From hodology to function.

In 1993 Francis Crick and Edward Jones appealed to the scientific community with an editorial in Nature entitled ‘Backwardness of human neuroanatomy’. Their aim was twofold: to ‘make a wide audience aware’ of how little is known about human brain anatomy, and to highlight the urgent need for ‘new methods to solve this problem’ (Crick and Jones, 1993). Around that time a group of researchers based at the NIH were working on a new magnetic resonance technique called diffusion tensor imaging (DTI) (Basser et al ., 1994). Almost 15 years after publication of the Crick and Jones editorial, diffusion-based imaging methods represent perhaps the most concrete response to their appeal. Without doubt, diffusion-based imaging methods are revolutionizing the field of neuroimaging along two fronts: for the first time we are able to reconstruct white matter pathways in the living human brain and re-explore connectional anatomy after a long period of stagnation since the work of the early 19th century neuroanatomists; and diffusion-based imaging methods allow us to take a hodological (i.e. connectionist) approach to disorders of the brain, where a more narrow localizationist model has been found wanting (Catani and ffytche, 2005). In this issue of Brain , Schmahmann et al . (2007) report the results of a study where a recently developed diffusion-based imaging technique, namely diffusion spectrum imaging, is used to perform virtual dissections of a monkey's white matter tracts. Their article contains spectacular images of the major association pathways and their virtual reconstructions appear to replicate the details of the monkey's anatomy derived from tracing techniques. The reader might be surprised that Brain has overruled its editorial policy of rarely publishing normative data. However, the reason for this is that the article encourages the reader to reflect on two important issues in neurological science: the …