Does Learning Grow or Is it Built?

New ferns with dew dropletsIt’s Week 1 of the Connectivism and Connective Knowledge course. This is a massively open online course led by Stephen Downes and George Siemens. I believe 1900 people have signed up for the course, so it really is huge.

With something this big, no single person can follow all the conversations or absorb all the information. It’s simply not possible. I’m planning to try to delve more deeply into a few conversations, rather than skimming lightly on the surface of many.

This week, I’m taking that to perhaps an extreme level: it’s one particular phrase and a comment about it that caught my attention. I had read Stephen’s What Connectivism Is previously, but was intrigued by the embedded comments today. Gina Minks and Diego Leal used Diigo’s sticky notes to add comments on the reading. Unfortunately, I’m not sure that you can see their annotations unless you have the Diigo toolbar installed. I want to share a snippet of the conversation here though, so anyone can read it.

Stephen argues in his post about understanding being no “more than the process of making connections”:

The point is:
– there are no mental models per se (that is, no systematically constructed rule-based representational systems)
– and what there is (ie., connectionist networks) is not built (like a model) it is grown (like a plant) (Color emphasis mine)

Gina highlighted the phrase at the end, starting with “not built,” and added this comment:

I’m not sure I agree with this. If I need to learn something, sometimes I really need to work at thinking about the new information, trying to tie it to something else I already know, or look for more information to sort the new information out in my head. It is definitely work – I search for the connections in my existing frame of knowledge, and then look through all the relevant networks I have for something to help me learn the new information. To me, his is definitly more than just “growing” a new mental model.

Here’s my response:

If you’re connecting it to existing knowledge, isn’t that sort of like a new branch growing from an existing tree? I’m not sure it’s clear here, but from Downes’ other writing, I think this is more about it growing internally, driven by the learner, rather than constructed externally. I admit I struggle with this metaphor though, and I’m not sure I quite get what he’s saying. I don’t think Downes would deny that learning can be work, but he would likely characterize that work as growing rather than building.

It may be more helpful to think of it in terms of networks of people rather than what’s happening inside your head. If you try to build a network based on a model, from the top down according to rules, is it going to be successful, or will it always be artificial and forced? On the other hand, if you can provide an environment where relationships and connections between people naturally form, you grow an organic network.

What if learning, networks of thoughts or whatever, happens the same way as networks of people grow?

I’m still not convinced that I’m not completely off-base here trying to comprehend Stephen’s argument, let alone whether I think mental models exist or not. It’s an intriguing metaphor though, especially since I do tend to be constructivist and talk about learning in constructivist terms.

What do you think: does learning grow or is it built? What metaphor for learning makes the most sense to you?

Update: This discussion continues at Metaphors and Language of Learning.

Image: ‘Reaching out
Reaching out

21 thoughts on “Does Learning Grow or Is it Built?

  1. Hi Christy,

    Thanks for the great information in this blog. I am currently completing an instructional design degree and I have found your site very helpful.

    In response to Gina, you wrote, “If you’re connecting it to existing knowledge, isn’t that sort of like a new branch growing from an existing tree?” (Tucker, 2008) I completely agree. I think Gina is really saying the same thing as Downes. When a plant grows, it starts with a seed. From that point, everything develops as an offshoot of one plant. A leaf will never grow by itself—it has to connect to the stem, which is connected to the roots. Therefore, in order to encourage the growth of a new “leaf,” or bit of knowledge or skill, we must encourage connection to the main stem. To me, this represents a tie to prior knowledge, relating new concepts to something that students already understand.

    You also asked “If you try to build a network based on a model, from the top down according to rules, is it going to be successful, or will it always be artificial and forced?” (Tucker, 2008). This question really gets to the heart of the difference between connectivism and constructivism, as described by Downes. His summation of the practical differences between these methodologies was the following:

    “In the ‘model’ approach, personalization typically means more: more options, more choices, more types of tests, etc. You need to customize the environment (the learning) the fit the student. In the ‘connections’ approach, personalization typically means less: fewer rules, fewer constraints. You need to grant the learner autonomy within the environment.” (Downes, 2007)

    It seems to me that these statements are simply describing two different means of accomplishing the same ends. In school, we are relying on the fact that the “seeds” of knowledge have already been planted. In fact, we have no way of knowing how these seeds were planted or in which direction the plant is growing. Certainly, we can pretest a student’s knowledge, but we can never predict every bit of prior knowledge or misconception with which a student enters our classroom. Perhaps this is the point of the connectivist strategy. You stated that “if you can provide an environment where relationships and connections between people naturally form, you grow an organic network.” (Tucker, 2008) This extends to the learning environment as well. If you can provide an environment where students are allowed to form their own associations with knowledge, as opposed to constructing meaning out of the associations with which we provide them, then you have created a connectivist classroom.

