home

Teachers’ Use of Gestures while Asking Questions in the SLA Classroom Shawn Loewen, Nobuhiro Kamiya, Yeoreum Lee, Kimi Natatsukasa, Mark Shea

The role of gesture in second language learning has grown as an area of research in recent years (See for example, McCafferty & Stam 2008). Studies are investigating how learners acquire L2 gestures, as well as how gestures may facilitate L2 comprehension. The current study aims to describe what types of gestures occur during a common classroom discourse type: teacher questions. The motivation for investigating teachers' use of gestures during questions is to gain understanding of what types of gestures occur in specific contexts in the L2 classroom. Once such patterns of usage have been identified, then researchers can begin to examine their impact on L2 comprehension and learning.
 * Introduction**

Gesture has been defined as movements of the hands and arms (McNeill, 1992), and the role of gesture in communication has been a topic of interests in the fields of psychology and linguistics. Their studies on gesture have attempted to depict the close relationship between speech and gesture. McNeill’s (1992) extensive study on gesture, using video-taped data of spontaneous speech, found that gestures and speech occur simultaneously.  Furthermore, McNeill argues that the use of gesture contributes to listener’s comprehension because gestures render speaker’s thoughts visible and exhibit images that cannot always be expressed in speech. In addition, the accompaniment of gestures with speech illustrates images that speakers think are connected. McNeill identifies six types of gesture: emblem ,iconic, metaphoric, deictic, beat, and cohesive. Emblematic gestures are those that express a culturally ageed upon meaning, such as thumbs up conveying the meaning of 'good' or 'okay'. Iconic gesture refers to gestures that exhibit a close formal relationship to the semantic content of the speech they accompany. For instance, a speaker may create a hemisphere to indicate a bowl in the kitchen. Metaphoric gestures are similar to iconic gestures; however, they represent an abstract idea rather than a concrete object. An example of metaphoric gesture is when a speaker raises his hands up and depicted two entities with half-closed palms while describing the cartoon with two characters. Deictic gestures indicate objects and events in the concrete world, even where there is nothing objectively present to point at. An example of a deictic gesture would be pointing at a person, who happens to be the subject of a conversation. In beat gestures, the hand moves along with the rhythmical pulsation of speech, often occurring as two movement phases with the hand moving up and down or in and out. Lastly, cohesivegestures refer to gesture which serves to tie together thematically-related but temporally-separated parts of the discourse. Cohesive gestures often contain repetitions of a certain gestures.
 * Gesture**

Some educational studies have presented how teachers use nonverbal behavior to accommodate their verbal cues in the non-language classroom (e.g., Roth & Lawless, 2002a; 2002b). They observed the instructor's speech in science classrooms and depicted how the instructor used materials in the classrooms and pictorial gestures when explaining the complex ideas. Recent SLA studies analyzed the function of such nonverbal behavior in L2 classrooms. (e.g., Allen, 2000; Lazaraton, 2004; Sime, 2008).
 * Gesture and L2 Classroom**

For example, Allen (2002) observed and videotaped second year Spanish as a foreign language classroom. Using the transcription of the observed classes, which included both verbal and nonverbal information, Allen illustrated how non-verbal cues functioned in the classroom interaction. For her analysis, Allen employed a coding scheme which was developed based on Burgoon et al. (1989) and Ekman (1980). This gesture coding scheme shared several similar categorization as McNeil's. For example, //baton// and //underliner// in Allen's categorization is similar to Mcneil's //beats.// They both account for emphasis on specific words or phrase during speech. Another example is //pictograph// in Allen's categorization and //iconic// in McNeil's. They both depict the shape of the referent. However, Allen's categorization includes interpretation of emblematic gestures and the coding objects which are used along with gestures. Allen described Emblem gesture as "symbolic movements specific to a culture used to repeat, replace, or contradict the verbal message" and Artifact as "Physical objects used to communicate meaning." (P.160, Allen 2000). An example of Emblem gesture is [Kimi: any brief good emblematic gesture from our data set?]; and an example of Artifact is that the instructor pointed a vocabulary of blackboard while verbally mentioning the same vocabulary. Allen identified different types of nonverbal cues which she found in high school Spanish language classrooms. The salient features found in her studies were: emblems (e.g., thumbs-up), illustrators [ shawn: would these be our metaphoric and iconic gestures? Kimi: Shawn, Allen had sub-categories under illustrators and one of them is iconic and some others fall under metaphorics. we should go into the comparison of two coding schemes here? .] which illustrate the accompanying verbal message, affect displays such as facial expressions. Her study reminds researchers of the importance of taking nonverbal aspects into account when analyzing classroom data.

