Image, language, brain

Image, language, brain. Ed. by Alec Marantz, Yasushi Miyashita, and Wayne O’Neil. Cambridge, MA: MIT Press, 2000. Pp. 272. ISBN 0262133717. $60 (Hb).

Reviewed by Jyrki Tuomainen, University College London

Linguistic description of language behavior and accounts of brain functions underlying language processing are currently not compatible with each other. The background of this book edited by Alec Marantz and his colleagues is the Mind Articulation Project, which strives to bridge the gap between linguists and cognitive neuroscientists. To reach this goal, it tries to implement the successful approach of visual neuroscience to the study of language and the brain. The book, based on presentations given in Tokyo in 1998 at the first symposium of the Mind Articulation Project, is an excellent introduction for linguists and other professionals who are interested in brain imaging of experimental research on language and other cognitive functions.

Image, language, brain consists of two parts, ‘Language and the brain’ (seven chapters) and ‘Image and the brain’ (five chapters), and covers a wide variety of topics on cognition and brain functions. The twelve chapters and the introduction are written by forefront linguists, cognitive scientists, psychologists, and cognitive neuroscientists who bring on their expertise with a clarity one can only admire. Most of the chapters are accessible to readers without much of previous conceptual machinery required by cognitive neuroscience, or linguistics for that matter.

The introduction by the editors appropriately sets the scene for the book and the Mind Articulation Project. It provides the reader with many of the basic concepts and background information needed to tackle the difficulties in correlating mind and brain. Another excellent feature is that it presents testable predictions about how linguistic operations could be carried out in the brain. Furthermore, the introduction cautions against overinterpreting the results of brain-imaging studies, and one is reminded, rightfully, that mapping of brain activity to a function is most useful when much is already known about the brain area that is presumably activated by a specific function. The other side of the coin is that brain imaging of cognitive functions in its best form is a business in which solid functional models of cognitive function guide the interpretation of brain activation patterns. Thus, brain imaging of cognitive functions should not be just a hunt for ‘blobs’ (i.e. activation spots in the brain images) for which a function will be assigned using hindsight and without proper task analysis of behavior.

In line with the tone of the introduction, in Ch. 1, Noam Chomsky suggests that despite extensive progress and optimism in many areas that investigate brain, behavior, and cognitive faculties, we are still a long way from the solution of the ‘mind-body’ problem, and there is still a lot to do with regard to understanding the brain correlates of much simpler cognitive processes than speech and language.

In Ch. 2, David Poeppel and Alec Marantz present a valuable review of different neuroimaging techniques that are used in investigating speech processing (which in this context denotes speech perception). This is a valuable chapter in pointing out the difficulties in trying to tease out different processing levels in speech perception. One can conclude that we are far from understanding the different transformations the speech signal undergoes, and are even further from neuroscientific evidence of a single abstract phonological component that would subserve both production and perception of speech.

In Ch. 3, Jacques Mehler, Anne Christophe, and Franck Ramus give an account of the initial and often highly remarkable abilities of the infant to recognize complex auditory phenomena such as the rhythm of the mother tongue, voice of the mother, sensitivity to auditory contrasts that underlie phonetic categories, and so on. They present a version of their phonology class hypothesis (refined in their later publications), one corollary of which is that infants are sensitive to the rhythmic aspects of language. The reason for this might be that the fetus has access to the prosodic features of speech, which are more tolerant to the low-pass filtering properties of the womb and abdominal wall than, for example, the acoustic features that correlate with phonetic categories.

Ch. 4, by Willem Levelt and Peter Indefrey, begins with a review of the recent version of a model of lexical access in speech production by Levelt and his coworkers. The authors then summarize the meta-analysis of brain-imaging studies of speech production by Indefrey and Levelt (2000; further refined in Indefrey & Levelt 2004). This is an important meta-analysis in the realm of psycholinguistics and brain imaging and should be read by all who are interested in the psycholinguistics of speech production.

Echoing one of the general themes of the book, in Ch. 5, Edward Gibson’s research testifies for the primacy of behavioral models and suggests that most of the progress in understanding the processing of syntactic structures owes to behavioral studies. The latest version of his dependency locality theory is presented in some detail, focusing on two resource-consuming aspects of syntactic parsing, namely, performing structural integrations (i.e. connecting words into the structure for the input so far) and storage cost (i.e. usage of working memory in keeping track of incomplete dependencies).

Ch. 6, by Angela Friederici, introduces a tentative model of the neuronal dynamics of auditory language comprehension. One interesting and commendable aspect of this model is that it includes prosody as an important processing component that makes a lot of information about syntactic and phonological structures available for the language processor. Ch. 7, by Masao Ito, provides an account of control functions of cognition and language.

Part 2 begins with a chapter entitled ‘Imaging neuroscience: System-level studies of the physiology and anatomy of human cognition’, by Richard Frakowiak. It illustrates the systemic approach to brain functions and brain imaging, which emphasizes the functional connectivity of different brain areas underlying specific cognitive processes. Ch. 9, by Hiroshi Shibasaki, describes the central control mechanisms of voluntary movement studied by multidisciplinary noninvasive approaches. Ch. 10, by Kensuki Sekihara, David Poeppel, Alec Marantz, and Yasushi Miyashita, focuses on the inverse problem in neuroimaging and describes how an ‘eigenstructure-based’ approach can be applied to extracting localized cortical activity from magneto-encephalographic (MEG) data. This chapter is valuable to expert readers, but it is an oddball in this book because it requires a mathematical background that one cannot expect from the audience the book is intended for. Ch. 11, by John Reynolds and Robert Desimone, focuses on competitive mechanisms subserving selective visual attention, and finally, Ch. 12, by Yasushi Miyashita, deals with the neurophysiological basis of visual imagery.

Overall, this book is an elegant example of research on the mind-body problem; it provides an account of an enterprise harnessed with enthusiasm created by current advancement in the methodology in brain imaging, yet accompanied with words of warning not to use advanced brain-imaging technology just to ‘see what’s happening in the brain’ without theory and prediction of what the outcome might be. One can never emphasize too often that true progress in (cognitive neuro)science lies in clever research ideas directly related to models and theories about the human mind.