estimates from single (cross-sectional) measurements, and greater than the heritability of individual externalizing disorders examined in the same sample (for seven disorders, median h2 =.63, Iacono et al., 2003). Because all measurements contribute to estimates of the intercept, this study indicates the value of repeated assessments of an endophenotype due to the increased reliability that derives from using multiple measurements. Individual differences in rate of change (linear slope) were also heritable, although to a much lesser degree (h2 = .15). Nevertheless, the degree of overlap in genes influencing intercept and slope was quite striking, and the best-fitting model in this study indicated that the two sources of genetic influence could be considered perfectly coextensive. That is, the genes that influence intercept, or P300 amplitude at age 17, are the same as those that influence change in amplitude into early adulthood, indicating that developmental trajectories in P300 amplitude index genetic risk.