Lazaraton (2004) investigated instructor's use of gesture during vocabulary explanation in an ESL classroom. Because the majority of the vocabulary explained in the data set were verbs, she identified many kinetographic gestures. Her detailed description of instances of gestural explanation illustrated different types of gestures used during the explanation: emblematic gestures ("shurruging" gesture for the vocabulary "to forget"), pictographic gestures (imitating using a broom for the vocabulary "sweep", and whole body movement gesture ("tiptoe"). These gestures indeed provided additive meaning to the verbal explanation, and helped learner comprehension. [Kimi: Here, we can explain how these gestures would be coded in our coding scheme - but I am not sure if it is necessary. ]

In another study of gesture, Sime (2008) specifically investigated how ESL learners interpreted the gestures employed by their teachers. The ESL learners participated in a stimulated recall session in which they viewed video recordings of their classroom interaction and were asked to comment on the teachers' use of gesture and other non-verbal behaviors. Overall, their stimulated recall comments included the following three aspects: gestures helped learners' comprehension, gestures facilitated learning, and gestures functioned as feedback.

As both Sime and Lararaton mentioned, gestures seem to facilitate learners’ comprehension of input in general. However, to date, few studies have specifically focused on which and how gestures by L2 instructors facilitate comprehension of different types of input. In order to better understand how gestures facilitate L2 comprehension, it is necessary to investigate the types of gestures instructors use in relation to different types of teacher talk. In that regard, several studies have investigated the occurrence of gesture in relation to corrective feedback and negotiation of meaning (Davis, 2009; Kida & Faraco, 2008) and vocabulary explanations (Lazaraton, 2004). In fact, Lazaraton concludes that ‘gestures... are forms of input to classroom second language learners that must be considered a salient factor in classroom-based second language acquisition research’ (p. 79). To that end, the current study will investigate gesture within the context of teachers’ classroom questions.


 * Teacher Questions**

Teachers’ questions are an important and common component of second language (L2) classroom interaction (Long & Sato, 1983; White & Lightbown, 1984), and they have been divided into display and referential questions. Display questions seek to elicit information that is already known to the teacher and thus the question acts more as a test of learner knowledge. Referential questions ask for information that is unknown to the teacher and thus involve the teacher and learner in meaning-focused interaction. With the advent of communication language teaching, the use of display questions has been looked down upon because they do not provide opportunities for authentic, meaning-focused interaction. Instead, teachers have been encouraged to ask referential questions which give learners the opportunities to produce language from their own linguistic resources regarding topics that are of personal relevance. This kind of language production is viewed as optimal for language learning from a CLT perspective. Furthermore, such referential questions can facilitate the negotiation of meaning, which is another important construct in interactionist approaches to SLA (Walsh, 2002). However, Lee (2006) argues that display questions are an important resource for teachers and should therefore not be summarily dismissed. In addition, the dichotomy between display and referential questions has been questioned (Nunn, 1999). Regardless of the impact on learning that the two types of questions have, it is clear that they both may occur in the L2 classroom.


 * Research Questions**

Given the ubiquity of teacher questions in the classroom and the recent focus on gesture in the classroom, the current study addresses the following research questions.

1. What gestures accompany teacher’s questions in the L2 classroom? 2. Are certain gestures associated with certain types of teacher questions?

