Sept. 10, 2023
In a 2019 paper, researchers led by University of Arizona assistant professor of psychology and cognitive science Robert Wilson used machine learning to determine the ideal difficulty level to learn new things. The neural network they created, meant to mimic the human brain, learned best when it was faced with queries set to 85% difficulty, meaning it got questions right 85% of the time. "If you never make any errors, you're 100% accurate, well, you can't learn from the mistakes," Wilson said.