Literature Review

My dissertation is an exploration of a different approach to learning at Masters Level through an enquiry into methods of discourse that could be used in online social learning.

Table 1

Table 1 lists the key concepts identified as relevant to adult learners in formal, distributed, online, taught, social learning contexts.  Throughout this literature review, when I use the term teacher I am doing so in its broadest sense and am not implying a didactic pedagogy.  My literature review will first explore each of them in turn and then apply them to my own professional practice in my context of working in the field of online social learning at Ultralab. 

The learners I work with at Ultralab are adults who are required to be self-directed, are learning from experience, are using critical reflection, and for whom the primary agent of learning is through online communities.  For these learners, the requirement is for an online experience that promotes purposeful learning. 

In undertaking this literature review, I am attempting to re-conceptualise what is meant by online social learning in a way that capitalises on new technologies and varied approaches to discourse.  What is the model for online social learning that better equips participants with the skills and ability to both become more effective learners and to learn more effectively through different discourses?

I shall now deal with each of the concepts in table 1.

Social learning

I start from a philosophical position of recognising that learning is fundamentally a social process (Wenger 1998) and, as such, an effective model of online social learning needs to be supported by learning from experience with others - this locates learning in social participation through an ongoing process of negotiation of meaning.

"We are deeply influenced by our life experience, that learning contexts in higher education (as elsewhere) are social constructs, that knowledge is socially constructed and that when learners enter and experience higher education they enter a system that is not value free and where power is exercised that can influence the progress of learning of a student learner"

Brockbank and McGill (1998)

It follows that the context of the learner is the single most important factor of the learning experience.  In this, Brockbank and McGill (1998), emphasise the importance of the perception of the learner in their ability to think, feel and act.  They also identify the importance of setting up intentional opportunities for learning for learners.  The key question for teachers is how do they support the ability of learners to learn through social interaction given the socially constructed nature of knowledge, meaning is created in relation to others, then reflection and the creation of meaning is inevitably a social process.  The implication of this approach to teaching is the teacher as facilitator of learning where the focus is on the relationship between teacher and learner and how the process of learning is being managed with the emphasis on critical reflection applied to the subject and process of learning itself.

A body of literature, including work by researchers such as Lave and Wenger (1991) and Wenger (1998), is developing on the topic of Communities of Practice (CoP).  It identifies learning as a natural product of social participation in an informal apprentice model with the process of negotiation of meaning between the community members at its heart.  In seeking to use this model as a framework for learning, Wenger suggests that three key tests for CoP be applied - is there an identifiable domain, practice, and community?  Wenger suggests that unless at least two of these tests can be answered in the affirmative then a CoP will struggle to survive and it may not be a useful model to use in trying to understand how learning is taking place.

Learning in a CoP involves observation, instruction, experience, as well as conversation as a part of the learning mix.  More recently, the ideas around CoP have been applied to online environments, but in so doing it is important to recognise that the in the CoP identified and studied by Wenger, learning does not take place through conversation alone, but there are other important processes that are a part of this particular model of learning.

Brown and Duguid (1999), identify the non-formal learning in organizations that takes place through conversation around the 'water cooler'.  These individuals are not brought together in a formal way, but volunteer to support to each other through 'chance' contact.  Such groups may, if examined, demonstrate the characteristics of a CoP with the participants having a strong bond around a common enterprise defined by their practice and domain of knowledge. 

The extent to which different types of knowledge such as that which is more or less involved can effectively take place without formal structure and processes is one for debate. Could the knowledge, skills, and understanding for being an accountant be learned in the same way as the insurance claims processors observed by Wenger (1998) who learned about their work from each other through in a non-formal way through social participation.

Having identified social learning as a key strategy in my model of learning for my dissertation, I experimented with different forms of discourse using online technology as the agent.  This has included my patchwork website, the use of SMS messaging, email, Blog technology, and video.  These approaches add a richness and variety to discourse taking the learning protocol beyond that of social participation beyond simple conversation.

