George Siemens is an educational theorist working at the University of Manitoba, who noticed a major gap in the learning theories present. Neither Behaviorism, Cognitivism, nor Constructivism were able to take into account the changing world, especially the technological advances, according to Siemens. Rejecting those theories, he has developed the theory of connectivism.
In brief, the theory of connectivism uses the model of a network, such as a neural network. The individual in the center of the network is surrounded by nodes in the world, nodes such as a skill, an experience, or a concept. The key is for the individual to make connections to these nodes, just like the neural network has axons and dendrites to connect to the next nerve. Learning takes place when either a) more connections are made to new nodes, or b) the connection is reinforced in different, unique ways to become stronger. Siemens argues that the key is not what flows through the pipes of the connections, but rather the pipes themselves, how big they are.
Thus, to enhance learning, the educator needs to introduce an environment rich in stimuli to promote more connections. This would mean exposure to a variety of different ideas, thoughts, people and perspectives. The more connections a person has, the overall health of the network, the stronger it becomes to new challenges (much like the strength of a web being built with more strands of fiber).
But there is more. Controversially, Siemens suggests that learning can exist within the network that we have, that it is an entity that can exist outside of our minds. In essence, there is an element of potential here, just like "potential" energy being a different form than "kinetic" energy. When I'm faced with a new task in my life, my potential learning that exists in a solid support network of my delicious account, my twitter associates, and my email address book is what counts. In this respect, knowledge of the specific content is secondary to the knowledge that exists in strong searching skills or the presence of good connections that will provide me the answer.
It should be noted that Siemens doesn't feel every node is a person... that is just one example. What I personally like about Siemens model is that it is organic; its image of learning is one that grows, that ebbs and flows in different direction. Nodes are introduced into our experience all the time, and our connections are shaded on the basis of circumstances. For someone who learned all the state capitals under forced coercion by their elementary teacher, the connection to the node of "state capitals" is shaded by the connection of the "bad experience". This could hinder the connection and make the learning weaker, or it could strengthen the connection for those that associate with pain strongly.
What's key for me here is that we finally have a theory which can conceptualize forgetting. In behaviorism, cognitivism, and constructivism, once you learn the concept, you really should remember it forever. But in connectivism, it is easy to conceptualize what is happening. The node (we'll say it is the thought "bugs are icky") exists in the network. But over time, many connections are made to other "icky" things (death, cheesy movies, taxes). If there isn't a connection between the "icky-ness" of these new items and the old, suddenly the connection to the old (bugs) becomes overwhelmed by the connections to the new, stronger ones. It is forgotten.
To avoid forgetting, one has to have more connections. In our example, one has to experience that "bugs are icky" in several different ways in order to strengthen the connection. Or, strengthen the connection that "bugs are like death", and "death is icky". For an educator, the key is not to drill the concepts over and over into the student, because while it makes a temporary connection, it will ultimately be replaced if not supported by diverse connections. The key is to provide the diverse connections. Teach that concept in as many different ways possible.
This, for good or for bad, has big implications to our schools. Marzano has advocated for the systematic "guaranteed and viable curriculum", but that isn't a very organic system. The introduction of a lot of nodes, even some of them conflicting, is required to build a strong network. If you stop and ensure mastery after each node is introduced, and introduce one node at a time, you aren't going to get very far. You have to be willing for students to not master all the nodes, that they will assimilate the ones into their network which they can. This is a huge leap of faith for educators, who insist that a student must learn "x, y and z". Indeed, the Iowa Core is somewhere in the middle... it is a pared down set of essentials that Marzano would idealize as the guaranteed and viable, but those essentials are conceptual skills that require a strong set of diverse nodes to support. It isn't just the simple node of learning the state capitals.
Once again, we are back to the point that Alan November made, the point of the digital curriculum, the point where we as educators need to connect our students more to that outside world and let them form their learning being in relationship with those perspectives out there. That is how we can assess our effectiveness.