Why brain imaging doesn’t always hold up — and how experts say to fix it

The use of external validation in neuroimaging studies promotes robust research and reduces mistakes, scientists argue.