Critical reflection

Schön (1983) articulates a contemporary understanding of what is meant by reflective practice, although original ideas are traceable back to Dewy and others.   For Schön, the notion of a professional artistry was not a satisfactory explanation of how a professional's skills were honed 'on the job' so that they became expert at what they did.  Schön sought to understand how it is that professionals learn and develop their practice through combining the knowledge they had acquired in formal education with the experience they gained whilst practising their profession.  He identified a process of learning whereby professionals reflected on experiences they have had using propositional knowledge they had been taught as a 'tool' (reflection-on-action) and then took this knew understanding and knowledge they have created forward to real life situations where they reflected-in-action as they deal with situations as they arise.  These two occurrences of reflection were for Schön inextricably linked together with one feeding the other in a cycle of reflective learning.

In the context of my Masters dissertation, it is reflection-on-action that I is relevant to the focus of my enquiry is into methods of discourse that could be used in online social learning.   The online experience I am focusing on is an asynchronous one and as such is not linked to practice as it happens.

Atherton (2003) identifies a tradition of critical reflection in Higher Education based on the work of Mezirow who links the process of reflection about oneself to the concept of transformative learning "the process of reflecting back on prior learning to determine whether what we have learned is justified under present circumstances" (Mezirow, 1990).  This notion of critical reflection goes to the heart of the argument about what is the purpose of higher education and what are the influencing factors at work in shaping educational experiences.  What counts as "learning" for educational purposes depends on cultural, social, economic and political factors, because implicit in education as a deliberate enterprise is the notion of prescription.  Atherton (2003).  In my dissertation I am trying to push the boundaries of a dissertation at Masters level and in so doing am engaging in critical practice that challenges the orthodoxies of how a dissertation should or should not be approached.

In order to be able to reflect-on-action the learner needs to be able to describe the action to give them some data to reflect upon.  This could be an event, a conversation, or an abstract train of thought, but whatever it is unless it is turned from the tacit to the explicit it can't be reflected upon.  Brockbank and McGill (1997) see the understanding of this process and the ability to articulate action as the critical factor in supporting someone else in becoming a reflective practitioner "We emphasise the importance of naming the process and the importance, in dialogue, with someone who has the skill initially (and therefore to be modelled for the learner), to identify these dimensions of reflection".

Brockbank and McGill (1998) usefully identify five key requirements that promote reflection and reflective practice: dialogue; intention; process; modelling; and the notion of personal stance.  In using the term reflection, they are referring to both the process and the product, which is the creation of "meaning and conceptualisation".  I will briefly discuss the requirements of dialogue and intention as modelling and the notion of personal stance are dealt with in a later section.

Dialogue that is effective in promoting critical reflection is that which challenges the student on the edge of their understanding and experience of the world.  This challenge forces students to reassess previous held beliefs, identify and re-consider their own assumptions, and re-conceptualise their understanding linking new insights to the knowledge and understanding they already possess.  This view of a learning process can be likened to interpretations of Vygotsky (1978), where the Zone of Proximal Development is the area where a teacher focuses their effort, scaffolding the learner to explore and re-conceptualise ideas through dialogue.   And for Wenger (1998) it is through dialogue that we engage in social participation and negotiate meaning and learn. 

Brockbank and McGill (1998) draw an important distinction to be drawn here about what is meant by 'effective' dialogue for learning.  In itself dialogue does not necessarily promote deep learning as an exchange of views between two sets of people with engagement with the others position is unlikely to create much new understanding.  Deep learning is that which is transformational for the individual and has at its heart an active approach to learning where students seek to do more than simply absorb facts but seek to understand the subject, make links back to their own context and "feel themselves to be agents of their learning" (Marton, 1975).  Dialogue can be internal with oneself, but it is through the external dialogue with others that the most opportunity for insightful learning takes place.  "Dialogue that is reflective and enables shifts in assumptions about a person's sense of reality will be grounded in their experience and interactive.  It will be a dialogue engaging the participant's realities as opposed to that which is simply didactic." Brockbank and McGill (1998).

