Abstract
Regional dendritic variation across primate neocortical areas
has been suggested by several quantitative analyses (Scheibel et al., 1985;
Elston et al., 1996). The present study extends this research by quantifying
the basilar dendritic/spine systems of supragranular pyramidal cells across
eight regions of human cerebral cortex. Tissue blocks from the left hemispheres
of five neurologically normal adults were selected to represent each level of
Benson's (1994) hierarchical functional schema: primary cortex (Brodmann's area
[BA] 4 & 3-1-2), unimodal cortex (BA 22 & 44), heteromodal cortex (BA
6 & 39), and supramodal cortex (BA 10 & 11). Subsequently, primary and
unimodal areas were designated as less integrative regions (LOW); heteromodal
and supramodal areas were designated as higher integrative regions (HIGH). Ten
cells per block (N=400) were prepared by a modified rapid Golgi technique (Scheibel
& Scheibel, 1978) and quantified on a Neurolucida computer/microscope interface
system (Microbrightfield, Inc.). A nested analysis of variance revealed significant
differences across Brodmann's areas and between HIGH and LOW integrative zones
for all dendritic and spine measures. As predicted, dendritic/spine systems
in HIGH integrative regions were more complex than those in LOW integrative
regions. The present findings indicate that the dendritic/spine sytems of cortical
areas involved in the initial stages of information processing are not as complex
as those involved later in the processing stream, and further underscore that
dendritic complexity is roughly associated with the functional demands placed
on cortical regions. (Tissue generously provided by Dr. D. Bowerman, El Paso
County Coroner, and Drs. E. Orsini and W. Tyson of Denver's Children's hospital.)