of conventional expressions as one of the types of FSs, especially in the case of production of these expressions, is rare. Conventional expressions which are the focus of the current study consist of sequences such as I'm very sorry versus I’m feeling apologetic toward you” (Schmitt, 2013, p.2), and also “no problem”, “nice to meet you”, and “that'd be great” which native speakers use in specific contexts (Bardovi-Harlig, 2007, p.756). Conventional expressions play an important role in social contacts and the development of L2 pragmatic competence, and also accomplish specific pragmatic goals (Bardovi-Harlig,2012).

Multi-perspective intelligence, proposed by Detterman and Sternberg (1986), is an increasingly adopted approach. It explains various aspects of human learning. Based on this approach, intelligence is a multifaceted concept that has metacognitive, motivational, and behavioral aspects. Therefore, intelligence research should not focus merely on the cognitive aspect (Earley & Ang, 2003). Considering this approach, Earley and Ang (2003) examined the cultural aspect of intelligence and developed a conceptual model of cultural intelligence named (CQ). CQ is composed of four components: cognitive, metacognitive, motivational, and behavioral functions in a culturally diverse setting (Ang& Van Dyne, 2008; Ang et al., 2006). CQ, as an important aptitude that can be developed in competent people who are psychologically healthy and professionally experienced (Earley & Mosakowski, 2004), is thereby defined as the capability to function efficiently in culturally diverse environments or the ability to adapt to a new culture setting (Earley & Ang, 2003). This definition is consistent with general intelligence known as (IQ) which is “the ability to grasp and reason correctly with abstractions (concepts) and solve problems” (Schmidt & Hunter, 2000, p.3). Therefore, CQ is a complementary form of general intelligence (IQ), like emotional intelligence (EQ); however, CQ is a culture-free concept, not a culture-bound one as EQ is (Ang& Van Dyne, 2008; Earley & Ang, 2003; Earley & Gibson, 2002).

Among the four components, cognitive CQ stands for an individual’s awareness of cultural universals and cultural differences which are based on general knowledge of norms, conventions, and practices in different cultures received via experience and education (Ang&Van Dyne, 2008; Ang et al., 2006; Ng, Dyne &Ang, 2009; Van Dyne et al., 2009). Metacognitive CQ is referred to the cultural knowledge that a person consciously acquires and uses for social interactions in new cultural environments. While cognitive CQ relies on culturally bound thinking, metacognitive CQ requires reflective thinking about social interactions in different cultural settings. Therefore, metacogitive CQ is a conscious reflection of cultural differences before and during cross-cultural interactions, referred to as the conscious acquisition of cultural knowledge and control over thoughts and acts in culturally diverse settings (Flavell, 1979). For instance, after consciously observing Asians᾽' communication style, a western business executive with high metacognitive CQ would know when it is time to speak up and behave appropriately when meeting Asians (Ang & Van Dyne, 2008).

The third component, motivational CQ refers to an individual’s interest to direct energy and attention to learning about intercultural environments and functioning in different cultural settings (Ang, Van Dyne & Koh, 2006). For example, a Chinese person who can speak Japanese and likes communicating with those from other cultures would not hesitate to interact or communicate with a person from Japan (Ang& Van Dyne, 2008). Such examples show the role of motivational CQ in the comprehension of cross-cultural adaptation (Templer et al., 2006).

On the fourth and last component, behavioral CQ refers to the exhibited culturally appropriate verbal and nonverbal behaviors when interacting with people from different cultures (Gudykunst et al., 1988; Hall& Hall, 1959; Ng et al., 2009). According to Hall and Hall (1959), motivation for cultural understanding should be along with the capability to show the appropriate verbal and nonverbal behaviors. When an individual initiates and keeps face-to-face communication, he or she has no idea about the other person’s latent feelings or thoughts. However, that individual can pay attention to what he or she hears and sees in the other person’s facial and vocal expressions.

