Connectivism: Learning as Network-Creation

Connectivism: Learning as Network-Creation      George Siemens        August 10, 2005
Table of ContentsAbstract                                                             3Introduction                                                        4What is a Network?                                             5Types of Nodes                                                   6Forming Connections                                           8Creating Meaning                                                13Characteristics of Learning Networks                     17Use of Learning Networks                                          21Corrections within Networks                                 22Networked Learning and Connectivism                   23Ecology                                                              24Implications for Higher Education                              25
and Corporate Training Conclusion                                                          26 References                                                          28
Abstract Existing theories of a particular subject matter are typically revised and adjusted to reflect changing environments. At some point, due to continual revisions, the theories becomes so dichotomous and complex that it is no longer reflective of the subject it is intended to define and explain. At this point, the existing theories need to be replaced with models that more accurately reflect the link between theory and reality. The domain of learning is significantly hampered by progressive revisions of what it means to learn, to know, and to understand. A subset of connectivism, network forming, is presented as an accurate model for addressing how people learn. The test of any theory is the degree to which it solves problems and incongruities within a domain. The shortcomings of behaviourist, cognitivist, and constructivist ideologies of learning are answered in light of learning as a connection-forming (network-creation) process.
"Knowledge is of two kinds. We know a subject ourselves, or we know where we can find information on it." (Samuel Johnson)Introduction:
Our metaphors of learning have become tired and worn. Skinner presented the “black box” of behaviourism (we don’t know what happens inside, so we just focus on the behaviour). Ausubel and others presented a computer-processing model (inputs, processing, coding for retrieval and outputs). More recently, constructivism has been presented as a free-floating theory of learning as an individually constructed experience.
Underlying each theory is a deeper ideology and worldview. Philosophers, psychologists, theorists, and linguists have long debated the nature of learning and knowing. Is learning the process of aligning with objective, external knowledge and truth (objectivism)? Are learning and knowledge an interpretive process (i.e. we learn as we experience, and truth is revealed through our action and cognition) (pragmatism)? Or is learning a process where we create our own truth through our own perspective of the world (interpretivism)?
Something is missing in this debate. Knowledge and truth can exist in a variety of ways. Different perceptions of what it means learn (or possess knowledge) do not need to be seen as exclusive. To some degree, objectivism, pragmatism, and interpretivism provide partial insight into a specific aspect of the learning and knowledge process. The nature of the subject matter and the learners themselves impact which approach to knowing will be most beneficial to learners. In contrast to these established views of learning, connectivism presents learning as a connection/network-forming process.
A caution to readers: most papers are intended to allow the author to express what he knows or has come to understand. This paper is largely intended to be a public grabbling of personal dissatisfaction with the learning process as conceptualized over the last century. It is my intent that this paper will provide opportunities for dialogue and exploration with others on how a network-forming model of learning provides insight into our knowledge and information needs in today’s world. The discussion will be held on the blog, discussion forum, and email list at:http://www.connectivism.ca.
What is a network?
The beauty of networks is their inherent simplicity. A network requires at minimum two elements: nodes and connections. Nodes carry different names in other disciplines (vertices, elements, or entities). Regardless of name, a node is any element that can be connected to any other element. A connection is any type of link between nodes.
Various factors influence the capacity of nodes to form connections. Once a network has been established, the flow of information can move from one domain to another with relative ease. The stronger the connection between nodes, the more rapidly information will flow.
The information system underlying network creation includes:
Data – a raw element or small meaning neutral element
Information – data with intelligence applied
Knowledge – information in context and internalized
Meaning – comprehension of the nuances, value, and implications of knowledge
This information system is a continuum, and learning is the process that occurs when knowledge is transformed into something of meaning (and will then generally result in something that can be acted upon). During this process, l learning is the act of encoding and organizing nodes to facilitate data, information, and knowledge flow.
