Online pedagogy : Development of the communicative skills in Russian online courses

This study on online language pedagogy takes a sociocultural approach to examine how tasks designed for synchronous multimodal web-conferencing environment (MWCE) encourage students’ development of communicative skills and how students initiate control over their own learning. The data from the archived synchronous class sessions of the secondsemester high-school Russian as a foreign language course taught completely online were analyzed using Conversation Analysis. The data shows that a problem-solving component built in the task design seems to be a powerful tool as it invites peer collaboration on solving linguistic problems, urges students to take control of their own learning and offers opportunities for participation in instructional conversations with the instructor. The study also shows that multimodal environment allows students at different developmental levels and with different learning styles to participate in the co-construction of the task via the communication channel of their preference.


Multimodal Web-Conferencing Environment
Synchronous multimodal web-conferencing tools (e.g.Wimba, Blackboard Collaborate) have been used for online instruction for more than a decade, but only recently have educators begun to explore how students develop communicative skills in such environments and how tasks designed for multimodal web-conferencing environment (MWCE) need to be designed to facilitate language learning.The concept of multimodality, or multimodal discourse, is not unique to CMC.In fact, "all discourse is multimodal" (Scollon and LeVine, 2002, p. 1) and "is always and inevitably constructed across multiple modes of communication, including speech and gesture not just in spoken language, but through such 'contextual' phenomena as the use of physical spaces in which we carried out our discursive actions" (p.2).In the context of CALL, multimodality refers to a combination of several communication channels such as audio, text, graphic, and video.Modern MWCEs incorporate audio-conferencing, textbased chat sessions, live video, and an interactive whiteboard, components that can be used simultaneously, in different combinations, e.g.audio-conferencing in conjunction with chat or an interactive whiteboard, or independently, one tool at a time, e.g.audio-conferencing or chat.While synchronous MWCE simulate face-to-face classroom, the MWCE differs from the face-to-face environment in that in the MWCE, participants can receive verbal information through several channels simultaneously and, consequently, can attend and respond to the contributions of multiple participants using several modalities thus constructing multimodal multi-participant discourse.The change in classroom discourse patterns results in the changes in class dynamics, class management, pedagogy, task design, and provides learners with opportunities to participate via a modality that better fits their learning styles and to take greater control over their own learning.This pedagogically motivated study focuses on some of these issues and seeks to answer to the following questions: 1. How do participants co-construct the context of the task via multiple modalities in a way that facilitates development of communicative skills?
2. How does language learning occur in synchronous MWCE?
3. How do students initiate control over their own learning?

Socio-cultural learning and CALL
The Socio-cultural Theory (SCT) is used as a theoretical framework to study a task-based approach to language instruction in synchronous MWCE.There are several reasons for grounding this type of research in SCT.Drawing on a socio-cultural perspective, learning is understood to occur in interaction with others, learners themselves, and through culturally constructed artifacts.Communication technologies are widely used to mediate social interaction.Similarly, they can be used as mediation means for learners' interaction.A task, which is a culturally created artifact, can also mediate language learning (Lantolf, 2000b).Therefore, learners working towards a completion of the task in synchronous MWCE can interact via different communication channels with others, e.g.their instructor and peers.Learners can also use the communication channels to mediate their own language learning through private speech.
Socio-cultural Theory is used to explain how language learning is mediated by others, learners themselves, and the tasks in synchronous MWCE.SCT proposes that in the process of interaction, biologically determined capacities for learning "are reorganized into specifically human forms of consciousness which allow us to intentionally and voluntarily control our memory, attention, planning, rational thought, problem solving, and learning" (Lantolf, 2000b, p. 79).According to SCT, "the source of human consciousness resides outside of the head and is in fact anchored in social activity" (Lantolf, 2000a, p. 13).When communicating with others, learners transform the regulatory means exercised by other conversationalists into their own mental activities.In other words, they internalize, or begin to "possess", the psychological processes materialized in their cooperation with others (Stetsenko andArievitch, 1997, as cited in Swain, 2000).The psychological processes are mediated by semiotic tools and one of these tools is language (Swain, 2000).Since learning is a mediated process, language learning is also mediated and viewed as both an object of learning and a mediating tool.Lantolf (2000b) lists three types of mediation including (a) mediation by others, (b) mediation by learners themselves, and (c) mediation by cultural artifacts such as tasks and computers used for interaction.