Although intention is an apparently straight-forward and obvious point, it is however an important one.  Unless participants are clear about what it is they are supposed to be engaged in it is far less likely to happen except by chance.  Likewise, unless the intention is made clear then participants cannot know if they have the skills and understanding about reflection that is required to effectively participate.  This alignment of expectations between what is expected by the  teacher and what the student expects to are central to the chance for a successful reflective dialogue. "There is no end point in the search of knowledge and sound practice" only conversation "And conversation means taking seriously the critical view points of others" (Barnet, 1992).

In the method for my enquiry, I discuss the idea of the Patchwork Text (Winter 2004).  This approach to learning in Higher education is based upon the ideas explained here, but develops them further to include a reflection of the reflection initiated through collaborative learning groups.  This is a multi-layered approach whereby learners are 'working on the significance of reflection itself, that is learning about how they learn!".  In working on my own dissertation, I was not part of a formal learning group but did want to 'learn with others'.  To this end I created a series of learning activities that enabled me to work and co-reflect with colleagues using technology as an enabler of this process.

Facilitation of learning

In a formal learning context, the teacher is the 'engine' for creating the conditions for learning.  In this I include the teacher interactions and relationships with students and the online learning environment and experiences they create that combine to produce facilitated learning.  Throughout my dissertation, I am using a critically reflective model of learning and an important question from a practical standpoint is to what extent does the ability to critically reflect need to be taught?  Is it something that we can all readily do or is it something that to varying extents is a skill that needs to be learned.  If the latter is the case, then how can it be promoted through the activities of others?  What are the conditions required for reflection and how can they be developed and promoted in online social learning?

In an online context, Bradshaw, Chapman, and Gee (2002) identify the importance of the teacher.  Their definition of teacher is broad and includes the roles of tutor, assessor, coach, mentor, and facilitator.  Theses roles are not seen as mutually exclusive, but represent 'hats' that an online teacher may need to interchangeably wear.  Supporting theses interactions, they see the learner - learner and teacher - learner relationships as being of central importance.  For Bradshaw, Chapman, and Gee, the development of a 'community' of learners is characterised by conversations that demonstrate trust, sharing, support, openness, and disclosure and deep learning.

Processes and modelling are manifested by the actions of the teacher in providing the opportunity for learning based on the intended learning outcomes and the task/s being undertaken (learning activities) to achieve them.   Moreover, the teacher by their actions can impact on the student, so by explicitly modelling in their process critical reflection and taking on the role of a co-learner the critical reflective behaviours that are desired will be better promoted. "Most human behavior is learned observationally through modeling: from observing others one forms an idea of how new behaviors are performed, and on later occasion this coded information serves as a guide for action" (Bandura 1977).

The concept of personal stance (Brockbank and McGill, 1998) relates to both student and teacher and captures the context, personal experience, and viewpoints from which we approach a learning situation, "how we place ourselves, within any given learning context, whether formal or informal, is fundamental" (Salmon, 1989).  As a learner, our view of the material being covered is inevitably influenced by our perceptions of the teacher.  Teachers convey more than disembodied knowledge by their teaching as their interpretation and viewpoint is inextricably linked in shaping the knowledge they convey.  Likewise their enthusiasm or antipathy towards a chosen topic is bound up in the way they present it.  For Brockbank and McGill, the important issue is not just for student and teacher to recognise this but to actively engage in a dialogue of understanding each others position so that each can develop their own understanding from a position of transparency as the relationship teachers develop with their students is a large determining factor in success.  Research by Professor Russell Bishop (2004) into achievement in a school context, also identifies the teacher / pupil relationship as being the single most important determinant in success of the pupils.

Given that I am enquiring into formal learning contexts, the role of the teacher is of central importance.  To what extent have they create the conditions for learning that promote critical reflection and the relationship with the student are of central importance to the success of the learning being undertaken.