Types of Nodes
Virtually any element that we can scrutinize or experience can become node. Thoughts, feelings, interactions with others, and new data and information can be seen as nodes. The aggregation of these nodes results in a network. Networks can combine to form still larger networks (each node in a larger network can be a network of nodes itself). A community, for example is a rich learning network of individuals who in themselves are completed learning networks.
Nodes are characterized by a general sense of autonomy. A node may exist within a network, even if it is not strongly connected. Each node has the capacity to function in its own manner. The network itself is the aggregation of nodes, but can only exert limited influence on the nature of each node in the network.
While networks are simple in nature, numerous elements impact the flow and dynamics of connection creation. Elements and characteristics of a network include:
Content (data or information)
Interaction (tentative connection forming)
Static nodes (stable knowledge structure)
Dynamic nodes (continually changing based on new information and data)
Self-updating nodes (nodes which are tightly linked to their original information source, resulting in a high level of currency (i.e. up to date)
Emotive elements (emotions that influence the prospect of connection and hub formations).
Data and information are database elements (i.e. they need to be stored and processed in a manner which permits them to be dynamically updated within existing networks). As these elements update, the entire network structure similarly gains and benefits. In a sense, the network grows in intelligence. Knowledge and meaning, on the other hand, receive their worth from the underlying data/information elements.
Forming Connections
Connections are the key to network learning. Yet not every connection has equal weight and influence in the entire structure. Connections can be strengthened based on a number of factors:
Motivation – motivation is a difficult concept to fully detail. The difficulty arises in that motivation is influenced through our emotions and logic. An individual with a clear goal may have significantly greater motivation to learn a new subject. Keller’s (1987) ARCS model communicates part of the challenge evident in fostering motivation. Numerous factors of attention, relevance of information, our sense of competence, and satisfaction impact the likelihood of connection forming. As discussed in the section Cognitions, Emotions, and Learning section of this paper, learning is a process of encoding nodes and forming connections. Motivation determines if we are receptive to particular concepts as well as our desire to foster deeper network connections through the items listed below (reflection, logic/reasoning, etc.).
Emotions and how we feel play a large role in how we value nodes and permit the presence of contradictory perspectives. Consider as an example, the phenomenon of global warming. General consensus exists that we are at least partially to blame. For many people, however, only limited lifestyle changes have occurred. How can this concept be seen in light of connectivism (or network learning)? Quite simply – the emotional and motivating nodes (the network itself is simply a node in a larger network. The process scales and reduces based on the size of the network we are considering) evaluate perspectives of global warming in relation to daily lifestyle (commuting, recycling). Until the node of “global warming” begins to impact the quality of life for the individual, many may continue to be comfortable with their global warming node, and route around it. Others may be more sensitive, and place greater emphasis on the concept of global warming because it more seamlessly integrates with the existing network. Reworking and recreating a network at its core takes time. Emotions are the influencing factors that enact other nodes and apply weighting scales to the network elements.
Exposure – repetition is an excellent way of strengthening connections. A node grows in popularity (relevance) as more nodes link to it. Ideas that link strongly to other ideas are quickly integrated into the network. This is a big factor for the difficulty of personal change. The first glimmer of thoughts of change (quit smoking, exercise more, eat healthy) result in a rogue node. The node exists but has limited traction within the entire network. As the node begins to form its own connections with other nodes (sense of self worth, happier, greater productivity at work, feel better), it gains traction and begins to link and connect to a greater degree with other nodes. The tipping point occurs when the node itself has created a strong enough network to begin to influence the entire thought process (neural network). Once it is no longer a rogue node, it continues to embed itself as node that is used by the rest of the network. Innovation within corporations follows a similar path. New ideas and processes are initially seen as threats by the rest of the organization (organism). As a result, the node is treated as a fringe element and left largely unconnected. However, if the idea (node) has true merit, it will continue to form connections within the network until it creates a sub-network of connections within the larger structure. At this point, it has the capacity to influence the larger network that originally resisted it.
