Introduction

For the most part, the study of speech and language function can be based only on human data. The study of birdsong offers interesting parallels with a human infant's acquisition of language (Doupe & Kuhl, 1999). Further, the use of anatomical tracer studies (Romanski et al., 1999), single cell recordings (Tian et al., 2001), and metabolic brain studies (Poremba et al., 2004) in the monkey is directing attention to auditory processing streams (Kaas & Hackett, 2000; Rauschecker & Tian, 2000). As a result, inferences are being made from nonhuman primate research that can be profitably applied to research into human speech perception and comprehension (Scott & Johnsrude, 2003) and production (Wise et al., 2001). Nevertheless, because human communication is so complex and versatile, there are strict limits to what zebra finches and macaques can teach us about everyday human discourse.

The introduction of functional neuroimaging has given us the potential to view the functional anatomy of the normal human brain as it perceives, comprehends, and produces language. However, we had better state at the outset what this chapter will not do. It will not provide a detailed overview of all functional imaging studies of human communication. This is for two reasons. First, since functional imaging is used predominantly as an anatomical tool, attributing a particular brain function to a particular cortical or subcortical region, a conscientious “overview” may end up reading a little like a modern-day phrenology text, of the kind “here lies phoneme perception and there lies theory of mind.” Second, the literature is now vast and encompasses papers of real excellence as well as ones that are less memorable. We will confine our review to results that we consider have raised issues of interest in the study of the functional anatomy of language, issues that, in many instances, are still the subject of lively debate.