If we consider learning as a meaning-making process where people construct shared knowledge, it becomes a social dialogical activity in which knowledge is the result of an active process of articulation and reflection within a context (Jonassen et al.). An important element of this belief is that conversation is at the core of learning because knowledge is language-mediated. Within this context, what makes a conversation worthwhile and meaningful is how it is structured, how it is managed by the participants, and most importantly, how it is understood. In particular, conversation is essential in learning situations where the main goal is to generate a new understanding of the world (Bruner). Thus, if conversations can be seen as support for learning processes, the question then becomes how synchronous textual spaces mediate conversation and how chat affects learning.
Experienced Teachers Learning in a Collaborative Virtual Environment
We studied two different groups of experienced teachers from Kindergarten to Grade 12 (K-12) attending a Master of Education course entitled "Curriculum and Instruction". They communicated through a collaborative virtual environment (CVE) designed to enhance teachers' professional development: TAPPED IN™ (TI). We recorded their on-line conversations over six weeks. The teachers met twice a week for a two hour session and the data collected consisted of approximately 350-400 pages of text from transcripts. The following concerns, gleaned from an ongoing analysis of on-line conversations are of interest for this paper:
- The first concern has to do with the ability of teachers to concentrate on a task while managing multiple simultaneous conversations. The question is how to maintain the focus on the purpose of the goal oriented task. The second concern is related to the technical characteristics of a CVE and the teachers' feelings of being lost, too slow, or not understanding the point of the discussion. The question is how to deal with this confusion when the aim is to construct meanings from online discussion?
- The third concern is related the preceding points. It is concerned with the importance of a leader coaching and guiding experienced teachers online.
We examined these three concerns, using TI during the teachers' on line discussions. Our primary goal in the analysis was to determine i) whether the teachers could conduct their learning activities through TI and ii) how goal-oriented conversations might be affected by the constraints of TI. The following examples come from a personal recorder. Messages are numbered in order to show their position in the session and to show the distance between the messages sent.
Implications of Multitasking in Learning Sessions
In CVEs, participants have the possibility of performing several tasks simultaneously (Holmevik). This is especially true when participants hold more than one conversation at a time. Participants can talk to one person or to the whole group while also chatting privately with people in the same CVE's room, in the same CVE or even in other CVEs. But the possibility of being able to participate in multiple conversations becomes potentially confusing and disorienting for teachers wanting to achieve a specific task.
Let us give an example of how a main task (e.g. to share notes of pedagogical projects -- task 1) fragments into different tasks (e.g. learning a command -- how to create a note -- task 2; and socialise, express feelings and play with cows -- task 3).
(Note that the students are in fact experienced teachers and a teacher is leading the session. The goal of the on-line session is to read and discuss the different educational projects that the students should have written in virtual notes.)
The goal of the task became difficult to accomplish for teachers who were suddenly involved in more than one task at a time. In order to understand what is going on in this situation, participants had to accomplish extra work. They needed to filter messages and rank them to make the main objective of the session clear. In a goal-oriented session such as this, it is extremely important to keep track of the task as well as to concentrate on one activity at time. This entails a necessity to understand current threads in order to contribute to the object of interest for them as individuals and as a group member.
Implications of Multi Threads and Floor-Taking in Goal-Oriented Conversations
Perseverance with each message creates a parallelism that can become extremely disorienting to participants who intend to produce new understandings and not just maintain an awareness during on-line conversations.
The larger the number of participants in a conversation, the more likely it is for fragmentation to occur. The jumbled and quickly scrolling screen can be quite disconcerting. Yet as mentioned by Mynatt et al., even between two participants, multi-threading is common due to the overlapping composition of conversational turns. Participants write simultaneously and the host computer sends the messages out sequentially. Under these conditions, competing conversational threads emerge continuously. It becomes difficult to know who actually holds the floor at the time.
Here is an example showing a teacher -- student 2 -- looking for attention and trying to read and understand others' answers to his questions:
Student 2 did not read message 26 sent from the teacher with care. In fact, the teacher did explain that there is a part in the assignment where students have to meet in order to exchange ideas about individual projects. Yet although S2's question was answered, S2 still did not understand. A possible reason is that S2 could have been focussed on writing the next question.
Again, the teacher answered the question asked in message 29. However, S2 still did not understand in spite of S15 and S6 confirming that the teacher had already covered the question.
Student 2 finally understood when the teacher addressed him directly and repeated what the other students had said before. In order to be heard, the students repeated their questions until they had the answer from the teacher. With more than a handful of participants, this attention seeking strategy may make on-line conversations confusing. Goal-oriented conversations then easily degenerate, as mentioned by Colomb and Simutis. These authors point out that one of the most common problems in using CMC is keeping students on task. Even experienced teachers do not escape from the possibility of converting from an instrumental discussion to a social one due to different misunderstanding between interlocutors.