Patterning is one of the most significant elements of learning. Patterning is the process of recognizing the nature and organization of various types of information and knowledge. The shapes created by these structures will determine how readily new connections can be made. For example, a learner in the field of medicine (who is aware of the nature of their own learning network) may recognize pattern similarity of elements between her field and the field of philosophy. The recognition of these patterns can result in exponential knowledge growth through connecting similar network elements. Duplication can also be minimized. This concept is quite interesting in light of how network theory was popularized in recent literature. Sociologists have spent decades exploring and detailing network phenomenon in a social sense. Several years ago, physicists began exploring networks in greater detail. The significant pattern-similarity between these two fields of research was quickly noticeable (though some would argue that physicists like Barsabi (2002) have popularized network concepts somewhat exclusionary to the work of sociologists). Syncretically fusing experiences of various domains provides substantial benefit in the process of learning formation. Much new knowledge will be generated in the future as specialized fields become more aware of each other.
Logic is a cornerstone in the learning process. Much of what we know (and have learned) is the by-product of thinking and reflection (reflection is much like logic but permits greater interplay of emotions). The process of thinking involves organizing and structuring our learning networks. As a reflective activity, logic can provide time for nodes to form connections. Connections can be formed without conscious thought, but the process is substantially improved when directed by focused reasoning. Logic is an ingrained connection-forming task, evaluating and recognizing patterns between different concepts and network elements. Cognitive neurology is rapidly developing our understanding of logic and cognition.
Experience is also a significant aspect of network creation. A great deal of our learning comes through informal means. Experience is a catalyst for both acquiring new nodes and forming connections between existing nodes. Learners who graduate from university or college often have the information and knowledge nodes, but connections themselves do not form fully until the learner is active within his/her field. Experience in this sense is largely a facilitator of connection forming.
Can learning be both an influence and be influenced in the network forming process? If learning is the activity that occurs between knowledge and meaning, it is obviously an object that is being acted upon, that is it is an influenced element in network forming. Learning itself is also an influencing element in that the actual process is one of network creation and forming. The strong reflexive and iterative aspects of learning contribute to its frequent misclassification as largely a content consumption process. Learning cannot be viewed only as a passive (acted upon) or active (acting upon other elements) process.
How does a network view of learning relate to other established theories of the instructional process (instructional in the sense of formal education, though it is no longer an exclusively lecture format)? Chickering’s work on good practices in undergraduate education provides a strong link to networked learning theory:
Good practice in undergraduate education:
1. encourages contact between students and faculty,
2. develops reciprocity and cooperation among students,
3. encourages active learning,
4. gives prompt feedback,
5. emphasizes time on task,
6. communicates high expectations, and
7. respects diverse talents and ways of learning.
Similarly, Gagne’s Nine Events of instruction integrate with a network-forming theory of learning:
1. gaining attention (reception)
2. informing learners of the objective (expectancy)
3. stimulating recall of prior learning (retrieval)
4. presenting the stimulus (selective perception)
5. providing learning guidance (semantic encoding)
6. eliciting performance (responding)
7. providing feedback (reinforcement)
8. assessing performance (retrieval)
9. enhancing retention and transfer (generalization)
Creating Meaning
Meaning in a network is created through the formation of connections and encoding nodes. The presence of a new node, however, does not ensure learning. The addition of a new node within a network does not ensure knowledge transmission or transference of meaning. The node must first be encoded and connected to other elements of the network.
While much of the field of cognitive scientists follows network learning models, a difficult challenge is to explain the presence of contradictions. Learning schema (Driscoll, 2000) expresses that much of our learning is actually the process of creating hierarchical structures of knowledge that we organize based on experience and exposure to new ideas. In order to prevent cognitive dissonance, new ideas are subsumed (correlatively or derivatively) within the existing structure (Driscoll 2000, p.120). Most people, however, have significant contradictions in their reasoning. The presence of contradictions supports the notion that “rogue nodes” are part of our network. While cognitive dissonance may still occur, the rogue nodes can remain present if they are not significantly linked to the entire network (i.e. the nodes are part of the network, but are generally not active hubs or information relay points).
