Neuroimaging as a Tool for Functionally Decomposing Cognitive Processes

Both those who extol and those who castigate neuroimaging studies and their invocation in cognitive science often misconstrue the contribution neuroimaging is seeking to make, and is capable of making, to cognitive science. We do believe that advances in neuroimaging, including functional magnetic resonance imaging (fMRI), as well as such techniques as single cell recording, are important contributions to the experimental repertoire of cognitive science. We also anticipate that neuroimaging’s importance will increase with improvements in imaging technologies and techniques. Our objective here, however, is neither to extol nor to castigate neuroimaging, but to make clear what sort of contribution neuroimaging, when done well, can make to understanding and explaining mental phenomena. Many of those who adopt extreme views of neuroimaging are not themselves practitioners of the technique and fail to appreciate principles that are commonplace or platitudes among the expert practitioners. Of course, some who adopt extreme views are practitioners. Accordingly, though we are by no means expert practitioners, we start with some commonplaces that are not always transparent in reports of neuroimaging, but are generally understood by the researchers conducting the research (see, for example, Logothetis 2008) and need to be taken into account by those evaluating it: