The field of therapeutics for Alzheimer disease (AD) is moving toward early intervention with the presumption that intervening before significant CNS damage will be beneficial. According to this rationale, influencing the early pathologic events should prevent or slow cognitive decline. Toward that end, Farrell et al.1 in this issue of Neurology® have assessed cognitive changes in the early stages of the putative pathologic cascade occurring in AD. The authors studied 112 clinically normal participants in the Harvard Aging Brain Study with cognitive, amyloid PET imaging (Pittsburgh compound B [PiB]-PET), and tau PET (18F-flortaucipir-PET) measures. The authors categorized the participants according to Centiloid (CL) values in those who were PiB negative (<CL20) and those who were mildly PiB positive (20 < CL < 40) and correlated these amyloid levels with clinical measures. They hypothesized that certain cognitive measures are more sensitive to AD pathology at various stages of the development of the pathologic process, depending on the distribution of amyloid and tau deposition. They documented measures of processing speed over most participants in the early amyloid stages (<CL20) and found memory changes in the early amyloid and early tau stages (20 < CL < 40). They concluded that processing speed measures might be most sensitive during the early β-amyloid stages and memory measures more reflective of emerging tau and β-amyloid accumulation.