Meaning is not evaluated on only one level. It is a by-product of a complex process of evaluation and reflexivity. The process is iterative and messy. Optimal meaning-generation follows general attributes of systems: open, adaptive, self-organizing, and corrective.
Latent Semantic Analysis provides an interesting observation of how learning can sometimes occur significantly beyond the value contained within a particular element. This phenomenon can be explained by the process of including a new node within an established network structure. The new node suddenly provides connection and knowledge flow across the entire network. As a connective element, the node may serve as a hub through which new information is routed, or may instead simply permit connections between ideas and concepts that previously did not have connections with each other.
Siemens (2004) presents the subject of Latent Semantic Analysis in light of networked learning theory:
“Landauer and Dumais (1997) explore the phenomenon that “people have much more knowledge than appears to be present in the information to which they have been exposed”. They provide a connectivist focus in stating “the simple notion that some domains of knowledge contain vast numbers of weak interrelations that, if properly exploited, can greatly amplify learning by a process of inference”. The value of pattern recognition and connecting our own “small worlds of knowledge” are apparent in the exponential impact provided to our personal learning.”
As mentioned earlier, data, information, knowledge, meaning are the main elements of the learning cycle. Learning itself occurs in the domains of knowledge and meaning. Which nodes are activated and when is a function of two elements: logic/cognition and emotions. Learning has too long been considered an exclusively cognitive activity. Cognition and emotions are intertwined to the point of blurring – each feeding the other in a continual dance of feedback and reaction. Meaning making happens through this dance.
Learning is not the content consumption process the formal education system perceives it to be. A simple experiential example of how we approach and read a book illustrates this concept. If the subject is in an area that connects well to our existing viewpoints, the material is quickly absorbed and integrated. If the content is in conflict with the overall health of our learning network, the material is acquired (i.e. transformed into nodes within our learning network) more slowly.
How is the new information coded into nodes? Rarely is the information captured directly or completely mirrored from the original source (with the exception of quotations and poetry – two devices which tend to duplicate themselves completely). Instead, new information is coded as vague representations (reflective of the network in which it becomes embedded). Our manner of coding a new node into our existing network is at least partly personal and experiential. Acquiring new knowledge is not a direct transference process. Instead, the process is best symbolized as a process of attempting to transfer the original idea within the original context (context in this case includes the author’s context as well as the context of the learner at the time of encountering the new knowledge).
Meaning is transferred in a rich, but messy process incorporating the content, the context of learner and resource creator, as well as the cognitions and emotions of the learner at the time of knowledge acquisition. We do not learn from books (or any other information source) in the sense many psychologists have expressed through theories of cognitive and constructivist theories. Learning is a “door opening” process which first permits the capacity to receive knowledge, followed by encoding the knowledge as a node within our personal learning network. Of equal importance, is how our emotions and cognitions influence the creation of integration with other nodes within the larger network. A new node that links well with other nodes is quickly integrated. A conflicting node may still exist, but requires a greater period of time to establish an information transference route to other nodes and networks.
Characteristics of Learning Networks
Many attributes of networks from sociology and physics transfer easily to the concept of networked learning.
Small world effect is based on research carried on my Stanley Milgram. He discovered that most nodes within a network are connected by a fairly short path. Information flow from one domain of a well developed network to another generally requires a small number of “hops”. A learning network has similar short paths between information elements.
Weak ties are links or bridges that allow short connections between information. The concept of weak ties expresses the understanding that much of our information comes from weak connections to networks other than our own. Our personal network is populated with nodes that are integrated with (or at least similar to) our own. For substantial innovation to occur, we often rely on loose/weak connections to other less familiar networks. These connections provide a view into modes and manners of thought which differ significantly from our own.