The role of cultural intelligence in individual success in cross-cultural interactions has been widely recognized (Gardner & Lambert, 1972; Kitao, 1991). Earley (2002) remarked that intercultural misinterpretation is common and has great effects on institutions. To understand people from different cultures and to have effective intercultural comprehension, CQ should be taken into consideration as another construct besides emotional intelligence (Salovey & Mayer, 1990), social intelligence (Thorndike, 1920), and practical intelligence (Sternberg et al., 1981).

However, the role of CQ in second language learning has yet to be sufficiently explained (Ghonsooly & Shalchy, 2013). CQ is likely to influence the pragmatic production (Earley&Mosakowski, 2004) as well as the production of FSs(Wood, 2002) because language learning or performance is highly affected by both world knowledge and cultural information (Ghonsooly et al., 2013). Although there are some studies on the relationship between cultural intelligence and different aspects of teaching and learning the English language (e.g.,Ghonsooly & Golparvar, 2012; Ghonsooly & Shalchy, 2013; Ghonsooly et al., 2013; Khodadady&Yazdi, 2014;Nasiri&Ghadiri, 2016; Rafie et al., 2016; Tajeddin & Momenian, 2012), research on CQ is necessary to provide sufficient insights into why some people can make an effective intercultural communication when communicating in culturally different settings (Ang et al., 1990). As such, the present research aimed at investigating the role of CQ as a crucial factor in cross-cultural interactions emerging from racial, ethnic, and national differences (Ang et al., 2006). 

Moreover, there are some controversial debates regarding the role of language proficiency in the use of conventional expressions as well. Although research has shown that there is a positive relationship between the high level of proficiency and enhanced ability in the production of conventional expressions, it does not guarantee a native-like selection (Bardovi-Harlig, 2001).

Therefore, this research aimed to investigate the role of language proficiency and cultural intelligence in the use of conventional expressions among Malaysian ESL learners, as a culturally different environment, by addressing the following two research questions:

1. Is there any significant correlation between the level of proficiency of Malaysian ESL learners with their use of formulaic sequences?
2. Is there any significant correlation between the cultural intelligence of Malaysian ESL learners with their use of formulaic sequences?

Consequently, two null hypotheses are listed below as follows:

1. There is no notable relationship between the level of proficiency of Malaysian ESL learners and their use of conventional expressions.
2. There is no notable relationship between the cultural intelligence of Malaysian ESL learners and their use of conventional expressions.

2.  Methodology

To investigate the role of cultural intelligence and level of proficiency in the use of conventional expressions among Malaysian learners of English as a second language, the current researchers applied a correlational research methodology and t-test analysis first by descriptive analysis of the participants’ level of cultural intelligence on a 7-point scale ranging from strongly disagree to strongly agree and their ability in the use of conventional expressions on a four-point rating scale ranging from zero (cannot evaluate) to three (almost perfect). In the end, the correlation between the level of proficiency and cultural intelligence, as two independent variables, and the use of conventional expressions as the dependent variable was assessed.

 2.1. Participants

The participants were 50 Malaysian undergraduate students studying Pharmacy at USM during their first semester of the academic years 2014/2015. They were in their first, second, and third years of study. Twenty-five were considered as having a high proficiency level and 25 as having a low proficiency level, based on their IELTS and MUET (Malaysian University English Test) results. MUET results were considered because not all students had IELTS certificates. The current researchers considered students who had fallen under band 4 or 5 as high intermediate L2 learners and good users (Group 1) and students who had obtained band 2 or band 3 as intermediate and low intermediate L2 learners (Group 2).

MUET is a compulsory English proficiency test that includes four language skills (listening, speaking, reading, and writing) for all Malaysian students who are planning to pursue their degree in a local university (Thang et al., 2012). The average grade for MUET is band 3 and lower than that fall under band 1 or band 2; however, band 1 does not exist anymore and the maximum band is 6 which is nearly impossible to obtain. The participants selected through a simple purposive sampling involved 35 females and 15 males with ages ranging from 19 to 22 with an average age of 20.5.

