Definition of Learning
When dealing with training or the concept of the learning organization, it is useful to have a working definition of learning. Learning is like time: we all know what it is until we have to define it. I have spent a lot of time and effort thinking about what learning is - and have often found myself 'tied up in knots'.
So, I went back to basics and considered a very simple example of learning: how animals learn to avoid fire. Very simply, if an animal gets too close to a fire, it gets burnt. The animal remembers the pain and 'learns' that avoiding fire would be to its benefit. Next time it sees a fire it knows that if it goes too close it will get burnt, but if it keeps its distance it will be safe. So, for this specific set of circumstances, the animal has opted to change its behaviour. Of course, it could override its current learning and choose to approach the fire, but it is far more likely to be afraid and run away from the fire. From this simple example, I have defined the following working definition of learning:
"An organism is said to have learnt when it has increased its options for applying, to a specific set of circumstances, new or different behaviour which the organism believes will be to its benefit."We can deduce from this definition that in order to learn, an organism needs the ability to:
- sense what is going on in its environment;
- assess whether its response to an event is beneficial or harmful;
- remember the event, its response and the consequences;
- respond with a different behaviour.
Notice that I have used phrases like: 'increased its options' and 'ability to'. This is because learning is not always immediately followed by an observable behaviour - learning is often stored for future use.
The illustration of the animal and the fire is an example of 'single-loop learning': the learnt behaviour of running away from the fire was not questioned, and the behaviour is likely to be repeated every time a fire is encountered.
At some time during our history, someone questioned the target of putting as much distance between themselves and the fire as possible, and considered the benefits of coming closer to the fire for warmth and cooking. This is an example of 'double-loop' learning where the targets themselves are questioned.