A very large study finds a statistically significant difference between two brief attention exercises, but the estimated difference is tiny. Which interpretation is most defensible?