2.2. Instruments and Materials

A cultural intelligence scale (CQS), developed by Ang et al. (2006), was adopted, as one of the data collection instruments, for this study. The questionnaire consists of 20 items with four subscales whereby items 1-4 are related to meta-cognitive cultural intelligence; items 5-10 are related to cognitive cultural intelligence; items 11-15 are related to motivational cultural intelligence; and items 16-20 are related to the behavioral cultural intelligence aspect. A 7-point Likert scale was used for the items ranging from strongly disagree to strongly agree with values 1 to 7 assigned to them respectively (Rafieyan et al., 2015). According to Chen et al. (2011), a person with a high score having efficient communication proficiency with people from other cultures means that he/she can understand local practices and adapt to new cultures better than others. The content validity was evaluated by a panel of judges and the reliability was assessed by conducting a pilot study.

To assess the use of conventional expressions including expressions of apologies, leave-takings, gratitude, declining offers, warnings, requests, invitation, acceptance of a request, declining an invitation, acceptance of an invitation, deflecting thanks, an introduction, and an agreement, an oral Discourse Completion Task (DCT), developed by Bardovi-Harlig (2006), was applied as another data collection instrument. DCT test comprises 32 initiating and responding scenarios. In the responding scenarios (n=19), the participants were requested to respond to an interlocutor’s turn and in the initiating scenarios (n=13), the participants were requested to start an interaction (Rafieyan et al., 2014a; 2014b).The content validity of scenarios of the DCT test was evaluated by two lecturers who are proficient in the area of interlanguage pragmatics. The reliability of the scenarios of the DCT test was assessed through conducting a pilot study that involved over 30 undergraduate ESL learners in Universiti Sains Malaysia who were excluded from the actual study. The reliability coefficient evaluated through Kurder-Richardson Formula 21 (measure of reliability for a test) was 0.85.

2.3. Procedure

During the academic years 2014/2015, after dividing the participants into two proficiency levels (high and low) based on their IELTS and MUET (Malaysian University English Test) results, 50 questionnaire sets for the cultural intelligence scale were distributed among the participants: They were requested to respond to each item by selecting one of the numbers ranging from 1 (strongly disagree) to 7 (strongly agree) and enough time was given to them to reflect their responses to the items in the questionnaire. Then the questionnaire sets were returned to the administrator. The second set of data collection was through an oral discourse completion task which was conducted among the participants. They heard the scenarios through headphones and their oral answers were recorded from their headphones onto a digital file for data analysis. The participants had 7 seconds to answer. This given amount of time was purposely designed to increase the use of conventional expressions by the participants (Bardovi-Harlig &Vellenga, 2012).  

To measure language learners’ production of conventional expressions, firstly, the participant’s answers were transcribed by the first author, and the level of appropriateness in the use of these expressions by the participants was evaluated by two English native speakers using a four-point rating scale, as shown in Table 1, ranging from zero (cannot evaluate) to three (almost perfect). The interrater reliability was R=0.88. Table 1 shows the rating descriptions (adopted from Taguchi, 2013). Thirty-two scenarios were administrated among the participants. Each participant could obtain a score ranging from 0 to 96.

 Table 1

Rating Band Descriptions [adapted from Taguchi (2013)]

To measure the language learners’ cultural intelligence, descriptive statistics, mostly comprised of mean and standard deviation, were applied to explain and sum up features of the gathered data. Cultural intelligence (CQ) was indicated by a mean score on a 7-point scale whereby 1 (strongly disagree) displayed the minimum score and 7 (strongly agree) displayed the maximum score on the scale. A mean score of 4 indicated the average score (Rafieyan et al., 2015). In this way, a mean score above 4 stands for ahigh level of cultural intelligence, and a mean score below 4 stands for a low level of cultural intelligence. The mean score and standard deviation were generally calculated for all sub-scales of cultural intelligence including cognitive, behavioral, motivational, and metacognitive.

To find out the relationship between cultural intelligence and the production of conventional expressions, Pearson Product-Moment Correlation (r) was applied. Moreover, a t-test assessed whether the mean of two groups (students with a high level of proficiency and students with a low level of proficiency) are statistically different from each other in the use of conventional expressions or not.