Scale Free networks are detailed by Ravid and Rafeali (2004):
“In Scale Free networks the distribution of different network parameters act in an exponential fashion. The most interesting and most measured exponentially distributed parameter is the distribution of connections from each node outwards (Out Degree). This uneven distribution means that in these networks some of the members are connected to a lesser degree and some of the members are connected to greater degree, which is how they hold a senior position in the network (Goh, et al., 2002). Networks of this type are relatively resilient, but are not at all immune to attack. In other words, a random removal of network members (a crash) will not hurt its stability, but a directed removal of key points — hubs — will cause the network to quickly collapse. On Scale Free networks, the distribution of density or congestion is constant and is not dependent on the exponential coefficient of the distribution of the number of connections (Jeong, 2003)”.
Centrality relates to the structural position of a node within a network. Centrality details the prominence of a node and the nature of its relation to the rest of the network. Visual Analytics (2005) provides a short overview of how the centrality of a node is influenced by factors of:
Degree – Determines the root by identifying the object with the most direct connections to other objects within the network. This finds the object with the most influence over the network.
Closeness – Determines the root as the object with the lowest number of links to all other objects within the network. This finds the object with the quickest access to the highest number of other objects within the network.
Betweenness – Determines the root as the object between the most other linked objects. This measure finds objects that control the information flow of the network, sometimes referred to as “gatekeepers.”
To make networks effective, some type of monitoring and overall quality determination is required. In a learning context, our emotions and logic play the gatekeeper role. They determine which nodes take root, and which nodes are exposed to which connections. Our emotions and cognitions, however, don’t always work in harmony. Consider the case of an individual who has a fear of flying. Cognitively, this fear is largely unfounded (air travel is one of the safest forms of transportation). Yet emotionally, the node itself has taken root and formed connections.
Control and knowledge flow
Knowledge and information flow occurs through various nodes in a manner reflective of the existing network (a concept we usually refer to as perspective). Two learners may experience the same information, yet code the new node into their network in different ways. What is conclusive to one learner, may be absurd in the network of another.
How then does knowledge flow within a network? Which factors impact the process? If we tentatively ascribe life-like properties to our learning networks, we can partly answer this challenge. Any living organism seeks two primary functions: replication and preservation. Nodes within our networks follow similar aspirations. Established beliefs and learning often ensure that new information is routed through (i.e. contextualized) the existing network. New information is evaluated and coded reflective of the existing meme (or zeitgeist) of the learning network. A simple illustration: if one believes that people can’t be trusted, the activities those around will be interpreted through this framework (i.e. routed through our neural network and coded with meaning reflective of this larger view). Meaning is attached as an “add-on” to the knowledge source, ensuring that the existing network replicates itself. If the entire network is subsequently reconfigured according to a new meme, the knowledge itself stays, but the meaning is also reconfigured.
In a similar sense, when knowledge is introduced to a learning network which is contradictory to the established structured, the existing network, in an effort to preserve itself, attempts to route around or push the new node to the fringe (ensuring that limited connections are formed, and as a result, the new node does not gain significant status with the larger network). If the node does acquire a certain level of status, new knowledge may route through the node, permitting the node to begin replicating itself (i.e. encoding meaning to knowledge).
Flow inhibitors are elements internal to a network that reduce the possibility of information and knowledge flow. Most often this will include elements like biases, preconceived notions, or lack of flexibility. Legitimate flow inhibitors can be our own cognition and emotions. Some types of information should be inhibited due to poor fit with the existing network, or information that is simply false. External inhibitors also impact the flow of information between learners. The physical design of a space, the bureaucracy, or knowledge sharing culture of an environment will influence and determine how well information flows between networks.
Flow accelerators are elements and conditions inherent in a network that permit the rapid formation and distribution of information. Receptivity and motivation are two key accelerators. External attributes of an ecology or network also influence how well information flows. A culture of openness, recognized value of cooperation, and tools and time allotted for collaboration all contribute to accelerate network formation.
