Failure

I love sharing a headline with clients that I got from Science Directback in 2019.

“Learning is optimised when we fail 15% of the time”

It’s a robust study – The University of Arizona collaborating with Brown and Princeton amongst others. It’s maths based and tries to quantify objectively just how much failure is needed to hit the sweet spot. That point where something is not so easy we don’t get any stretch or learn anything new, but where we don’t fail so much that we have nothing to build on, or we give up entirely.

Robert Wilson et al conducted a series of experiments where they taught computers different tasks and tested in which situations and with what sort of error rate the computer learnt fastest.

Learning appeared to be fastest when the difficulty of a task was such that the computer responded with 85% accuracy. You may think this is interesting for anyone involved in the AI debate but what’s that got to do with humans at work? Well, when they reviewed previous studies of animal learning, they found the same ratio held true.

The research also evidenced what many people in learning fields know intuitively already – that we learn best through experience and examples, and working iteratively through learning from getting things wrong to get things more right.

They used a simple example of a radiologist who “gets better at figuring out there’s a tumour in an image over time…you need experience and you need examples to get better…”

My neuroscientist partner in crime Dr Iain Price PhD, who I co-wrote DOSE with as a result of a lockdown project on wellbeing, helped me to understand why this 85% rule would be true in terms of developing our brains:

My LinkedIn poll last week showed you agree. 90% of respondents voted failure as being key to learning.

However (as there so often is with something that is easy to turn into a business buzzword or mantra), there is a real problem in applying this cognitive opportunity to your real life.

We don’t like failure – however good it is for our brains and however smarter it could make us.

The title of an article published by the Association for Psychological Science summarises it beautifully – ‘Not Learning from Failure – The Greatest Failure of All’