I wrote this article in September 2013. Somehow it got lost in a corner of my hard drive until I re-discovered it eleven years later.The ideas still seem to have currency and so I’m published here for the first time.
Stages of learning
After many years of listening to how people debrief activities on training courses Penny Tompkins and I have noticed that in answer to our question “What did you learn?” participants tend to answer in one of four ways – they describe:
- What happened in the activity
- What happened in comparison with what they know or expected
- An insight or new conclusion
How what was learned could be applied elsewhere.
We have also noticed that most answers are type 1 or 2 answers, fewer are type 3, and fewer still type 4. More recently we have linked the lack of answers involving applications to Taleb’s idea of domain dependence (the inability to take an idea or behaviour outside of the originating situation) and Kahneman’s notion of ‘substitution’: if people find a question hard to answer they will, often unconsciously, answer an adjacent easier question instead (Thinking, Fast and Slow).
Examples of the above four responses are:
1. What happened in the activity
X did/said …
I felt …
We got confused.
This happened, then this happened etc.
2. What happened in comparison with what they know or expected
It was difficult.
It’s just like …
It was fun. I liked the exercise.
Surprisingly we all had different metaphors.
They are powerful questions.
The feedback was useful.
3. An insight or new conclusion
I realised how much can be achieved with so few questions.
I had to find a way to park my impulse to offer suggestions.
I now understand what is meant by …
I see what I have been doing wrong.
I’ve got a new tool for my toolbox.
I have an idea what clean is all about.
4. How what was learned could be applied elsewhere
I’ve set myself a goal to …
I could use this in my work to …
I’m wondering how I can adapt this to …
I’m going to start using this on Monday morning.
These four kinds of answer seem to form both a logical sequence and an organisational hierarchy. We suspect that many people will go through these steps one at a time and in this order. And because each one “transcends and includes” (Wilber) the previous kind of information they form a series of nested stages. This therefore is an example of a simple developmental model for learning.
The Kolb learning cycle
A well-know model of learning is the Kolb Learning Cycle:1
Kolb makes the point that Experiencing, Reflecting, Conceptualising and Experimenting are four ways of “grasping new information”, and that by going round the whole cycle a richer, fuller learning takes place.2
Missing from the above cycle is any notion of ‘desired outcome’. Fortunately, learning can and frequently does take place without a desired outcome, and in many situations desired outcomes help to set a direction for learning. Without a desired outcome the learner will need internal criteria and signals for ‘valuable, useful, etc.’ in order to select from his or her innumerable experiences, reflections and conceptions. Even with a desired outcome these signals are useful since some of the most interesting learning happens serendipitously.
With or without a desired outcome there needs to be some way to calibrate whether the process of learning is working or not – and for that evidence is required.
Interestingly, Kolb puts Experiencing at the top of the cycle, presumably because it is a common starting point, but I have always put it at the bottom to signify that the logical hierarchy of concrete experience being ‘lower’ (less inclusive) than abstract conceptualising.
We make a distinction between Conceptualising using existing models to make meaning of Experience (i.e. top-down approaches), and Conceptualising by creating an internally coherent model out of the Experience (i.e. bottom-up modelling).
Comparing Kolb to our stages of learning
If you are familiar with the Kolb learning cycle you may have seen overlaps with our four steps/stages model. (If you did, you will be using Stage 2 of our model.)
Our four stages do not map one-for-one onto Kolb’s cycle. While Experiencing would have happened in the activity that is now being debriefed, both our Stages 1 and 2 are Reflective processes. Stage 3 can be mapped directly onto Kolb’s Conceptualising. As long as Experimenting means “testing implications of concepts” mentally (as well as physically) then it is equivalent to our Stage 4.
Moving round the cycle
Although most attention gets paid to the four ways of processing information (Kolb calls them “learning preferences”), it is also interesting to consider what happens in the arrows? How do the transitions occur?3
A few contextually-clean questions can encourage the stream of learning to flow:
Note *: There are two types of Experimenting (each leading to a different kind of Experiencing):
Physical behaviour in the physical environment. Experiencing is then automatic since, systemically speaking, every action produces a reaction and therefore feedback
Imaginative behaviour (e.g. mental rehearsal, role play, exploring a metaphor landscape, NLP future pacing). The resultant Experiencing is useful feedback for the self-system but since it does not happen in the physical context it is less observable and more personal.
Assuming the learner wants to experiment with imaginative behaviour, useful questions are:
And then what happens?
And what happens just before you [behaviour]?
And when/where will you [behaviour]?
And what needs to happen for you to [behaviour]?