Uses of Learning Networks
Networks are constantly forming. As a dynamic process, networks can aggregate into larger structures (a network of networks). Networks can also be deconstructed into smaller structures. For example, everyone has some type of personal learning network. When an individual works for an organization, they bring their network with them, combining as part of the larger network of the corporation. In the course of our daily lives, we move among numerous networks. We are constantly acting upon and being acted upon.
Recognizing that we are continually moving in and out of networks provides an important starting point for rethinking corporate and higher education. Instead of seeing the artificial construct of a program or course as the point of learning, we can view the process of “living life” as a constant learning process. As we acquire new nodes, form new connections, aggregate into larger networks, or deconstruct into smaller structures, we are continually learning and adapting – interacting dynamically with the world around us.
Corrections within Networks
Not all nodes within a network continue to remain relevant. As an intelligent network, our minds are continually reshaping and adjusting to reflect new environments and information. Corporations undergo a similar process. Nodes that are no longer valued are “weakened” within this environment.
Weakening can occur in many ways, but the most obvious is a loss of connections within the network. For example, if I believe in the Lock Ness monster, this belief can exist as an unobtrusive node because it doesn’t generally impact my daily activities. As a result, the node is largely ignored (information and ideas are not routed through this node). As I encounter new sources of information critical of the concept of the Lock Ness monster, I may eventually weaken the node sufficiently to eliminate its relevance in my neural network. In the same sense, a learner who continually encounters new information and knowledge, will dynamically update and rewrite his/her network of learning and belief. If on the other hand, the node itself is critical (that is, it is a hub or is heavily connected), weakening will only happen over a long period of time, or through seismic shifts across the entire network, assuming that the emotional nodes which route information critical of beliefs permit fluidity of new ideas, instead of simply using new information through the perspective of existing beliefs (see the discussion on emotions above).
Learning networks are self-organizing. A designer or instructor can influence the creation of new nodes, but the receptivity (and the nature of the existing learning network within a learner) will determine how effectively new information is integrated. Rocha (1998) defines self-organization as “the spontaneous formation of well organized structures, patterns, or behaviours…”. The injection of new nodes within a learning network can often be the instigating influence to a rapid reorganization. If the node significantly informs the existing structure, it can quickly become an illuminating hub (see the earlier discussion of Latent Semantic Analysis).
Networks are adaptive. They constantly adjust and transform in reaction to the world around. Nodes within the network continually update themselves, accruing ongoing benefit to the entire structure. In a sense, we can see this phenomenon in the field of human knowledge growth over the last half-century. The stunning advancements of science and society can largely be attributed to the increased capacity of people and organizations to connect with each other.
Networked Learning and Connectivism
Networked learning is a subset of connectivism. In presenting the original theory of connectivism, I presented eight attributes:
Principle 1: Learning and knowledge rests in diversity of opinions.
Principle 2: Learning is a process of connecting specialized nodes or information sources.
Principle 3: Learning may reside in non-human appliances.
Principle 4: Capacity to know more is more critical than what is currently known
Principle 5: Nurturing and maintaining connections is needed to facilitate continual learning.
Principle 6: Ability to see connections between fields, ideas, and concepts is a core skill.
Principle 7: Currency (accurate, up-to-date knowledge) is the intent of all connectivist learning activities.
Principle 8: Decision-making is itself a learning process. Choosing what to learn and the meaning of incoming information is seen through the lens of a shifting reality. While there is a right answer now, it may be wrong tomorrow due to alterations in the information climate affecting the decision.
Networked learning relates largely to the second principle of connectivism: network forming. Additional elements will be explored in future papers.
Ecology
Networks need to occur within something. For our purposes here, this “something” is best defined as an ecology. While an ecology does have some similarities to a learning network, it possesses some unique elements that set it apart. A network is largely a structured process. Nodes and connectors comprise the structure. In contrast, an ecology is a living organism. It influences the formation of the network itself. For example, each learner in a college possesses a personal learning network. The health of this network is influenced by the suitability of the ecology in which the learner exists (in this case, the college). If the ecology is healthy, it will permit networks to flourish and grow. If the ecology is not healthy, networks will not develop optimally. The task of educators and trainers is to create and foster a learning ecology that allows learners to quickly and effectively enhance their existing learning.