This study employs an observational and descriptive approach to the occurrence of teachers' questions and gestures in the classroom. Researchers recorded ESL classes, and the subsequent video data was analyzed.
 * Methodology**
 * General Design**

The observations were carried out in two classes in the Intensive English Program (IEP) at a large midwestern university. Class A was a low intermediate content class focusing on American culture. Class B was a beginning reading and writing class. The teacher for Class A was a female native speaker of Japanese who had previously taught Japanese as a foreign language for two years and English as a second language for one year, both in the United States. The teacher for class B was a male native speaker of Japanese with twelve years of EFL teaching experience in Japanese junior high schools, as well as three years of ESL and Japanese as a Foreign Language teaching experience in the United States. Both instructors were members of the research team conducting the study, although they conducted no observations of their own classrooms and did not take part in transcribing or coding the data from their own classes. In addition, when the video observations were done, the research questions of the study had not yet been specified. Instead, the goal was simply to record L2 classroom interaction. In this way, the teacher researchers were not aware of the purpose of the study at the time the data were collected, and any reactivity regarding the object of investigation due to teachers' prior knowledge of the topic was avoided.
 * Setting and Participants**

Although data from the students were not analyzed in this study, demographic information regarding the students is provided to help contextualize the classrooms under investigation. Class A consisted of 10 students, and Class B consisted of 11 students. Sixteen of the students were female, and 5 were male. The average age was 25, and the first languages of the students were Arabic (13), Chinese (4), Korean (2), Kurdish (1) and Bambara/French (1)

Each class was observed for six hours, providing a total of twelve hours of classroom observation. Four of the researchers conducted the observations, with each researcher observing three hours of class. In addition, only a single researcher was present for each observation. A digital video camera was placed in the corner of the classroom and was used to record the class. The camera primarily followed the teacher, but occasionally panned to follow student interactions. In addition to the camera microphone, the teacher wore a lapel microphone attached to a digital recorder. Of the twelve hours of observations, roughly three hours were not video-recorded due to in-class breaks and technical difficulties, leaving roughly nine hours of video data. Portions of video in which the teacher was off-camera for more than one minute were then removed, leaving 8 hours and 15 minutes of data for the analysis.
 * Procedure**

Each researcher transcribed their own observations. The data were first transcribed by listening to the audio recordings from the teacher's microphone. After that, the transcripts were revised while watching the video recordings. Next, questions asked by the teacher were identified and coded as display or referential questions, according to the taxonomy reported in Walsh (2006). In coding the gestures that occurred during the questions, the researchers identified any gestures initiated during the question in addition to gestures that may have been initiated before the question but continued through it. Gestures were coded according to a modified taxonomy developed by MacNeil (1992) and reported in Lazaraton (2004), with elements taken from work by Allen (2000). The categories of gesture used were //emblem, iconic, metaphoric, deictic, beat, artifact,// and //nodding.// It was possible for a single gesture to include multiple types.
 * Analysis**

> [The categorization of Lazaraton is based on McNeill. It may be better to quote McNeil? What do you think?]. shawn: i think quoting Lazaraton is fine, but we should definitely give her article more space in the lit review] > > > In order to ensure reliability in data coding, four researchers each coded three hours of the data. After several rounds of meetings to discuss coding anomalies, the average final average agreement amongst the four raters was 78% in identifying questions, 91% in coding the type of question and 84% in coding gestures. > > The overall frequency and percentage of display and referential questions was calculated. In addition, the types of gestures that occurred in these questions were tabulated. Finally, specific recurring gestures in data set were identified. > > > **Results** > The first research question asked about the types of gestures that accompany teachers' questions. As shown in Table 1, a total of 1,143 questions were identified in the video data. Of these questions, just over half were not accompanied by any type of gesture.
 * Emblematic gestures are " symbolic movements specific to a culture used to repeat, replace, or contradict the verbal message" (Allen, 2000. p.160). An example of emblematic gesture would be a thumbs-up for an affirmative response.
 * Iconic gestures are "closely related to the semantic context of speech" (Lazaraton, 2004, p. 84), and as such, they represent concrete, imagible semantic concepts. Iconic gestures might take the form of a pair of fingers "walking" or hands sketching out a "circle."
 * Metaphoric gestures "represent an abstract idea rather than a concrete object or action" (Lazaraton, 2004, p. 84). This is perhaps the most flexible category, including many types of meaningful gestures. An example of an metaphoric gesture might be a hand moving in a circular motion to mean, "again."
 * Deictic gestures are those that have "a pointing function, either actual or metaphoric" (Lazaraton, 2004, p. 84).
 * Beat gestures consist of the hands moving "with a rhythmical pulse that lines up with the stress peaks of speech" (Lazaraton, 2004, p. 84?)
 * Artifact gestures incorporate "physical objects used to communicate meaning" (Allen, 2000, p. 161).
 * Nodding was coded as the vertical movement of the head.