3. Results

Table 2 presents the findings of the correlation between the ability of Malaysian ESL learners in conventional expressions usage and their cultural intelligence. Table 3 represents the descriptive statistics regarding the difference between the levels of competency in the production of conventional expressions by the participants with two different proficiency levels and Table 4 represents the degree of difference between two different proficiency levels in the production of conventional expressions. Levene’s test (1961) for equality of variances determines if the two groups with different proficiency levels in the production of conventional expressions have nearly the same or different amounts of variability between scores. 

Table 2

Relationship between Cultural Intelligence and Production of Conventional Expressions

**There is no correlation at the 0.032 level (2-tailed)

Table 3

Descriptive Statistics

Table 4

Independent-Samples T-Tests (Difference in the Ability in the Production of Conventional Expressions between Group 1 and Group 2)

To indicate the magnitude of the difference between the two groups, the effect size statistics using eta squared was calculated. The eta squared value obtained in the studyis0.25 which is a very large effect size. Twenty-five percent of the variance in the production of conventional expressions is explained by language proficiency.

4. Discussion and Conclusion

The current study aimed to investigate the relationship between the Malaysian ESL learners ‘ability to interact effectively in culturally different environments and their level of proficiency with their use of conventional expressions as one of the types of FSs. 

The results suggested that the difference between the two groups with different proficiency levels in the use of conventional expressionsis statistically significant. However, no relationship was found between the two variables (their level of cultural intelligence and production of conventional expressions). Therefore, on one hand, the findings reject the null hypothesis which states there is no significant relationship between L2 learners with a high and low level of proficiency in the appropriate use of conventional expressions, and on the other hand, the findings accept the null hypothesis which states there is no significant relationship between the level of cultural intelligence of Malaysian ESL learners and their appropriate use of conventional expressions. The findings obtained in the present research are consistent with the findings obtained in the previous research conducted by (Matsumura, 2003; Rafieyan et al., 2015; Taguchi, 2006, 2011, 2013; Tajeddin &Momenian, 2012).

The results derived from the present study regarding the relationship between the level of proficiency and the production of conventional expressions can be explained through the fact that a higher level of proficiency requires both sociopragmatic and linguistic knowledge which in turn can be affected by several factors including individual differences such as motivation, attitude, anxiety, and self-confidence (Gardner et al., 1997; Dӧrnyei, 1998; Liu, 2012), teaching methods (Clément et al., 1994), sociocultural factors (Razmjoo & Movahed, 2009) autonomy, age, learning environment, and curriculum designers (Koosha et al., 2011; Liu, 2012; Razmjoo & Movahed, 2009; Souriyavongsa et al., 2013) can be considered as one of the most important factors in the use of conventional expressions (Bardovi-Harlig, Rose &Nickels, 2008; Taguchi,2013).

Based on the literature reviewed, students with a higher level of proficiency are closer to the native speakers’ selection in the use of conventional expressions, and that second language proficiency plays a significant role in the appropriate use of these types of expressions. Therefore, the overall appropriateness in the use of conventional expressions is significantly related to the EFL L2 learners’ language proficiency. However, only a little difference has been reported in the types of linguistic expressions used by L2 learners with different proficiency levels. Moreover, discourse and grammatical proficiency which is also related to the level of proficiency seems to affect the quality of conventional expressions (Taguchi, 2006). General proficiency is considered a substantial factor in the appropriateness and grammaticality of conventional expressions (Taguchi, 2011).

These findings cannot be extended to the relationship between CQ and the use of conventional expressions in general, as they belong to a broad category in language and not merely to the conventional expressions. Pedagogically, teachers are advised to provide opportunities for L2 learners to be exposed to the second language culture by incorporating their teaching resources with authentic materials (Krashen& Terrell, 1983; Rafieyan et al., 2014a; 2014b). Because, Cultural intelligence, as an ability to communicate effectively across cultures, plays a significant role in the individual’s success in cross-cultural interactions. The results obtained from this research would benefit the curriculum designers and teachers by boosting the communication skills and pragmatic competence of ESL learners. However, more studies should be conducted with different instruments and methodology in a different context to ensure more accurate results.