Implications for Higher Education and Corporate Training
The connectivist view that learning is a network creation process significantly impacts how we design and develop learning within corporations and educational institutions. When the act of learning is seen as a function under the control of the learner, designers need to shift the focus to fostering the ideal ecology to permit learning to occur. By recognizing learning as a messy, nebulous, informal, chaotic process, we need to rethink how we design our instruction.
Instruction is currently largely housed in courses and other artificial constructs of information organization and presentation. Leaving this theory behind and moving towards a networked model requires that we place less emphasis on our tasks of presenting information, and more emphasis on building the learner’s ability to navigate the information (i.e. connectivism).
Blogs, wikis, and other open, collaborative platforms are reshaping learning as a two-way process. Instead of presenting content/information/knowledge in a linear sequential manner, learners can be provided with a rich array of tools and information sources to use in creating their own learning pathways. The instructor or institution can still ensure that critical learning elements are achieved by focusing instead on the creation of the knowledge ecology. The links and connections are formed by the learners themselves.
Conclusion
Those who struggle to create an adequate theory of learning must admit that the process is much like stumbling in the dark. So much of our thought structure is shaped by hidden assumptions evident in our existing learning and educational systems. When attempting to move away from established approaches, a period of confusion and disorientation ensues. Many in education are beginning to venture into this transitory stage. We are moving from formal, rigid learning into an environment of informal, connection-based, network-creating learning.
Those theorists most closely aligned with the new landscape are also those who most readily acknowledge that the process is one of coming to know, rather than of knowing. The developing structure of technology, neural research, institutional reorganization (from hierarchy to network), and social impact of learning under new ideologies, is evolving too rapidly to be effectively detailed as “this is what it is”. The moment this declaration is made, the environment has shifted. Learning is an in-process activity. We need to lay aside the desire to know, and embrace instead the desire to continue to learn. Knowing is no longer a destination (though in fairness, it never was, but our learning designs and institutions created it as such). Knowing is a process of walking in varying degrees of alignment with a dynamic environment.
This simple paper is not intended to provide a complete (or even coherent) theory of networked learning. My main purpose is simply to open slightly the door leading into the larger room of modern learning needs that each reader can explore in their own time and manner. I invite readers to participate in the discussion on the connectivism website:http://www.connectivism.ca.You have to be confused before you can reach a new level of understanding anything  - Dudley Herschbach – Nobel Prize winner (Chemisty).
References
Barabási, A. L., (2002) Linked: The New Science of Networks, Cambridge, MA, Perseus Publishing.
Chickering A. , Gamson Z., (undated) Seven Principles for Good Practice Retrieved on August 10, 2005 from: http://honolulu.hawaii.edu/intranet/committees/FacDevCom/guidebk/teachtip/7princip.htm
Driscoll, M. (2000). Psychology of Learning for Instruction. Needham Heights, MA, Allyn & Bacon
Keller, J., M., (1987). The Systematic Process of Motivational Design. National Society of Performance and Instruction, 26(9) (p. 2).
Ravid R., Raefali S., (2005). Asynchronous Discussion Groups as Small World and Scale Free Networks Retrieved August 10, 2004 from http://www.firstmonday.org/issues/issue9_9/ravid/
Rocha, L. M. (1998). Selected Self-Organization and the Semiotics of Evolutionary Systems. Retrieved August 10, 2004 from http://informatics.indiana.edu/rocha/ises.html.
Theory into Practice (undated) Conditions of Learning, (R. Gagne) . Retrieved on August 10, 2005 from http://tip.psychology.org/gagne.html
Visual Analytics (2005). Centrality. Retrieved August 10, 2004 from http://www.visualanalytics.com/Products/Placements/Centrality.cfm