Table 1
 * |||| Teacher 1 (K) |||| Teacher 2 (N) |||| Total ||
 * Gesture || Frequency || % || Frequency || % || Frequency || % ||
 * Present || 343 || 49.7 || 164 || 36.2 || 507 || 44.4 ||
 * Absent || 347 || 50.3 || 289 || 63.8 || 636 || 55.6 ||
 * Total || 690 || 60.4 || 453 || 39.6 || 1143 || 100.0 ||

> Table 2 reveals the types of gestures that accompanied questions. The most frequent type of gesture was metaphoric, accounting for one third of the data. The next most common category was the use of multiple gestures (30%). Deictic gestures and artifacts were the next most frequent categories (15% and 10% respectively). Finally, emblems, iconic gestures, beats and nodding occurred relatively infrequently in the data. > > > > Table 2
 * |||| Teacher 1 (K) |||| Teacher 2 (N) |||| Total ||
 * || Frequency || % || Frequency || % || Frequency || % ||
 * Emblem || 1 || 0.3 || 7 || 4.3 || 8 || 1.6 ||
 * Iconic || 7 || 2.0 || 10 || 6.1 || 17 || 3.4 ||
 * Metaphoric || 104 || 30.3 || 64 || 39.0 || 168 || 33.1 ||
 * Deictic || 54 || 15.7 || 24 || 14.6 || 78 || 15.4 ||
 * Beats || 18 || 5.2 || 11 || 6.7 || 29 || 5.7 ||
 * Artifact || 43 || 12.5 || 11 || 6.7 || 54 || 10.7 ||
 * Nodding || 2 || 0.6 || 1 || 0.6 || 3 || 0.1 ||
 * Multiple || 114 || 33.2 || 36 || 22.0 || 150 || 29.6 ||
 * Total || 343 || 67.7 || 164 || 32.3 || 507 || 100.0 ||

> The second research question asked about the types of gestures that are associated with different types of questions. The results reveal that referential and display questions occurred almost equally in the data, with 561 referential and 582 display questions. Table 3 and 4 shows that just over half of the referential questions contained no gestures, although there was some difference between the two teachers (52% versus 68%). As in the overall data, metaphoric gestures were the most common. > > > > > > > Table 3: Referential Questions
 * = Gesture Type ||||= Teacher 1 (K) ||||= Teacher 2 (N) ||||= Total ||
 * =  ||= Frequency ||= % ||= Frequency ||= % ||= Frequency ||= % ||
 * = No gesture ||= 213 ||= 52 ||= 101 ||= 68 ||= 314 ||= 55.9 ||
 * = Emblem ||= 0 ||= 0 ||= 2 ||= .1 ||= 2 ||= .03 ||
 * = Iconic ||= 1 ||= .2 ||= 7 ||= 5 ||= 8 ||= 1.4 ||
 * = Metaphoric ||= 67 ||= 16 ||= 13 ||= 9 ||= 80 ||= 14.2 ||
 * = Deictic ||= 45 ||= 11 ||= 4 ||= 3 ||= 49 ||= 8.7 ||
 * = Beats ||= 10 ||= 2 ||= 3 ||= 2 ||= 13 ||= 2.3 ||
 * = Artifacts ||= 26 ||= 6 ||= 6 ||= 4 ||= 32 ||= 5.7 ||
 * = Nodding ||= 1 ||= .2 ||= 1 ||= .1 ||= 2 ||= .03 ||
 * = Multiple ||= 50 ||= 12 ||= 11 ||= 7 ||= 61 ||= 10.8 ||
 * = Total gesture ||= 200 ||= 81.0 ||= 47 ||= 19.0 ||= 247 ||= 100.0 ||
 * = Total ||= 413 ||= 73.6 ||= 148 ||= 26.4 ||= 561 ||= 100.0 ||

