Participants were biological offspring of centenarians enrolled in the New England Centenarian Study, an ongoing nationwide study of centenarians and their families. From 560 sibships, the oldest living sibling was chosen, yielding 352 female and 208 male centenarian offspring. A power analysis revealed that a total sample of 200 subjects was needed to detect a difference of 2.5 points or more in mean scores between the centenarian offspring and reported normative mean scores with an α = 0.05 and β = 0.80. Thus, anticipating some losses to follow-up or refusals to participate, a random sample of 308 unrelated subjects was selected (168 women and 140 men, reflecting the expected proportions of women and men for an average age of 75). Of the women, 24 were unavailable, 14 had died, and five refused, resulting in an analytical sample of 125 women. Of the males, 15 had died, and four others could not be reached, resulting in an analytical sample of 121 men.
