Dear Christian,
At the risk of stepping in before Alexander has a chance to, here are
a few comments on your questions—which as you've no doubt been
discovering are ones that come up quite often, and that different labs
have very different approaches to answering. ;-)
> In all papers that I read, where activations of the prefrontal cortex in fMRI group studies have been found, the loci of these activations were almost always defined by the respective Brodmann areas they (supposedly) reside in. Anatomical regions like the dorsolateral prefrontal cortex, for example, are consistently described as being located in an area that overlaps BA 9 and BA 46.
>
> If refering to Brodmann areas in order to specify the anatomical locations of activations in fMRI group studies is rather imprecise, why is this practice so common in neuroimaging? (I assume not every lab is using probability maps and the use of Brodmann areas in fMRI research predates the invention of those maps).
I think there are probably a few reasons why Brodmann area labeling
has persisted:
1) It reflects an implicit acknowledgement that underlying
cytoarchitectonic characteristics are probably important, which I
think is intuitive.
2) There are areas that may seem slightly awkward to describe based
purely on macro anatomical landmarks, but are on the surface clearer
when referring to Brodmann areas. For example, if I say something
about the "anterior temporal lobe", that could refer to the superior,
middle, or inferior frontal gyrus (or fusiform, etc…) on the lateral
or medial aspect. If I say "Brodmann Area 38", that narrows it down a
bit in a shorthand way.
3) Brodmann areas appear in the widely-used Talairach & Tournoux
atlas, which makes them seem easy to label and definitive.
Of course, given the challenges that Alexander points out, the result
is that most studies that report Brodmann areas are probably using
them in a rather generic way that does not really reflect anything
about the underlying cytoarchitecture, and if anything may be a bit
misleading. (Including the many problems with the Talairach &
Tournoux atlas.) Thus I would agree wholeheartedly with Alexander's
recommendation to either use some sort of probabilistic Brodmann area
labeling, or stick to macro anatomical landmarks that you can see in
your structural images. (In my "anterior temporal lobe" example
above, it would be rather trivial to better describe a region using
purely macroanatomical features, which would provide specificity in a
straightforward way without appealing to vague Brodmann-area
definitions.)
See also, e.g.:
Brett M, Johnsrude IS, Owen AM (2002) The problem of functional
localization in the human brain. Nat Rev Neurosci 3:243-249.
Devlin JT, Poldrack RA (2007) In praise of tedious anatomy. NeuroImage
37:1033-1041.
Hope this helps! Happy localizing. :)
Best regards,
Jonathan
--
Dr. Jonathan Peelle
Department of Neurology
University of Pennsylvania
3 West Gates
3400 Spruce Street
Philadelphia, PA 19104
USA
http://jonathanpeelle.net/
|