Abstract
The present study explored the relationship between dendritic/spine complexity and cortical processing demands. Eight cortical areas were chosen from the left hemispheres of 4 neurologically normal subjects (23, 32, 34, and 69 years), with two blocks representing each level in Benson's (1993, 1994) cortical processing hierarchy: Primary (areas 3,1,2 and 4), Unimodal (areas 22 and 6), Heteromodal (areas 39 and 44), and Supramodal (areas 10 and 11). Based on functional criteria, these regions were subsequently grouped into SIMPLE (Primary and Unimodal) and COMPLEX (Heteromodal and Supramodal) cortices. After staining with a modified rapid Golgi technique, the basilar dendritic systems of 10 supragranular pyramidal cells per tissue block (N=320) were quantified on a Neurolucida system (Microbrightfield, Inc.). Dependent measures were total dendritic length (TDL), mean dendritic length (MDL), dendritic segment count (DSC), dendritic spine number (DSN), and dendritic spine density (DSD).
Despite interindividual variation, neurons in the COMPLEX sample
consistently exhibited higher TDL (11.1%), MDL (3.5%), DSC (8.1%), DSN (35.7%)
and DSD (18.3%) values than those in the SIMPLE grouping, with significant differences
in DSN across all four cortical areas (F (3, 6) = 6.921, p < .02). These
findings suggest that dendritic complexity roughly reflects the computational
demands placed on those systems. Substantial DSD differences were apparent across
all dendritic branches, but especially in distal segments. Age-DSN and Age-DSD
correlations suggested that neurons in COMPLEX areas were more susceptible to
age-related spine decreases than those in SIMPLE areas. (Tissue generously provided
by Dr. D. Bowerman, El Paso County Coroner, and Dr. R. Sherwin, Penrose Hospital)