Korva is a scroll of the internet’s best ideas.
Fail first, then learn it
Seeing something many times doesn't mean you'll remember it well. Everyone has seen the Apple logo countless times, yet when people are quizzed on whether the bite is on the left or the right, or whether there is a stem or a leaf, they aren't sure. The classic version of this demonstration used a penny.
Drawing the logo while looking at it helps. Drawing it without looking helps more. You struggle, you start to question the bite and the leaf, and then you look at the logo again. That second look produces better learning than skipping the struggle, even if the drawing came out wrong.
A retrieval failure followed by corrective feedback is how people learn through life: they make a mistake, then correct it. Someone who always does it the right way never notices the features.
Why does failing first help learning?
Psychologist Robert Bjork calls this kind of effort a desirable difficulty: a harder way of studying that feels worse in the moment but sticks better. Trying to pull an answer out of memory strengthens the paths that lead to it, while rereading only builds familiarity. A wrong guess also makes the correct answer stand out when it arrives. Bjork and his wife Elizabeth have studied this effect for decades at UCLA.
Why can't people draw a penny they have seen thousands of times?
In a 1979 study, Raymond Nickerson and Marilyn Adams asked people to draw a US penny from memory, and most got major features wrong or in the wrong place. People could still tell a real penny from a fake one, because recognizing an object takes far less detail than reproducing it. Everyday objects only need to be distinguished from other things, so the brain keeps just the features that do that job. The same gap between recognizing and recalling is what makes multiple-choice tests easier than essays.
Why does correcting a wrong answer stick so well?
Studies of the hypercorrection effect, such as work by Brady Butterfield and Janet Metcalfe in 2001, found that people correct their confident mistakes more reliably than their uncertain ones. The surprise of learning that a firmly held answer was wrong seems to draw extra attention to the correction. That attention helps the right answer replace the wrong one. In their experiments, errors made with high confidence were more likely to be fixed on a later test.
More from Andrew Huberman
