Failing This Many Times Could Help Us Learn New Things More Efficiently

Scientists have calculated what they say is the percentage of times we need to fail in order to most efficiently learn something new.

In what researchers have dubbed the Eighty Five Percent Rule, failing 15 percent of the time and succeeding the remainder is the optimum way to gain new skills and information. Otherwise if the challenge is too easy we don't learn. Too hard, and we're likely to be put off and give up.

To arrive at this percentage, the authors of the paper published in the journal Nature Communications taught computers simple tasks, like telling the difference between patterns or reading and sorting handwritten numbers.

They found the machines learned fastest when they got the task wrong 15 percent of the time, and succeeded 85 percent.

And by looking at existing research on learning in humans and animals, like monkeys, they found this ratio also appeared to give the best results when it comes to learning.

Robert Wilson, lead author of the study and University of Arizona assistant professor of psychology and cognitive science explained in a statement that the work puts an idea known as the "zone of proximal difficulty," already in use in the field of education, on a mathematical footing.

Wilson explained this approach could be used in what is known as perceptual learning, where we hone our knowledge over time. He used the example of a radiologist learning to tell between an x-ray which shows a tumour and one which doesn't.

"You get better at figuring out there's a tumor in an image over time, and you need experience and you need examples to get better," he said.

"I can imagine giving easy examples and giving difficult examples and giving intermediate examples," Wilson said.

"If I give really easy examples, you get 100 percent right all the time and there's nothing left to learn. If I give really hard examples, you'll be 50 percent correct and still not learning anything new, whereas if I give you something in between, you can be at this sweet spot where you are getting the most information from each particular example."

Relating the findings to students he gave the following advice. "If you are taking classes that are too easy and acing them all the time, then you probably aren't getting as much out of a class as someone who's struggling but managing to keep up."

Wilson said: "The hope is we can expand this work and start to talk about more complicated forms of learning."

Speaking to Newsweek, Wilson stressed: "Definitely do not take away that 85 percent is some magic number we should be aiming at all the time. That only applies in the very limited settings we considered. A more useful takeaway is that perfection isn't great for learning—we need to make some mistakes in order to learn and if what you're doing isn't challenging, then you're probably not learning as well as you could."

"This is especially true for kids—and as parents and educators we need to make sure we're not overemphasizing perfection at the cost of learning. Of course, when things are too difficult we don't learn either, so it's equally important not to push your kids into things they aren't ready for!" Wilson said.

Pointing out the limitations of the study, Wilson told Newsweek the team only considered what are known as binary classification tasks, for instance identifying whether a photo features a cat or a dog, rather than more general challenges. The study also didn't look at very fast learning in humans and or animals.

This article has been updated with comment from Robert Wilson.