Learning from experience

The importance of context and personal experience has already been highlighted in the conditions required for critical reflection.  The knowledge that practitioners possess is tacit or implied knowledge until it is made explicit.  Oakenshot (1962), uses the terms 'practical knowledge' and technical knowledge', "Technical knowledge is capable of written codification; but practical knowledge is expressed only in practice and learned only through experience with practice".   Brunner (1996) goes further and describes the state when we know how to do something but are unable to explain conceptually what we are doing, or perhaps even why. This view of the 'psychology of work' is described by Brunner as "Knowing as Doing" and it has implications for how we learn, in that we understand by doing something other than just talking about it.

Eraut (1994) emphasizes the centrality of context to knowledge, in that context influences not only what knowledge gets used but also how. The learning of knowledge is not a separate act in itself, but the learning takes place as the knowledge is applied to the situation and transformed to meet the particular context thereby creating new knowledge. Transferring knowledge to new context is in itself not a straightforward process of taking a 'template' a replicating it elsewhere as the unique set of circumstances will require new learning to be developed.

Kolb (1984, p41) defines learning as "the process whereby knowledge is created through the transformation of experience. Knowledge results from the combination of grasping and transforming experience" (Kolb 1984, p41).  Kolb is identifying the important role that experience plays in learning in contrast to cognitive learning theories that give primacy to cognition over affect and behavioural learning theories deny a role for subjective experience.  Kolb's model proposes two dialectically related modes of transforming experience: concrete experience - abstract conceptualization and reflective observation - active experimentation.  The model implies choice for the learner in selecting the learning abilities they will use for a particular context of learning.  This fits well with the concept of self-direction and recognises that some will prefer to learn through experience and others through abstract conceptualization using symbolic representation (reading).  Equally when transforming or processing information, we may prefer to either observe others (watchers) and reflect, or engage (doers) in the activity ourselves.

Individuals display this choice between concrete or abstract and reflective or active in "patterned characteristic ways" (Kolb, 2001) and these are their learning styles.

The four basic learning styles Kolb identified pose a challenge for any learning design and are particularly challenging for work online.

• Diverging - prefer to work in groups, listen with an open mind and receive personalised feedback.
• Assimilating - prefer readings, lecturers, exploring analytical models, having time to think things through.
• Converging - prefer to experiment with new ideas, simulations, laboratory assignments, and practical applications.
• Accommodating - prefer to work with others to get assignments done, set goals, do field work, test different approaches to completing a project.

The recognition of the importance of learning from experience has clear implications for learning in an online social context.  It emphasises the importance of realising that learning takes place away from the online context and if it is to be successfully supported it will utilise more than just online conversation but will have a range of strategies that recognise the learning that takes place elsewhere through working with others, trying things out, and reflecting on experiences.

Self-directed learning

There is a clear tension here between the role of the teacher in empowering individuals to be self-directed but at the same time providing the conditions for purposeful learning.  The same importance of the question around the innate ability of students to be critically reflective, is relevant to what extent is adult learners are naturally self-directed.   Martin (2002) offers four choices as to what self-directed learning might be "Is it a characteristic of all adult learners? Is it a goal toward which adult learners should be moved? Is it an instructional method naturally employed by adults? Is it an orientation toward a trait or learning style that individuals possess in varying degrees?"

Brockbank and McGill (2003) identify a desire to support transformational or deep learning within Higher Education.  By this they mean the development of the individual who as a self-directed learner has the capacity to increasingly take control of their learning and operate in a self-critical way in terms of analysing their chosen discipline and approach their own learning.  This developmental view implies the need to support the learner towards a state of self-direction in their learning.  Hiemestra (1994) emphasises the "responsibility for accepting any consequences when individuals take control of their learning.  Hiemestra and Sisco (1990) refer to the support as  "individualizing instruction" which is the mechanisms external to the individual that make it possible for them to become self-directed learners and these are largely influenced by the teacher.

Together these views identify as important the capacity of the individual to make informed choices about how their learning will take place, but also accepting that with this comes the need to accept that there are 'hazards' to this approach and that mistakes or poor choices will be made.  These descriptions of the characteristics of the self-directed learners point towards the need for conditions that support the learner in working towards self-directed learning, but also identify the need for the learner to be self-directed internally.

