If neuroimaging is the answer, what is the question?

It is unclear that we will come to a better understanding of mental process es simply by observing which neural loci are activated while subjects perfo rm a task. Rather; I suggest here that it is better to come armed with a qu estion that directs one to design tasks in ways that take advantage of the strengths of neuroimaging techniques (particularly positron emission tomogr aphy and functional magnetic resonance imaging). Here I develop a taxonomy of types of questions that can be easily addressed by such techniques. The first class of questions focuses on how information processing is implement ed in the brain; these questions can be posed at a very coarse scale, focus ing on the entire system that confers a particular ability, or at increasin gly more specific scales, ultimately focusing on individual structures or p rocesses. The second class of questions focuses on specifying when particul ar processes and structures are invoked; these questions focus on how one c an use patterns of activation to infer that specific processes and structur es were invoked, and on how processing changes in different circumstances. The use of neuroimaging to address these questions is illustrated with resu lts from experiments on visual cognition, and caveats regarding the logic o f inference in each case are noted. Finally, the necessary interplay betwee n neuroimaging and behavioural studies is stressed.