Mediation by others: Instructional conversations
Mediation by others occurs within a learner's zone of proximal development (ZPD) defined as "the distance between the actual developmental level as deter-mined by independent problem solving and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers" (Vygotsky, 1978, p. 86, italics in original).The definition suggests that ZPD is where learning as a form of collaborative effort takes place.This type of collaboration termed as "an instructional conversation" by Tharp and Gallimore (1991) and "a collaborative dialogue" by Swain (2000) is a critical construct for instructed SLA as it pinpoints a dialogic nature of learning which occurs when learners are engaged in "assisted performance" with an expert.Ohta (2000) cautions that for learning to occur, learners should not receive assistance with what they are already able to do and should be provided with an opportunity to develop an ability to work independently.The concept of instructional conversation (IC) is not new to CALL.Meskill and Anthony (2010) conceptualized this instructional technique for synchronous and asynchronous learning environments and developed a set of strategies that can be used to assist learner performance.

Mediation by self: Private speech
The notion of the learner's private speech which "tends to look like one half of a dialogue between individuals with a close personal relationship" (Lantolf, 2000b, p. 88) is another critical concept for SLA.Through private speech, "we gain control over our ability to remember, think, attend, plan, evaluate, inhibit, and learn" (Vygotsky, 1987, as cited in Lantolf, 2000b, p. 88).In a sense, private speech is the evidence of learners' progress in their development towards internalization of knowledge and is an intermediary product before "higher forms of consciousness arise on the inner plane" (Vygotsky, 1987, as cited Lantolf, 2000a, p. 15).In sociocultural framework, consciousness, or conscious awareness, refers to learner autonomy, or learner ability to regulate their own cognitive processes and, therefore, their own learning, which is encouraged by social interaction.Thus, the goal of socio-cultural learning is not only to create opportunities for learners to be able to perform tasks independently, but also to provide opportunities to learn to be in control of their learning.

Tasks as a mediation means
Lantolf (2000b) considers a task as a cultural artifact mediating language learning because it "requires learners to process language pragmatically in order to achieve an outcome that can be evaluated in terms of whether the correct or appropriate propositional content has been conveyed" (Ellis, 2003, p. 16).While tasks are meaning-oriented, they may be designed to compel students to use specific forms.Tasks are also "intended to result in language use that bears resemblance, direct or indirect, to the language used in real-world situations" and "can engage productive or receptive, and oral or written skills, and also various cognitive processes" (p.16).Since tasks provide learners with opportunities to make decisions about language use, participation in communicative tasks encourages learners' self-regulation as it "requires the learner to mobilize and orchestrate knowledge and abilities in a direct way which will become a catalyst for learning".
Scholars working in the field of computer assisted SLA also have begun to explore "the nature of the unique technology-mediated tasks that learners can engage in for language acquisition" (Chapelle, 2001, p. 2).Drawing on findings from research conducted in face-to-face environment, Chapelle suggests criteria for evaluation of CALL tasks.These criteria include language learning potential, or focus on form, meaning focus, learner fit, authenticity, positive impact on learners, and practicality.Wang (2007) suggests combining the criteria of language learning potential and focus on meaning because, as she argues, they are inseparable.Studies conducted on the effectiveness of tasks in MWCE also demonstrate that learners' language proficiency is a key factor for the successful completion of the meaning-focused task.Even more advanced learners make interruptions for form-focused negotiation with their instructor and expect corrective feedback from their online tutors during audio-conferences.These findings indicate that incorporation of the elements that draw learners attention to linguistic features in the meaning-focused tasks seems to be valuable for learners of different proficiency levels and necessary for beginners.Since beginners have limited linguistic repertoires, they "require a larger number of stimuli and more structured activities" so that they could also develop accuracy while being engaged in meaningful interaction.
Although tasks are viewed as mediation tools, it is not much known how they mediate language learning in the MWCE and how mediation by others, e.g. by the instructor and peers, can be built into the MWCE tasks (Hampel, 2010).Hampel reports that one of the problems related to the task implementation in a MWCE is that learners do not always view the task as opportunity for collaboration, but use an instructor "as an intermediary" instead of interacting with peers.Hampel notes that "the task instructions would have to include some of the functions that the teacher and peers have in face-to-face setting" (p.141).These functions include providing scaffolding to assist students in eventually developing autonomy "and to move from highly structured activities to more open ones where learners would take control, that is from individual tasks . . . to more collaborative ones where students would co-construct their own knowledge" (p.142).

Data and participants
The data stem from the archived synchronous class sessions of a second-semester high-school Russian as a foreign language course taught completely online.Three times a week students met with their instructor for fifty-minute sessions in Wimba, a MWCE which allows for real-time voice-and text-based interaction.Wimba Live Classroom also has an interactive whiteboard that can be used to display Power Point slides, pictures and documents, as well as to type, draw and hand-write using the drawing tool.There is also a live video, but this function was not used on regular bases.
Fourteen students from one school and two students from another one enrolled in this course offered by blendedshools.net(BSN) logged in from their school computer labs.Each student sat in front of a desktop computer which was supplied with a headset and a microphone.BSN is a US-based notfor-profit educational organization which provides live language instruction throughout several states and offers over 70 different courses in 10 different languages for elementary through college-aged learners.

Designing tasks for synchronous MWCE
When designing tasks for synchronous MWCEs, the criteria for the appropriateness of tasks for CALL (Chapelle, 2001) were used.The following considerations were also included in the task design: 1. integration of the problem-solving component which urges students' planning of their responses and drawing attention to meaning and form; 2. capacity for simultaneous engagement of multiple participants via different channels; 3. opportunity for scaffolding via multiple channels.
All tasks used for language instruction encouraged multiple students' participation through audio-and text-based channels and provided learners with an opportunity to participate in IC.These tasks include matching exercises, short dialogues, short narratives with fill in the gap components and picturesupported narratives.This paper discusses only the task in which students had to create short dialogues.The goal of the short dialogues was to teach how to enquire where people live.To provide students with an opportunity to master the language, the task included a series of short dialogues that repeat the same language.Figure 1 demonstrates a slide designed based on Meskill and Anthony's (2010) recommendations.Each slide includes a task toolkit, or "a set of language elements

FIGURE 1
Slide design for short dialogues task that make up the focus of a given language task" (p.4).The task toolkit displayed on each of the slides contains a model of the dialogue that students had to create.Each slide includes two pictures of two different countries to support two dialogues.The people's and countries' names to be used in the dialogues are typed in Russian in their base forms above the pictures.
To accomplish the task, participants had to solve several linguistic problems: 1. to read and understand the meaning of the words above the pictures; 2. to use an appropriate form of the verb жitь 'to live'; 3. to substitute the person's name with a personal pronoun in the response to the question; 4. to use the noun expressing location (the country) in prepositional case.
Instructional conversations were conducted in students' L1 because of their limited L2 proficiency.

