This book is predicated on the relatively uncontentious notions that discourse patterns – what people do when they talk or write – can provide trained observers with information about cognitive functions and affective states in speakers and, further, that cognitive functions and affective states may be signs of integrity of neurological function and structure. Neurolinguists, psycholinguists, aphasiologists, psychiatrists, psychotherapists and speech pathologists all take some variation on assumptions like this as their point of departure in studying brain–behaviour relationships and treating some neurological and affective disorders. However, discourse – people's talk and text – is inherently complex and apparently unstable and, worse, the neurological substrate and processes that support even superficially simple things like ‘how words are represented in the brain’, let alone ‘what happens in brains when people talk’ are matters of active debate and investigation rather than scientific givens. In the face of so much uncertainty and complexity, most of the work done on language–brain relationships has, very sensibly, centred on theoretically discrete and/or methodologically isolatable phenomena associated with particular semantic, morphosyntactic or phonological structures or processes. Work on discourse in clinical environments as another means of investigating neurocognitive (dys-)function, although often called for, has been less common.

This situation is changing now because of technological developments and, we think, a sea-change-like shift that is taking place in attitudes to brain–behaviour relationships.