How much can we learn about the mind by using brain imaging techniques like fMRI? In a recent NYT article, David Brooks has argued that we can learn very little. He notes how multiple brain regions are responsible for a single psychological process like working memory, and a single brain region like the insula is involved in multiple mental states. Tim suggests that systems neuroscience is well aware of the first problem, and that we might just abandon some mental states as natural kinds to solve the second. There are other real problems for fMRI, but only if it is not used in conjunction with other methods.