Knowles (1984) model for adult learning pulls together the ideas of self-direction and learning from experience, but also adds the notion of life-, task-, or problem-centered orientation to learning as opposed to a subject-matter orientation.  This belief that personal significance is important can be traced back to the work of Dewey (1916), "the alternative to furnishing ready-made subject matter and listening to the accuracy with which it is reported is not quiescence, but participation, sharing in an activity.  In such shared activity the teacher is the learner, and the learner is, without knowing it, the teacher".

For Knowles (1980), a logical outcome of these assumptions is the use of a collaborative teaching model that involves the learners as partners.

In the online context, Stephenson (2001) developed a paradigm grid where he identified the importance of aligning the expectations of learners with those of the teachers in terms of the pedagogical approach to be taken.

Fig 1 Online Paradigm Grid (after Stephenson 2002)

The case being made for teacher and student to work together in developing ths conditions for self-direction is a strong one, however In Figure 1 it can be seen that if the teacher and learner are to use the discourse of the self-directed learner, then the actions of both need to reflect this if conflict and misunderstanding about purpose is to be avoided.  Self directed learning is located firmly in the SE quadrant, but if the teachers actions or students expectations are located in the NW quadrant there will be conflict. 

Argyris and Schön  (1974) recognized that professionals although able to explain their approach to a particular problem (espoused theory), may in fact practice a different approach altogether or a theory-in-use.  I would suggest that this tension may manifest itself particularly in online contexts where the 'invisibility of students' and relative lack of ability to see what they are doing tempts teachers to move toward the teachers controlled / specified tasks as a means of dealing with their own insecurities.  The use of technology, and the model of learning we adopt is vital in empowering learners to control and direct their own learning although empowering learners is not the same as leaving them to their own devices.

Learning through online technology

At the heart of my approach to my dissertation is online technology.  This is founded on my belief that the opportunities it offers are significant in terms of promoting critically reflective approaches to learning that are more than conversations in online learning communities.  Although identified as a separate key theme of my enquiry, it is integral to the whole process, but it is in terms of the findings and conclusions that the theme is most significant.

Analysis of practice

This review of Ultralab's and by association my practice is an attempt to identify the underpinning philosophy and theories used in the development of the Ultralab model for online learning.

Over the past 5 years I have developed my professional understanding in the field of 'online social learning' within a set a set of orthodoxies that have been developed at Ultralab since the mid 90s.

Online Learning Communities have been a key strategy for the delivery of continuous professional development programmes for school leaders in England over the past 5 years.  Ultralab has been at the forefront of some of theses projects including Talking Heads (TH), an online non-formal professional learning community for Headteachers (TH team, 2002) and other large projects such as the formal learning communities as a part of the National Professional Qualification for Headteachers (NPQH) programme for aspiring Headteachers (Bradshaw, Chapman, Gee, 2002).  My experience of working at Ultralab in online communities has included TH and facilitating the MA module 'Online Teaching and Facilitation' (OLTAF).

The theoretical starting point for Ultralab's work in online learning communities are a collection of papers written by Heppell, Ramondt, and Chapman.  As well as these papers, it is reported in conversation by these and other Ultralab staff that there was also much discussion that lead to a conceptualisation of what online learning communities are, how they function, and how they should be supported.  My Blog contains some discussion with Ultralab colleagues around Ultralab online learning orthodoxies, and in particular a discussion on discourse presentation and framing which sheds some light on their thinking in this respect and the high level of importance attributed to it.

More recently, published work by Bradshaw, Powell, and Terrell have built upon these ideas introducing additional perspectives and ideas in the domains of online learning, community learning, and adult professional learning. 

The collection of papers in table 1 (below) is selected from those published and available from Ultralab.  I have selected the 5 papers listed as a representative of the literature published by Ultralab relating to online learning communities run for adult learners.  I shall be focussing my critique on the conceptual background rather than a general critique of the papers.