Methodology
Participants' interactions during each task were transcribed using transcription conventions adapted from Jefferson (1984).Conversation Analysis (CA), which treats human interaction as "organizationaland procedural" and conversation "as an emergent collectively organized event" (ten Have, 2007, p. 9, italics in original) is used to explain how, in synchronous MWCE, participants co-construct the context of the task and learning.In CA, the basic unit of talk organization is a turn taken by a conversationalist voluntarily or by selection of the current speaker (Sacks, Schegloff and Jefferson, 1974).Turn-taking organization of the face-to-face classroom interaction differs from that of ordinary conversation in that only the instructor is to select the next speaker and "direct speakership in any creative way" (McHoul, 1978, p. 188, italics in the original).The general rule for both formal and informal face-to-face interaction is that one person talks at a time.Synchronous MWCE does not put such restriction on speakership because in MWCE, conversationalists can participate via text-and audio-based modalities which allows for greater variation in turntaking organization.
Since the goal of the study is to observe how learning takes place, it is important to determine what to take as evidence of student learning.In SCT, learning is defined as a successful "assisted performance" (Tharp and Galliomore, 1991), which is a learner correct response prompted by an expert.In CA, learner self-correction is termed self-repair (SR).In SLA, self-repair is viewed as "evidence of noticing an observable behavior from which we can infer that a learner has engaged in some monitoring strategy or has noticed a production error (Kormos, 1999, as cited in Smith, 2008, p. 85).Self-repair can be self-or other-initiated.Other-initiated self-repair (OISR) occurs after someone other than the speaker notices a problem in speaker's utterance and prompts the speaker's self-repair in the next turn.Self-initiated self-repair (SISR) "occurs when a learner corrects his or her own utterance without being prompted to do so by another person" (Foster and Ohta, 2005, p. 420).Thus, in the sociocultural context, OISR and SISR correspond to two different developmental stages.OISR occurs within a ZPD and is indicative to learning taking place.SISR is qualitatively different from OISR as it occurs at a later stage when the structure has not been fully internalized yet, but the learner is already able to notice the problem and regulate its production independently.

Task as a mediation means
The first example 1 demonstrates how the problem solving component built in the task design encourages multiple students to collaboratively construct the meaning of the words denoting countries using the chat while instructor uses audio-conferencing tool to talk to Jack and Harry, the students who volunteered to perform a dialogue.Although the instructor does not tell students to start working on the task, three students, Terry, Sam, and Nick start meaning negotiation of the words Franci 'France' and poni 'Japan' displayed on the slide in the chat.When the slide appears on the whiteboard, Terry makes incorrect guesses about the meaning of these words (lines 3, 5).Sam conducts repair on Terry's utterance "Paris" by typing "France and" (line 8), but fails to correct Terry's utterance "China" by incorrectly referring to the word poni 'Japan' as Asia (line 13).Nick attends to Sam's repair "France and" and collaborates with him by correcting Terry's utterance "China": "That's Japan, not China" (line 15).Although Sam's utterance "Asia" precedes Nick's repair turn (lines 13 and 15), Nick corrects Terry's, but not Sam's utterance.Since Nick's utterance is separated from Sam's by the instructor's audio-based input (line 14), apparently, Sam's utterance "Asia" became available in the chat box when Nick was still typing his own response.
(1) 1 T: We have a couple of more minutes, so we can do one more slide, 2 so who is now?Now, I remember Jack [Brown 3 Terry: T: um [wanted to ask a question 5 Terry: T: and who will help him?7 (3) 8 Sam: france and 9 (2) 10 Terry: EVAN!!!!! 11 T: YEAH! Harry!If I e-mail that you were working today, 12 T: [probably your facilitators will be so happy, ok, Jack and Harry 13 Sam: [asia 14 Which one do you choose?15 Nick: that's Japan, not China 16 Terry: :P 17 (5) 18 T: Jack

