The authors developed multivariate models to examine the association between Clinical Dementia Rating (CDR) scale scores and the spatial distribution of paired helical filament tau (PHF-tau) pathology. Severity of tauopathy was examined in 14 cortical regions of interest (ROIs). Classification trees were used to test the independent hierarchical contributions of each ROI to dementia. Multiple-regression and cluster analyses were performed to determine the relative contributions of select ROIs to dementia. Dementia (CDR > or = 1) was modeled as a dichotomous variable. Autopsy material was obtained from 124 demented and nondemented elderly patients. All ROIs except the hippocampus made significant contributions to dementia. However, they were not independent. In multivariate models, only a single step (Step 7 in a hierarchical progression), which contained four ROIs, contributed significantly to dementia. A classification tree resulted in a single decision split, suggesting that only Step 7 ROIs need be considered. A total of 89.4% of the cases were correctly classified (p < .0001). Twelve discrepant cases all had superimposed vasculopathy that might also have affected the function of Step 7 ROIs. The transition to clinical dementia was associated with the presence of tauopathy in A9/10, A22, A23, and A39. Animal studies suggest that these represent a single distributed cortical network focusing on prefrontal regions that provide "executive control" over complex goal-directed behaviors. A22, A23, and A39 provide major afferents to other frontal systems and have previously been implicated in very early clinical Alzheimer's disease.