Bateson’s levels of learning
By going round the cycle a few times, a new qualitatively different kind of learning becomes possible: Learning how you learn. An example would be for a person to Reflect on their process of debriefing (i.e. to Reflect on how you Reflect).4
The notion of learning to learn evokes Gregory Bateson’s levels of learning (my simplified version):5
Learning III | is “change in the process of Learning II”. i.e. Learning by modelling self-in-relation-to-the-wider-system and being able to change the process of learning to learn. |
Learning II | is “change in the process of Learning I”. i.e. Learning from self-modelling and being able to change the process of learning. Commonly known as ‘learning to learn’. |
Learning I | is learning by “correction of errors of choice”. i.e. Learning by changing behaviour in response to feedback from others and the world. Commonly just called ‘learning’. |
The first time we go round Kolb’s cycle and our stages we are operating at Bateson learning I. When we have done that a few times we can use how we go round the cycle – our process of learning – as the content for further cycles.
Bateson pointed out that higher, more inclusive levels of learning involve progressively learning how lower-level processes work so there is increased choice in the range of processes available, e.g. a person who is attempting to go from Learning I to II might say “I commonly learn by doing X, Y and Z. Now I am going to use W, X, U to learn.” While this will involve a change of behaviour, more importantly it also involves a change to the process of learning.
Symbolic Modelling
From a Symbolic Modelling perspective, learning is a kind of change. The whole learning cycle can happen once or many times during a single coaching/therapy session. The great value of working with embodied metaphors is that the whole cycle can take place within the confines of a metaphor landscape. And the psychoactivity of that landscape provides the surprising kind of feedback required for the next iteration of Experiencing and learning.
A discussion with Brian Birch on facebook.com/groups/CleanLanguagePrivateDiscussions/ prompted my thoughts about the Kolb learning cycle during a Symbolic Modelling and Clean Language session. I identified:
a. Indicators of where, at any moment in time, the client is on the Kolb cycle
b. Which clean questions invite the client to attend to each of the four kinds of learning.
Thus,
a. As a Symbolic Modelling facilitator we could notice when the client is:
| Concrete Experiencing | Describing the experiences they are having in the here and now, including psychoactive responses and in-the-moment changes. |
| Reflective Observing | Naming symbols in their metaphor landscape, and describing their attributes, location and relationships (across time and space). |
| Abstract Conceptualising | Self-modelling sequences, structures, patterns and drawing conclusions. |
| Active Experimenting | Within their metaphor landscape: exploring the effects of their desired Outcome, trying out potential solutions/Remedies, and interacting in new ways with Problems (see our PRO Model). Asking themself ‘what ifs’. Physicalising their metaphor |
b. The clean questions that, broadly speaking, invite a client to attend to each of the four kinds of Kolb categories are:
| Concrete Experiencing |
|
| Reflective Observing |
Plus within the current perception, questions about relationships across space and time:
|
| Abstract Conceptualising |
Plus across perceptions, a series of time-related questions in combination:
|
| Active Experimenting |
|
It is also worth noting that in Symbolic Modelling, David Grove’s desired outcome question, And what would you like to have happen?, frames the whole change/learning process.
So, what have you learned from this article?
Notes
1. While we will call it the ‘Kolb learning cycle’, Kolb acknowledges that he got it by taking together “Dewey’s philosophical pragmatism, Lewin’s social psychology, and Piaget’s cognitive-developmental genetic epistemology.” DA Kolb et al. (1999) Experiential learning theory: Previous research and new directions in R. J. Sternberg and L. F. Zhang (Eds.), Perspectives on cognitive, learning, and thinking styles. 2000.
2. This is in line with Maturana and Varela’s dictum that “All knowing is doing. And all doing is knowing” (The Tree of Knowledge). Each one of Kolb’s ways of “grasping new information” involves “doing” and therefore “knowing” (and vice versa).
3. More recently the Kolb’s have extended the number of Learning Styles from 4 to 9. They also talks about “learning flexibility” – the ability to either match the learning style to the context, and/or move between learning styles. In this way learning becomes a dynamic adaptive process rather than a fixed state. See: learningfromexperience.com/downloads/research-library/the-kolb-learning-style-inventory-4-0.pdf
5. Bateson did specify Learning 0 and Learning IV, but given that at Learning 0 behaviour “is not subject to correction” and Learning IV “probably does not occur in any adult living organism on this earth”, I have not included them here. For an in depth review see Paul Tosey (2006) Bateson’s Levels Of Learning: a Framework For Transformative Learning? epubs.surrey.ac.uk/1198/1/fulltext.pdf
