Article Text

Download PDFPDF

The new phrenology: the limits of localizing cognitive processes in the brain
  1. Geraint Rees

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Edited by W R Uttal (Pp 255, £27.50). Published by MIT Press, USA, 2001. ISBN 0 262 210 177

    The notion that particular areas of the brain mediate specific functions is central to neurological diagnosis. In the 19th century, phrenologists made the rather bolder suggestion that particular cortical functions were reflected in the shape of the overlying skull, a claim not thought credible nowadays. However, William Uttal thinks that a modern variant of phrenology is alive and well, in the shape of neurologists, neuropsychologists, and cognitive scientists who think that particular regions of cerebral cortex mediate specific cognitive functions. Uttal rejects the notion that higher cognitive processes (such as memory or attention) can be localised to particular regions (or networks) in the brain. At best he suggests that this is naïve, at worst impossible. He argues trenchantly on two fronts. Firstly, he suggests that all attempts to provide a psychological taxonomy of cognitive processes have failed. If there is no taxonomy, then there can be no localisation. Secondly, he argues that experimental data claiming to localise specific cognitive processes in the brain (for example, from functional neuroimaging) are inconsistent and unreliable. The book thus presents a highly critical appraisal of the founding assumptions of cognitive neuroscience and behavioural neurology. Unfortunately, the data presented represent a very selective overview and interpretation of the field, and should be read with caution rather than enthusiasm. Uttal has a tendency to generalise inappropriately; specific problems with interpretation of a single experiment become general assertions that the technique is unable to provide any useful information. Similarly, generic scientific problems (such as deciding the appropriate level of statistical significance) are presented out of context as specific problems for cognitive neuroscience and functional neuroimaging. It is always important to challenge the assumptions underlying any area of science, so I applaud Uttal for having written this interesting and thought provoking polemic. But it should be read carefully, and the claims made should be taken with many pinches of salt.