To be able to 'send' a message is not equivalent to claiming the 'floor'. An important extra task that teachers have to do in CVEs before sending a message is to think about how it meets the goal of the discussion. Looking for coherence and understanding is a must in learning situations and this becomes a great challenge in online learning sessions.
On the other hand, different modalities of communication in CVEs may add richness and depth to online conversations when participants can anticipate constraints. Consider another group of teachers. They are discussing readings, and make great use of multiple modalities, such as gestures, to reframe misunderstandings. These gestures provide back channel information and other visual signs.
Here is one example of what a group of teachers does in order to avoid embarrassing situations.
As Mynatt et al. express, "the availability of multiple modalities gives complexity to the interactional rhythm, because people have choices about what modality to use at any particular moment and for any set of conversation partners" (138).
Given these pros and cons of CVEs, the challenge of holding an on-line educational discussion requires the teachers to reestablish the context and control the underlying the sense of the conversation. This challenge could be also regarded as an exigency of the medium that 'invites' teachers to structure their conversations in order to encourage meaningful discussions.
Importance of a Teacher of Teachers
The problems mentioned earlier may be solved more easily when there is a leader at hand. Since these difficulties mainly arise at the start of learning the communication environment, it might be proposed that a leader is most critical in this phase. A comparison of two groups' interactions with and without a leader supports the intuition that a leader is crucial for keeping the learning on track even though the participants are experienced teachers.
In this example, the task that the group performed was the same: "learning to attach an icon to their ellipses representing their presence in the system".
Table 1. Data related to groups with and without a teacher
|Groups||Learners||Icons attached||Messages produced||Time employed|
|1. Without leader||12||0||549||56 min 8 sec|
2. With leader
|9||4||644||1h 27min 52 sec|
Fig. 1 Comparing flow and categories of the messages sent by the groups
These frequencies confirm that teachers without a leader have more problems than the group with a leader. The number of successful icons attached by the groups (0 and 4 icons) demonstrates this claim. What happens is that the number of messages related to 'Task' decrease and those related to 'Relation' increase when there is no leader present -- a result which would be unsurprising among most people who have worked in 'real' classrooms. Messages produced and coded as 'Playing' and 'Feedback' also show a considerable difference between groups. Finally, categories such as 'Whisper' and 'Artifact' present in comparison to the others minimal differences between groups.
A leader is a must for the smooth development of on-line conversation. The leader is a sort of mediator between the pedagogical task of the on-line conversation and what appears on the screen. The leader's task is to show which threads are important to follow or not and how messages should be read on the screen. Like an orchestra conductor, the leader coordinates tasks and makes sense of individual actions which are part of a common product and the quality of the on-going conversation.
This ongoing research has demonstrated three important concerns surrounding experienced teachers' professional use of CVE.
First, teachers chatting online have to anticipate the lack of assurance "that what gets sent gets read" and that gaining the floor in a CVE is "that one's message draws a response and in some way affect the direction of a current thread" (Colomb and Simutis). Teachers have to learn to negotiate turn-taking sequences behind the screen. When chatting, a person's intention to speak is not signalled. Overlapping and interruptions do not exist and non-verbal communication requires knowledge of gesture commands. Negotiating turns in online conversations is concerned with how people express information and what they express. In educational discussion, turns are generally taken when messages either present a good formulation of ideas, express controversial thinking, raise an issue that allows someone else to participate, or provide knowledge on the topic at hand.
Second, teachers should learn to collaborate in online conversations. It is essential to be aware that people are writing a text while they discuss. The quality of the conversation will depend on one hand, how teachers manage the discussion and, on the other hand, the opinions they elaborate together.
Third, teachers need leaders in online discussions. A leader has to be able to anticipate the text that the participants are writing. The leader has the responsibility of meeting pedagogical goals with a participant's messages. The leader has to show the coherence or incoherence of the discussion and raise issues that improve the level of the written interaction.
These issues are extremely important in a context where people learn through conversations.
As Laurillard has mentioned, "academic knowledge relies heavily on symbolic representation as the medium through which it is known. ... Students have to learn to handle the representations system as well as the ideas they represent" (27). Therefore, it is necessary that learners know and think about the rules of online discussion in order to adapt technical commands and effects to their needs. But these rules are in contrast to what participants expect from online conversations. Teachers want to perform their tasks with support of a computer program; they do not want to learn the computer program per se.
CMC in learning activities must be based, not on visionary claims about technology as an all-purpose tool for automatic teaching/learning, but on specific accounts of how and why the technology affects the user's achievement of specific goals.