    Dr. Ormrod (2009) discussed the misalignment of the classic information processing theory, which views the brain as functioning similarly to a computer. This model does not adequately describe human thinking because our brains go beyond cause and effect analysis. Our thinking is spread in multiple directions, as described by Ormrod, Schunk, and Gredler (2009), who state that “messages go every which way…across areas that handle very different sensory modalities and motor functions” (p. 35-36). While this supports the connectivist concept, Ormrod et al (2009) questioned the feasibility of implementing this theology, stating “One problem with the connectionist approach is explaining how the system knows which of the many units in memory to activate and how these multiple activations become linked in integrated sequences” (p. 82). This is where the role of the instructional designer becomes crucial, sparking the “activation” of memory units and guiding how they become “linked” in the brain.

    The question was “Does learning grow or is it built?” I say that learning grows around the scaffolding that we build individually as we accumulate new information, just like a vine grows on a fence. True learning occurs after the basic connections have been made, when we can really allow our thinking to branch off in every imaginable direction. I think of it like playing the piano. When you first learn to play a piece of music, it is choppy and hesitant while you think about every note. It is only after many hours of practicing that you learn to express the “feeling” of the music, something that can only be explored once the basic connections have become automatic.


    Downes, S. (2007, February 3). What connectivism is. [Blog message]. Retrieved from

    Laureate Education, Inc. (Producer). (2009). Information processing and the brain [DVD]. Baltimore, MD: Dr. Jeanne Ormrod.

    Ormrod, J., Schunk, D., & Gredler, M. (2009). Learning theories and instruction (Laureate custom edition). New York: Pearson.

    Tucker, C. (2008, September 7). Does learning grow or is it built? [Blog message]. Retrieved from

    1. You said of the “models” (constructivist) and “connections” (connectivist) approaches that they “are simply describing two different means of accomplishing the same ends.” You’re right that these are two different means. I don’t think that connectivist approaches are right for every situation. For true novices with very little prior knowledge or skills, more structure and guidance and even direct instruction can be effective. Heck, I think behaviorism even still has a place. Anyone who works with three-year-olds will tell you that positive and negative reinforcement etc. are solid techniques for that age group. With adult learners, especially experienced ones, that kind of control over learning is not as appropriate.

      Although I wasn’t thinking about this when I originally wrote this post, rereading it today I see connections to the current discussions on whether or not you can manage informal learning. I agree with Jane Hart; you can’t manage informal learning. You can support and encourage it, and you can create environments where it can thrive, but you don’t have the amount of control that the word “manage” implies.

      This is an area where the idea of internal or external control is important. “Building” and “scaffolding” to me imply more external control; those are things someone else can do for you or to you. “Growing” seems more internal; no one can “grow” you. In that respect, it’s like the idea of training versus learning: you can train me, but you can’t “learn” me.

  2. @Paul, I hadn’t thought about the connection between the growth metaphor and a PLN/PLE, but that is the type of learning that connectivism explains well. Thanks for helping me extend the metaphor a little more!

  3. I fully agree with the growth metaphor. I’ve been using it on my personal website for years to distinguish elements of the my personal tree growth that add to my personal garden. Roots, branches, leaves, bonsai etc. The metaphor conveniently describes my PLN.

  4. @Nancy, I agree; the metaphors here are very intriguing. My brain is feeling very stretched right now. I wrote another post about the metaphors as an attempt to summarize what has been discussed.

    @Gina, I don’t think I saw your comment as meaning that people consciously think about building learning. Just reading the language in your comment though, I think you do see learning as more structured than fluid and dynamic.

    @Ken, I think your last comment helped me realize I should focus more on the metaphor rather than the literal chemical-physical process. The metaphor will affect my language and my approach to designing learning.

    @Virginia, some great language here extending the two metaphors. I wonder if we need different metaphors for different contexts or types of learning. The metaphors affect our approach, but maybe having multiple metaphors will help us see problems several different ways.

    @Sarah, thanks for dropping in. I’m glad you’re enjoying it. This is very deep conversation; it’s taken me several days and two word clouds to gather my thoughts enough for a second post. I feel like I understand parts of this, but that I’m still missing something. Some connection that I need just isn’t there yet.

    It’s fun to explore anyway, even if a year from now I may look back at these posts and wonder why I was so oblivious!

  5. I’m afraid I have nothing intelligent to add to this conversation other than to say how I am enjoying it, even if I’m not sure I understand it. Thank you all for providing food for thought.

  6. Kia ora Christy! Tēnā korua Nancy and Stephen!

    Connecting the ‘correct’ neurons is a bit like getting the ‘right’ plug-in for things to work properly. Learning can be viewed from different perspectives.

    When it comes to ‘what is learnt’, I’m not so sure that knowing about the neural connections helps, no more than does understanding how the plug-in works.

