Is learning a discrete or emergent phenomena?

Is Learning Discrete Or Emergent

The traditional view of learning is that it is the transmission of truth from one human mind to another (see here for my take on this subject).

In modern terms it is all most like computer file transfer protocols exchanging data from one knowledge retention source to another. Therefore if there is an improper transmission of the data, the fault must lie in the exchange protocol.

The scientific method fits well with this view of learning as a language exchange protocol as researchers can go on to break down the act of learning into smaller and smaller discrete parts of study. This scientific reductionism is both incredibly powerful for helping us understand how things work in the world, and for helping to find general theories that can predict behavior into the future.

However, it is not so good at understanding emergent phenomena.

To illustrate, let’s take a look at the Game of Life developed by John Conway.

It was originally developed as a mathematical model for the growth and death of cells. It is played out on a square grid with cells coming into existence and dying, and is controlled by three very simple discrete rules.

  1. Birth – if an empty cell is surrounded by exactly 3 populated cells, a new cell is formed and is placed in that position.
  2. Death by overcrowding – if a cell is surrounded by 4 other cells, it will die (the cell marker is removed and the area becomes empty).
  3. Death by loneliness – if a cell is surrounded by fewer then 2 other cells, it will die (the cell marker is removed and the area becomes empty).

At this point, we have explained the underling rules of the system scientifically. We have broken down the model to its most simplistic, and understand the sum of all the processes.

Something strange starts to happen, however, when we activate the model from certain start conditions.

The cells begin to move across the board in a steady pattern. These are known as glider patterns (one of these glider patterns is illustrated bellow), and their occurrence is not readily predictable from the rules governing the system.

This emergent phenomena of walking was the first one to be noticed, but is by no means the last. Certain start conditions have revealed “puffer trains” and “glider guns” which produce whole new cell colonies.

These phenomena are the product of all the interactions within the system as a whole and as such do not lend themselves to a reductionist explanation If you would like to play around with the game of life there is a simulator for it here.

Game_of_life_animated_LWSS

Emergent “properties are a property of the whole, not the property of the parts, and cannot be deduced from the properties of the parts.” ((Jamshid Gharajedaghi, Systems Thinking 2nd ed., page 45)) This makes them especially problematic to observe. Measuring them with traditional analytical tools only gives us a snapshot of one of their states. In effect we are forced to measure the manifestations of the system, and not the system itself.

The distinction between the two types is possibly best understood by the following diagram, also taken from Systems Thinking, 2nd ed., page 45.

emergent property

What other systems out there have been linked to emergent or type II phenomena? The classics are things like love, life, happiness, etc. But, is learning a type I/discrete system, or a type II emergent system? And if so, what are the implications for how we study learning?

I believe that learning is an emergent phenomena. Viewing it as such will help us to reconcile some of the apparent contradictions between different schools of thought in the education debate.

I also believe that looking at it from this holistic perspective may provide new insights and move the debates forward in new directions. To that end I present four conclusions that can be drawn if learning is truly a type II emergent phenomena.

  1. Initial start conditions are incredibly important.
  2. The interplay of even the smallest element can have a sometimes disproportionate impact on learning.
  3. We can only measure the manifestations of learning not learning itself (and that these manifestations can be faked).
  4. Learning is the outcome of an ongoing process, and if the system supporting that process comes to an end, the outcome will also end.

(for a more poetic reflection on learning try here)

Related Posts

Share This

Share on facebook
Facebook
Share on twitter
Twitter
Share on linkedin
LinkedIn
Share on email
Email
Share on print
Print

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.