Monday, 3 August 2015

The myth of "trial and error"

Trial and error, often accepted as a shorthand for "learning from experience, is not enough to be an effective way to learn. At some stage you need trial and success if you are to make progress.

Image from
Learning from failure needs to be tempered with learning from success. If all you do is fail, you do not learn. 

I will give you a personal example.

I am a devotee of puzzle books - I find them to be effective for relaxation, brain-training, and whiling away the 20 minutes of air-travel between take-off, and switching on your electronic devices. I was intrigued by a graphical logic puzzle called Slitherlink and in a fit of bravado I bought a book of "advanced" puzzles. 

I made no progress at all. Every attempt was a failure, and I could not work out any of the tactics or heuristics that would allow me to progress. With no success, I could not build my understanding. Eventually I capitulated, bought a book of "moderate" puzzles, and began to build up a repertoire of tactics and insights. 

The same is true when trying to learn at work. Learning from failure and error can be helpful when the reason for failure is clear, and where you can derive lessons about how to avoid the error next time. When the reason for failure is not clear, and when you have no success, then you cannot learn. Learning from failure alone is very difficult. However as soon as you have a success, then learning can move into overdrive. 

Thomas Edison is often quoted as this point, as saying (when persevering in developing a filament for the electric light) "I have not failed 10,000 times. I have not failed once. I have succeeded in proving that those 10,000 ways will not work".  However it must be remembered that we have a workable electric light bulb not because Edison failed 10,000 times, but because he succeeded once. Without the success, Edison was none the wiser about the material that would function as a lightbulb filament. 

Here's another Edison quote - "I speak without exaggeration when I say that I have constructed 3,000 different theories in connection with the electric light, each one of them reasonable and apparently likely to be true. Yet only in two cases did my experiments prove the truth of my theory".  Certainly Edison operated through rigorous trial and error, but without the trial and two successes he would not have been able to progress his invention.

Trial and error alone is not enough. We truly learn through trial and eventual success. 

No comments:

Blog Archive