We ace making great progress in singling out those among us who are at
high risk of cancer, whether on the basis of epidemiologic characteri
stics that predict risk in the absence of a firm knowledge of mechanis
m or, more recently, by specifying the genetic site of a factor that u
nequivocally puts a specific person in danger. We have been less capab
le of identifying factors that, given awareness of increased risk, det
ermine whether or not a person adopts an appropriately self-protective
behavior. This article reports the experience of female twins whose c
otwins have a diagnosis of breast cancer. Such individuals can be assu
med to have been presented with incontrovertible evidence of their ele
vated risk almost as much as if they had been found to have a dangerou
s gene. Their subsequent actions can shed light on why persons do not
always take steps to protect themselves from a clear but future danger
. It is evident that cognition is not the sole, or even the most impor
tant, determinant of health promoting behavior. Among the others are b
eliefs about cancer causality and the state of personal health, the de
tails of the medical experience of relatives, and the degree to which
perceptions of that experience have intruded into consciousness. The g
ap between rational and actual efforts to reduce personal risk is grea
t, and if technology is to offer us the benefits promised, we must giv
e priority to narrowing that gap.