Identifying a preclinical phase of Alzheimer's Disease (PCAD) that is distinct from cognitive changes in healthy aging continues to be a major research focus. Combining neuropsychological and neuroimaging methodologies should improve our ability to differentiate healthy from pathological aging, although studies that utilize both methods often result in equivocal findings, possibly due to variability in cognitive test performance that may be capturing distinct phenotypes. One method of capturing this cognitive variability is to utilize contrasting neuropsychological tests to identify subgroups representative of distinct cognitive phenotypes, and determine whether differences in brain morphometry support these classifications. We review several approaches to defining cognitive subgroups, and we consider the possibility that cognitive asymmetry might provide one means of identifying both functional and structural changes associated with aging and dementia.