Table 1

The University for Industry project (UfI) sought to bring together a diverse group of learning professionals (including trainers, museum education officers, teachers, etc) into a non-formal learning process with largely unstructured opportunities for learning and little instruction from the part of the facilitators of the project.  This paper (Heppell and Ramondt, 1998) articulates Ultralab's formative view of online learning with little addition or deviation until subsequent papers of Bradshaw, Powell, and Terrell (2002).

Heppell and Ramondt (1998) clearly identify the philosophy in informing the design of the University for Industry project as being constructivist.  In terms of pedagogical design, this had implications for developing an environment where discourse is at the heart of the learning process.  Further, it leads onto the notions of a learning community and the key concepts identified by Ramondt and Heppell that characterise such a community:

• The centrality of self directed participation to learning communities
• That a sense of ownership can be engendered through the provision of tools for learning
• The evolutionary process inherent in online communities
• The role of informal, social conversation and the exchange of stories to build a sense of community
• The role of describing place to create a sense of familiarity and ease
• The influence of computer software on shaping the dialogue

Ramondt and Heppell (1998)

It appears from these key concepts that Heppell and Ramondt placed a high emphasis on the 'self generating' nature of online learning communities enabled by empowerment of users through 'usable' technology.  It also appears that 'informal' conversation was seen as the primary learning protocol.  The research showed that this lack of structure posed a problem for some researchers "On several occasions some participants requested, and indeed anticipated, more structure."  However Ramondt and Heppell believed that this would conflict with a key concept of self-directed learning "These suggestions were resisted because the project aimed to engender self-directed learning, and this may in fact have been part of the dilemma."

A premise underpinning the project was that "online communities flourish when the participants are self-directed and participate in designing their learning."  To promote learning, three routes to learning were identified and offered to the participants.  Learning about learning; where www resources were offered as starters to stimulate reflection and discussion with experts encouraged to join in.  Learning about good practice; where participants started discussions on topics of professional interest.   Learning through exploration; where participants constructed synchronous www based 'filed trips'.

The project was run around sessions each of which had a topics of focus.   Whatever the focus, it appears discussion was the 'tool' used for learning.  Clearly notions of self-direction and reflection were though important, but the conceptual model for these is not clear.   Was reflection viewed as simply thinking about something, or was it Kolb's "active, persistent, and careful consideration", or Boud, Keogh and Walker's "recapture their experience, think about it, mull it over and evaluate it', or Schön's reflection on-action, or Eraut's particular view of reflection in-action.  Whatever Heppell's and Ramondt's conceptualization of reflection, there is no clear articulation of this or evidence of this being articulated to the participants.

Likewise, the Heppell and Ramondt identify self-direction as important.  However, their conceptualisation and articulation of what it means to be a self-directed learner in an online learning community is not evident.  This point is discussed when I review self-direction in the following section.

A research finding from Ramondt and Heppell indicated that they believed that this open structured approach with an eclectic collection of participants was a successful model "The hypothesis that an eclectic mix of professionals with a broad but common focus ("learning" in this case) can learn from and with each other already can be viewed with confidence."  The mixture described has at best a loosely defined domain of learning, but the lack of common enterprise, and by definition the manufactured nature of the community would seem to suggest that at best this was a fragile community that would struggle to thrive over a longer period of time beyond the duration of the research project where the 'impetuous' of the project itself creates some participation.

Bradshaw, Powell, and Terrell (2002) in their presentation to Bera outlined components of a conceptual model for learning in online communities.  Philosophically they concur with the earlier expressed underpinning constructivist philosophy, but also begin to identify the importance of different approaches other than that of learning through conversation.  For example, they identify a metacognative approach to learning "the thinking about thinking which informs decisions about what to do next" (Eraut, 1994, p144 in Bradshaw, Powell, and Terrell, 2002) as having potential implications for learning in online communities.  They also identify other models and theories such as Lave and Wenger's work on Communities of Practice (1991), which locates learning in social participation.  However, in what might be an over-simplistic adoption of these theories for online communities they focus on conversation as the tool for community learning.  The research by Lave and Wenger was into communities that live an work together and learning for them was more than just conversation but included demonstration, collaboration around solving problems, observation, and instruction.  