    The fact is, that both the ‘correct’ neural connection and the ‘right’ plug-in means that things operate properly. I’m not assuming that the neural connections are always exactly the same from person to person for the same learning, no more than I assume that the same plug-in that works fine on one computer will likewise work well on another.

    It is the operation that’s important when it comes to learning. How a learner operates after appropriate learning has occurred is what is assessable, not how they learnt it.

    Ka kite

  7. Hi Gina here! I just want to clear up something. I wasn’t implying that we intentionally construct new meanings. I think learning happens more organically than a learner saying, “My, that is a new thought. Let me query my existing knowledge, so that I can construct a new meaning that makes sense to me”. I think our brains work to make connections for us to provide meaning to new concepts automatically.

    However, I think especially in academic situations when we are presented with a new set of knowledge, we do intentionally search for something familiar that helps us scaffold to the new information.

    Good instructors help us create those mental queries. To me that is creating a bridge from old info to new info.

  8. Christy,

    To add to that metaphor, my first thought when I looked at the two terms of “build” and “grow” was that buildings have a static foundation on which to build. Therefore, there needs to be a solid foundation before new learning can be built upon.

    However, many times we learn something, then relearn it as it makes more sense in a different situation. This fits more of the “growing” metaphor, as the basis from which something grows also is changing (either deeper roots or in the case of a child, the skeleton). There might be growing pains during the process as the foundation tries to offset new ideas that just don’t “fit” into what the basis was. This makes “growing” a lot more dynamic than “building” knowledge.

    On the other hand, like you, I don’t see that connectivism is a replacement for behaviorism or constructivism. I think that different stages of learning, development, environments, and social pressures are going to make one learning theory more relevant than another. I definitely embrace connectivism as a learning theory, but not an overarching one that is appropriate for all circumstances and time periods. If that were the case, then we could go back to the middle ages and void all learning during that time period because of the lack of “connections” made due to the isolationism.

  9. @George, my problem with “learning as advancement of ideas” is that it doesn’t really apply to most everyday learning. When a 3rd grader learns the multiplication tables, it isn’t advancing any ideas for the wider world. In this class maybe that’s what’s happening, and that can be learning, but I don’t think that’s a general enough phrase to cover a lot of basic learning.

    @Ken, you are always welcome to write long comments here. The difference between complicated and complex is a good one here. Learning is both simple and complex, and that’s why I’m having trouble with the deep philosophical discussions. I think there may be value in applying different learning theories to different situations. As you said, constructivism works sometimes but not always. Connectivism might be a great theory to explain something like the MOOC (in fact, maybe the only theory) but perhaps not a 3rd grader memorizing math facts. If I’m working with a toddler, well, behaviorism provides a decent explanation of what will be successful. Behaviorism persists because it does have predictive value in some situations.

    @Stephen, I know you will argue that the toddler is making new connections between neurons, so connectivism does apply there. Ditto for the 3rd grader memorizing math facts. But while that explains the chemical-physical response of the body, I’m not sure how that helps me figure out a good way to teach math facts to a third grader who is struggling. I’m not disputing the accuracy, but I’m not convinced of the usefulness of looking at learning at the level of neurons. I’m not sure how to apply that to what I do. I’m trying to get it, I really am, but maybe I need to understand learning at the level of metaphor first before I can connect that to the concrete level of neurons.

    @Diego, I do think that Gina implied that we have conscious control over how we connect what we learn. Now, I think we can try to make connections, but learning isn’t always as structured as the metaphor of building implies.

    Which brings me to…
    @Virginia, I think your point about the structure is good, especially since it is that structure imposed from an external plan. We don’t nurture buildings, but we can nurture learners. Maybe rather than thinking about scaffolding we should be thinking about fertilizer or some gardening metaphor.

    Interesting ideas everyone. Thanks for all your thoughtful comments. I’m still wrestling with these ideas, but it’s great to think them through with all of you.

  10. As Diego points out, I feel “build” means there is an implied structure to the building. In addition, building usually has some structure imposed through a “plan”, interaction with external forces, and is systematic in how it is developed. I think of “growth” as organic, with little control as to how it is developed. However, for something to grow quicker, there needs to be support and nurturing, so like building, there might be some outside pressures. However, in the case of growing, the outside pressures might aid in the growth, but are not necessary for growth. On the other hand, building requires some intention and cannot be done without a plan, support, or structure.

  11. When I see a phrase like “learning as advancement of ideas” I always have difficulties because of the vagueness inherent in the terms.

    The word ‘idea’, for example, has a 2,500 year old history, and has meant different things to philosophers at different times. When a person writes “learning as advancement of ideas” I have to ask what – physically, chemically, organically – he or she has in mind.