> Table 4: Display Questions
 * = Gesture Type ||||= Teacher 1 (K) ||||= Teacher 2 (N) ||||= Total ||
 * =  ||= Frequency ||= % ||= Frequency ||= % ||= Frequency ||= % ||
 * = No gesture ||= 134 ||= 48.3 ||= 188 ||= 61.6 ||= 322 ||= 55.3 ||
 * = Emblem ||= 1 ||= .03 ||= 5 ||= 1.6 ||= 6 ||= 1.0 ||
 * = Iconic ||= 6 ||= .2 ||= 3 ||= 1.0 ||= 9 ||= 1.5 ||
 * = Metaphoric ||= 37 ||= 13.3 ||= 51 ||= 16.7 ||= 88 ||= 15.1 ||
 * = Deictic ||= 9 ||= 3.2 ||= 20 ||= 6.5 ||= 29 ||= 5.0 ||
 * = Beats ||= 8 ||= 2.8 ||= 8 ||= 2.9 ||= 16 ||= 2.7 ||
 * = Artifacts ||= 17 ||= 6.1 ||= 5 ||= 2.6 ||= 22 ||= 3.8 ||
 * = Nodding ||= 1 ||= .03 ||= 0 ||= 0 ||= 1 ||= .01 ||
 * = Multiple ||= 64 ||= 23.1 ||= 25 ||= 8.2 ||= 89 ||= 15.3 ||
 * = Total gesture ||= 143 ||= 55.0 ||= 117 ||= 45.0 ||= 260 ||= 100.0 ||
 * = Total ||= 277 ||= 47.6 ||= 305 ||= 52.4 ||= 582 ||= 100.0 ||

> **Patterns in the discourse** > In addition to identifying the different categories of gestures that accompanied questions, the researchers also noted specific gestures that occurred relatively frequently in the discourse. Interestingly, some gestures were seen mainly only from one of the two teachers (see Table 5).

Table 5  > Pattern 1 – Artifact to chin > There were numerous instances where Teacher 1 was holding an artifact, usually the textbook or the handout, in both hands at about stomach level, and as she asked the question, she raised the artifact and brought it towards herself, resting it either against her chest or chin. Most instances of this gesture were with display questions, asking the whole class to respond. Two examples are shown below. > > T: Then can you tell me what is affiliation? > T: Where are they in the map?
 * |||||| Teacher 1 (K) |||||| Teacher 2 (N) |||||| Sum ||
 * Gestures || Display || Referential || Sum || Display || Referential || Sum || Display || Referential || Sum ||
 * Artifact to chin || 42 || 6 || 48 || 0 || 0 || 0 || 42 || 6 || 48 ||
 * Arm(s) extended || 22 || 6 || 28 || 5 || 1 || 6 || 27 || 7 || 34 ||
 * Pointing student || 7 || 34 || 41 || 3 || 10 || 13 || 10 || 44 || 54 ||
 * Pointing artifact || 38 || 31 || 69 || 5 || 8 || 13 || 43 || 39 || 82 ||
 * Raised hand || 1 || 14 || 15 || 0 || 2 || 2 || 1 || 16 || 17 ||
 * Sum || 110 || 91 || 201 || 13 || 21 || 34 || 123 || 112 || 235 ||

> Pattern 2 – Arm(s) extended > > A pattern involved both teachers was extending either one or both arms out at her/his side with the palm(s) slightly upturned as if she/he is welcoming students' response. There were more instances of this gesture that went with display questions than with referential questions, asking the whole class to respond. Two examples are shown below. > > T: But, what does, what does it mean? Heartwarming movie? > T: What was the problem? > > While the above gestures were coded as metaphoric, there was at least one gesture that was identified as iconic. When Teacher 1 asked //what class is offered? what is offered? Offered?//, she raised one arm, bent it back to her shoulder and then extended it in an ‘offering’ motion.