In articulating a view of self-direction, Hiemestra's work is used to explain that for individuals self-direction is "the ability to assess, plan, implement, and perhaps evaluate effort"  (Bradshaw, Powell, and Terrell, 2002).  This conceptualisation of self-direction begins to address the deficiencies in some of the earlier Ultralab, although it stops short of articulating this in terms of a model for online social learning.  That is what does this look like in terms of the facilitation of participants in these these online communities?

In recognising Palinscar's strategies for structured dialogue "summarizing, question generating, clarifying and predicting" (Palinscar, 1996. in Bradshaw, Powell, and Terrell, 2002) conversation is identified as the process for learning.  In doing this, however, Bradshaw et al are moving away from the notion of 'chance conversation' towards a view of conversation with a purpose that may be facilitated in some way. In using the term 'chance conversation', I am describing a relatively unstructured approach to learning with broad topics of conversation but with little use of other approaches to learning that could be supported through online communities.  In some ways, theses conversations can be likened to conversations around the 'water cooler' identified amongst others by Brown, Duguid, (1999) when discussing informal 'communities of practice' in organizations. I would contest that 'chance conversation' is but one protocol of social online learning.

The Talking Heads Report (TH team, 2003) is an extensive 'history' of the approaches and strategies used to promote online learning communities for one of the largest research projects of its kind in the world.  The project ran for some four years and was aimed at bringing Headteachers around the UK together in non-formal learning communities.

The report identifies conversation as a primary strategy for learning in online communities "Palinska (2001) is most explicit in considering dialogue.  It is by this means that learners are provided with scaffolded instruction.  It is through this scaffolding that social learning takes place (Brunner, 1996).  It is recognised that learning through dialogue and reflection is most appropriate for deep (Biggs & Moore, 1993) or independent learners (Jih & Reeves, 1992)" (TH team, 2003).   Notions of conversation, reflection, and a social approach to learning are put forward as components of a model for online learning.  However, there is a little if any articulation of how these espoused values are to be applied to a model of online learning.

Running parallel to the talking Heads projects were a raft of online learning communities linked to formal learning programmes at the NCSL.   A report by Bradshaw, Chapman, Gee (2002) outlines the Ultralab model learning in these communities.   Once again, at the heart of the approach taken is conversation and reflection "Powerful online tools can support learning programmes by combining pre-existing or purposely designed course content with collaborative online learning environments and reflective dialogue."  Again, reflection is described as being at the heart of the learning process, but little explanation or indication is given as to how this would manifest itself as a model for online learning. 

There is a strong link being made between the conversation in the online community and workplace practice "In engaging with online learning communities, learners are synthesising knowledge and reflecting on their practice."  However, this is a 'passive view' of incorporating workplace practice in the learning process, as it does not imply action simply the sharing of experience through chance conversation. 

Conclusions

My review of Ultralab practice indicates three main findings:

• a high reliance and emphasis on 'chance conversation
• a belief in closed, but 'expertly' facilitated communities
• a philosophical commitment to reflective practice with little explanation of what it is or how it is to be achieved
• a belief in the value of self directed learning with little explanation of what it is or how it is to be achieved

Figure 2

Figure 2 attempts to illustrate the characteristics of discourse that are possible in online communities depending upon the conditions encouraged by the teacher.  The grid illustrates how the 'framing' of the discourse influences the opportunity for learning.  Chance conversation resulting from a discourse with an ill-defined purpose and facilitation that lacks strategies other than to keep the conversation going.  A discourse with a clear purpose and a facilitator with a range of strategies that include the developing of the individual's ability to critically reflect and to become a self-directed learner will result in a purposeful discourse with a potential for 'deeper' learning.

I would argue that it is in the bottom left hand quadrant that the most impoverished experience of learning takes place where there is no clearly identified purpose to the discourse and facilitation fails to use a range of approaches to learning and relies on chance conversation.  This is in fact the dominant model for online community learning.