Assisted performance
The next two examples show how learning takes place in the course of IC.
When working with Jack and Harry, the instructor does not provide explicit corrections, but initiates student SR by calling students' attention to meaning and form.Since after the five-second pause (line 17) Jack did not ask a question, the instructor calls on him (line 18), but Jack responds that he does not know how to say the word 'Paris' (line 19).Jack's response shows that he has not read the word Franci 'France', so the instructor prompts Jack's self-repair by asking Jack to read the word and by underlining it on the slide (lines 19-22).Then, she monitors other students' attention as well by asking who can read it (lines 24, 25).
oqenь horoxo/very well done, but Jack self-corrects by reading the word Franci 'France' (line 28) and then asks a question, "Kto жiv t v Franci " 'Who lives in France' (line 30).Although the task toolkit indicates that the question has to start with the question word Gde 'Where?',Jack improvises and asks a different question.While the instructor accepts Jack's improvisation, she recasts Jack's question because of the phrase v Franci 'in France' (line 30).Jack uses the wrong case of the noun and does not change the preposition v 'in' to its alternate vo 'in' used if it is followed by words beginning with /v/ or /f/ sounds followed by another consonant.In his response, Harry uses the noun in the correct case, but he substitutes the name Michelle with the masculine pronoun although the slide specifies that Michelle is a girl.The instructor circles the words Mixelь (girl) on the slide and prompts Harry's self-repair by stating that Michelle is a girl (lines 35-36).Harry immediately self-corrects by using the feminine pronoun (line 37).
After Harry provided the correct response (ex.3), the instructor returns to the model and asks how students can answer the question beginning with Gde? 'Where?' (lines 39, 41, 43, 44).Terry's message in chat (line 42) seems to be an incorrect translation of the word gde 'where', but nobody corrects him.The instructor calls on Kevin who seems to understand the question Gde жiv t Mixelь?'Where does Michelle live?' because he answers in English, "I am

Students' control of their learning
Evidence that students control their own learning is reflected in the chat transcripts.As has been already discussed, Terry, Sam, and Nick voluntarily negotiate the meaning of the words to be used in dialogues.There are two more pieces of evidence demonstrating directly and indirectly how students regulate their learning though private speech.One of them is when Terry uses English letters to type the word 'Japan' as it sounds in Russian, "japania" (line 53).Since private speech is available to others in face-to-face environment, Kevin transfers this feature to MWCE and adjusts it to the new environment.If Terry said this word using audio channel, it would interfere with the instructor's and Kevin's talk, so he transforms the private speech to the text-based format.
The direct evidence of students' self-regulation comes from the text-based dialogue between Terry and Sean.Terry uses chat to call Evan, "EVAN!", using capital letters to emphasize that he says his name loudly (line 10).Sean refers to this message by saying "Terry it doesn't deal with class so stop say Evan" (lines 31, 32) Terry responds, "He wants to read russian thats hot it has to do with class so :P" (line 40).Although there is no evidence of what students were doing at that moment in the lab, this conversation implies that Evan reads words from the slide out loud.Apparently, Terry uses chat to call Evan in order to not disturb other students, but Evan does not notice his message; so, apparently, Terry starts calling Evan offline thus disturbing Sean.This chat conversation demonstrates that while some students may not make evidence of their learning available in MWCE, they may stay focused and practice language offline.

Concluding Remarks
Several implications can be made from this pedagogically-motivated study on socio-cultural learning in synchronous MWCE.A problem-solving component built in the task design seems to be a powerful tool for both encouraging students' participation and facilitating their learning.Thus, the problem solving component kept students focused on the task, invited peer collaboration on solving linguistic problems, and offered opportunities for learning via IC.It also facilitated the development of student self-regulation means and provided them with opportunities to become autonomous learners.The study also shows that multimodal environment allows students at different developmental levels and with different learning styles to participate in the co-construction of the task via the communication channel of their preference.Thus, students who seemed to be more comfortable with oral production participated via the audio channel.Students who needed assistance with interpreting word meaning invited peer collaboration in the text-based modality.Even students' production of private speech was realized in two different modalities.That the students opted to focus on different language-related issues while working towards the completion of the same task gives evidence to the fact that they were able to identify their needs and control their own learning.
The instances of successful assisted performance demonstrate that synchronous MWCE provides students with opportunities for learning.The instructor's monitoring Jack's attention on the word Franci 'France' resulted in his reading of the word, whereas her drawing Harry's attention on the gender of a pronoun prompted Harry's self-repair.With the instructor's assistance, Kevin was able to ask a question and to transform regulatory means exercised by the instructor when assisting Harry with the grammatical gender into his own mental activity when conducting SISR on the incorrectly used pronoun.Although this study advocates for sociocultural learning and a task-based approach to language instruction in synchronous MWCE, more research is needed to better understand this type of learning environment and teaching techniques that provide learners with the best learning experience.