What has functional neuroimaging told us about the mind? So many examples, so little space.

is it that one distinguishes between psychological theories? The dominant strategy is to test for dissociations. Let us say that theory a (T a) is a proposal about a mechanism that underlies some behavior. Theory b (T b) is an alternative proposal about a mechanism to account for the same behavior. Put differently, according to T a , there is some system of psychological processes that underlies performance in some task. According to T b , performance in this same task is mediated by a different system of processes. Both theories account for the same set of data. Of course, were it the case that T a and T b always made the very same predictions about all behaviors, distinguishing between the theories would be a matter of taste, not a matter of science. T a and T b are interestingly different only if there is at least one circumstance in which they predict different behaviors. It is this circumstance that allows the theories to be distinguished. Let us consider one example raised by Coltheart (this issue) to see what impact neuroimaging data have on distinguishing between alternative theories. Smith and Jonides (1997) reviewed what we knew at the time about the cognitive (not neural) architecture of working memory. That review featured the model of Baddeley, which theorizes that working memory for verbal information is mediated by a different cognitive system than working memory for spatial information. Let us call this T a. The alternative theory not recognized by Coltheart (this issue), but an obvious contrast to T a , is that the architecture of working memory is singular in form, no matter the type of information stored. According to this theory, T b , there is a single system of storage and rehearsal processes that operates on both verbal and spatial information. These two opposing theories can be resolved through behavioral evidence, neuropsychological evidence (lesion data), and neuroimaging evidence, all of which support Baddeley's (1992) theory of separable cognitive systems mediating working memory (T a). Dual-task paradigms have been used successfully to demonstrate the separability of spatial and verbal aspects of working memory. For example, den Heyer and Barrett (1971) conducted a dual-task experiment in which subjects had to perform a primary task of remembering either the identity or the spatial location of letters on a grid. During the retention-interval, subjects were given a secondary task that was either …