ABSTRACT
The existence of modular processes in human cognition is now widely accepted as one of the ``givens'' of the ®eld. The notion that all human cognition is undertaken by a unitary system ± from the most basic perceptual processes to thought and language ± is virtually impossible to sustain. On the other hand, it is also very hard to accept that all cognitive processes are performed by independent modules, if only because some mechanism would be needed to coordinate them, and this process would require some degree of domain independence. Both Baddeley (1986) and Newell (1990) have recognised this, the ®rst with the concept of the central executive, and the second with the quest for uni®ed theories of cognition. Given that we recognise the existence of both modular and domain general processes, we need objective means for determining between them. We are left then with the following problems.