    To me, learning *just is* the creation of new connections between neurons, or the strengthening of existing connections between neurons (this corresponds roughly to long term and short term learning, respectively). These connections are formed as a result of chemical-physical reactions caused by experience; patterns of connectivity develop from experience according to known principles of association.

    Nowhere in this picture do I see anything being ‘built’ or ‘constructed’. Perhaps one day neuroscience (or psycho-kinesis) will have advanced to the point where these connections are ‘built’, but for now, we have to form them the old-fashioned way: by growing them.

    Indeed, when I look at learning from this perspective, I find I have no real point of contact with the older theories from which to offer a criticism. When people talk about ‘constructing ideas’ I literally don’t know what they’re talking about.

  12. Hi Christy!

    When I think about my personal learning process, I think I agree with you more than with Gina. What she implies (I think) is that the learner have a high level of conscious control in selecting how the new knowledge will be tied to the existing one. I don’t think this is the case.

    Even if I’m trying to learn about relativity, it’s likely that, unconsciously, I’ll try to relate (tie) the words involved to other contexts of my life. Unconsciously, I tap into the existing neural networks in my brain, and try to adapt them to recognize a new pattern of information.

    So, to me, this looks more like growing (in unexpected and undirected ways) rather than building. Like Ken says, building brings to my mind bricks disposed in an ordered and structured way (I don’t think that means necessarily linear, though).

    John Medina presents a clear case of how information in stored in the brain (according to brain science) in his book Brain Rules. No defined structure, no progressive storage. Just the creation and recreation of connections between neurons. So, the process seems to be closer to something growing out (or expanding?) than to something being built.

    Just my 0.02 cents. 😀


  13. Tēnā koe Christy!

    You speak of constructivism and building from what’s already known perhaps in a different context. This works well for some learning. In Newtonian Physics for example there are many instances where constructivism work superbly. A case in point the works splendidly is learning the equations for rotational motion, inertia, rotational momentum, rotational energy etc by relating directly to (perhaps) already known translational motion, mass, momentum and translational kinetic energy.

    It’s when the disciplines are quite different that any attempt to draw on previous knowledge in a constructivist way can bring about problems for the learner. For instance, Newton’s laws of motion do not work at extremely high speeds. Einstein proved this, and he introduced what was called his laws of relativity. A contructivist approach to learning relativity (especially Einstein’s special case) does not help the learner.

    In short, constructivist methods don’t always work. In some cases they work up to a point then fall down when the particular is looked at. In other cases, they simply don’t work, right from the start.

    So. When it comes to learning something where there IS no previous knowledge that one can draw on (either as a model, metaphor or other idea) you are right – one has to do ‘work’ to learn.

    My experience as a student at university gave me some insight as to how learning works for me. I am a plodder when it comes to trying to learn. I don’t give up easily. Consequently I discovered all sorts of things on the way that assisted my learning. One of them was simply dogged familiarity.

    When a student comes across something that is new and so different that it cannot be understood, there are usually several resons for this. One is the language associated with the new knowledge. If the language is not familiar, then that alone can bring difficulties to the learner. Familiarity with new words and their meaning helps the learner think in terms of the new piece of knowledge that they are learning.

    This is the next step, which is the ability to think in terms of the new ideas/concepts, brought about through knowledge of the new vocabulary.

    The next is, of course, application of the new-found knowledge – Bloom’s taxonomy comes to mind at this point – and so on and so forth.

    (This might be a long comment 🙂 )

    Learning is not necessarily linear. Building suggests that learning might be linear. I suspect that growing is probably better, for growth, as you know, means more than just one brick on top of another (linear). This is the same difference you might find between complicated systems and complex systems. A machine could be considered a complicated system. We can build on that. A growing child is complex. It’s progress through development is through a complex series of processes we call growth.

    Similarly with a tower that one can build and a growing tree.

    I think that learning is more like the growing tree.

    Virginia Yonkers gave what I think is a superb metaphor for learning in comment she put on one of my posts. Her description of it being like a river or stream made a special impression on me.

    Ka kite

  14. Hi Christy,

    Good question. I wonder, however, if we can eliminate the friction between the two concepts (grow vs. build) by replacing the term with “learning as advancement of ideas”. This is a concept that Scardamalia and Bereiter use (though, they explicitly use the term build). Aside from the word usage, the concept of learning as advancing ideas in a field or discipline seems right to me. Rather than simply absorbing the knowledge that has been created by others, learning is the active participation of individuals in moving the field forward.

    Research, for example, is not concerned with discovery what already is known by most people in a field. It is concerned with advancing knowledge. And, as I look at this rather large course unfold, I’m left with a similar feeling. When the dust settles, we may not agree whether we built or grew knowledge, but I hope we can state that we advanced discussion, we participated, and that we learned.


Leave a Reply