> Pattern 3 – Pointing student > > On many occasions, the teacher nominated specific students to answer questions by pointing to them, and also by sometimes saying their names. There were more instances of this gesture with referential questions than with display questions. One of the reasons why the gestures in Patterns 1-3 were likely to be used with a certain type of question is that both teachers tended to ask display questions of the whole class while referential questions of each individual. Two examples are shown below. > > T: What were you gonna say, XXX (student's name)? > T: Yeah, you didn't come?

> Pattern 4 - Pointing artifact > > There were even more occasions when the teachers pointed artifacts in their hands (textbooks or handouts) or the chalkboard while asking questions. There were roughly equal number of instances between display questions and referential questions that went with this type of gesture. Two examples are shown below. > > T: Did you find North California in the map? > T: Do you think they can have more information?

> Pattern 5 – Raised hand > > When the teachers asked questions in which students had to identify their membership in a specific group, the teachers often raised one hand as they asked the question. Although this is a typical gesture seen in classroom in any subjects. Teacher 1 showed this gesture much more often than Teacher 2. This is partly because Teacher 1 used more gestures than Teacher 2 overall. Contrary to the pattern 1, most instances of this gesture was with the referential questions because the teacher was mostly checking the membership whose answer she/he could not anticipate. Two examples are shown below. > > T: Who won? > T: Who wants to watch Mona Lisa Smile?


 * Discussion**

What are common between the two teachers

The results showed that over a half of questions do not accompany any kinds of gestures no matter which type of questions are asked. This tendency did not differ between the two teachers. [nobi: do we have any gesture studies in which we can compare these numbers with? K imi: Not that I know of but probably we can connect this study with Lazaraton (2004). Her study only included vocab explanation, but she did observe some metaphoric gestures during explanation And the instructor in Lazaraton is also Japanese! ] One of the speculative reasons for this is that both teachers were Japanese. In Japan, gestures are not commonly used in society when speaking. Therefore, teachers from different background might show a different picture.

Another common feature between these two teachers was the type of gestures used. For both teachers, metaphoric gesture and multiple gesture were the ones used the most among all kinds of gestures, which together consists of over 60% of all the gestures. Many words that were expressed with metaphoric gesture were abstract. It is possible that the teaches speculated that their students may not be able to understand these words because these do not have concrete images and also owing to their low level of English. Then, they used gestures in order to help them better understand what they were trying to convey. The fact that both teachers used a great deal of multiple gestures show that their gestures were often not simple; rather they were complicated. However, it should be noted that multiple gesture also includes a fairly simple gesture, such as pointing the blackboard or the handout in their hands.

Some gestures were used only a few times by the two teachers: Emblem, Iconic, and Nodding. There are not many kinds of gestures that can be coded as Emblem. Also since we focused only on gestures that accompany questions, there were only a few instances of Nodding. Therefore, it is understandable that these two gestures were not seen frequently. On the contrary, the reason for infrequent use of Iconic should be addressed. One of the reasons might be that, as described above, when the word is rather concrete, the teachers thought that the help of gesture was not necessary. Therefore, they did not use any gestures. It is also possible that the teaches simply used more abstract words than concrete words.

What are different between the two teachers

As shown in Table 1, Teacher 1 asked more questions than Teacher 2 (690 and 453, respectively). This could be due to the type of instruction they had when the data were collected and also the difference in the amount of data available. Therefore, these numbers can not be compared. However, it is possible to compare within the same teacher. As shown in Table 3 and 4, Teacher 1 asked 413 referential questions and 277 display questions (59.9% and 40.1%, respectively). On the contrary, Teacher 2 asked 148 referential questions and 305 display questions (32.7% and 67.3%, respectively). Thus, Teacher 2 used strikingly more display questions than referential questions. Notice that Teacher 2 used more display questions than Teacher 1; nevertheless, Teacher 1 used far more referential questions than Teacher 2.

Regarding gestures, one difference can be seen in the percentage of questions that accompany gestures. As shown in Table 1, Teacher 1 used gestures for almost a half of the questions whereas Teacher 2 used only 36%. Therefore, Teacher 1 used gestures more frequently than Teacher 2. When we look at this difference in light of types of questions, an interesting pattern shows up. As shown in Table 3, on the one hand, Teacher 1 used gestures 81% of all the gestures used with referential questions. On the other hand, as shown in Table 4, Teacher 1 used only 55% of all the gestures used with display questions. Therefore, Teacher 1 and Teacher 2 used almost the same number of gestures for display questions. This could be attributed to the large number of display questions that Teacher 2 used as described above.

Lastly, differences between the two teacher emerged when examining typical gestures used. As discussed above, some gestures were seen only or mainly from Teacher 1. As shown in Table 5, the total number of these typical gestures is strikingly different between Teacher 1 and 2 (201 and 34, respectively). This could simply be due to the larger number of gestures used by Teacher 1 than Teacher 2; however, remember that there were more display questions asked by Teacher 2 than Teacher 1. Even only the data for display questions are compared, the difference is still vast between Teacher 1 and 2 (110 and 13, respectively). In fact, for Teacher 1, the total number of display questions with typical gesture was larger than that of referential questions despite the fact that she asked more referential questions than display questions (413 and 277, respectively). In contrast, for Teacher 2, the total number of display questions with typical gesture was smaller than that of referential questions despite the fact that he asked far more display questions than referential questions (305 and 148, respectively). These mismatches together show that Teacher 1 tended to use the same gestures repeatedly especially when she was asking display questions whereas Teacher 2 tended to use a variety of gestures no matter which type of question he was asking.

The cause of these differences need to be examined. First of all, gender could be one factor. Some studies reported that women tend to use more gestures than men in story retelling in L1 (Hostetter & Hopkins, 2002, Nicholadis, Pika, Yin, & Marentette, 2007) and L2 (Nicholadis et al., 2007). [nobi: I found two more studies about gender difference, but they are rather old: 1984 and 1995. And this is not the main topic of this paper, so I did not include them here. Dr. Loewen, what do you think?] Another factor could be teaching experience. Teacher 1 had three years of teaching experience whereas Teacher 2 had over 15 years. [nobi: are there any gesture study in light of teaching experience or age? Such as the more one is experienced, the less gesture one uses? Kimi: I just emailed you one study... ] [nobi: Kimi. I got one for gender, but not for age or teaching experience.] Another speculative factor lies in class, rather than teachers. Class A was a content class whereas Class B was a reading and writing class. [nobi: again, any studies?] It is likely that in a reading and writing class, teachers tend to ask questions that have certain right answers that they know (i.e., display questions) more than those questions whose answers they do not know (i.e., referential questions) as they are teaching skills of language. On the contrary, in a content class, the topic of discussion is often different from language itself, such as culture, personal experience, and life. Therefore, it is possible that teachers in a content class tend to ask more referential questions than display questions. The last prospective factor is the level of class. Class A was in the low intermediate level whereas Class B was in the beginning level. Teacher 2 might have thought that it would be more appropriate to ask display questions rather than referential questions as the former seemed easier to answer than the latter because answers are often prepared in advance or written in textbooks. Nevertheless, considering the level of the students, Teacher 2 should have used more gestures than Teacher 1, which did not prove to be true in the present study. Therefore, the whole picture may be more complicated. In order to further examine these issues, studies with larger sample sizes of teachers are indispensable. Also, the change of use of gestures must be tracked over a longer period of time. These may become themes in further studies of the combination of question and gesture.

References > McCafferty, S. & Stam, G. (2008) > Walsh, S. 2002. Construction or obstruction: Teacher talk and learner involvement in the EFL classroom. //Language Teaching Research, 6(1//), 3-23.

Nicoladis, E., Pika, H., Yin, H., & Marentette, P. (2007). Gesture use in story recall by Chinese-English bilinguals. Applied Psycholinguistics, 28, 721-735.

Hostetter, A. B., & Hopkins, W. D. (2002). The effect of thought structure on the production of lexical movements. Brain and Language, 82, 22-29.

>> >>