The overarching aim of our current funding (NIH Grant # R01 AG063949-01) is to repurpose the Louisville Twin Study?s original goal of understanding the etiology of child cognitive development into a true lifespan developmental twin study of cognition, including delineation of the etiology of preclinical symptoms of Alzheimer?s disease and related dementias. In the current project period, the third independent aim encompasses modeling effects of cognitive development on cognitive functioning in midlife, adjusting for confounding genetic and environmental factors. Throughout the life of the Louisville Twin Study (LTS), however, recorded ability scores have been age standardized, thus making it impossible to estimate effects of true intraindividual change in cognitive ability over any period of development. Problems with the age- standardized approach are compounded by the administration of different tests (e.g., moving from the Wechsler Intelligence Scale for Children (WISC) to the Wechsler Adult Intelligence Scale (WAIS)) and test editions (e.g., updating from the WAIS-III to the WAIS-IV) that supposedly measure the same latent cognitive abilities. Whether they actually do so is an empirical question that has not been addressed previously in any longitudinal twin study of cognitive ability. All LTS twins? cognitive ability protocols are available, so individual items could be used in future analyses. We propose using vertical linking methods that allow the administration of validated cognitive tasks without requiring age-standardizing cognitive test scores. Child and adult test content typically is not the same, but there are some items that were administered in childhood that can be retained in adult tests. For the current data collection, three age-appropriate items from the WISC Block Design, Similarities, Vocabulary, Arithmetic, and Information subtests will be administered to all returning LTS adults (projected N = 750). As children, LTS twins completed approximately 10 cognitive ability tests, on average, half of which were WISC tests. With an estimated five WISC protocols administered to each twin, on average, there are greater than 73,000 individual subtests to archive and vertically link child and adult Wechsler ability scores. Item Response Theory models subsequently can be used to place ability scores on the same scale so that ability trends can be studied and compared across different periods of child and adult development. Twins? latent cognitive ability test scores, thus, will be put on the same scale from age 7 through midlife. Our efforts in the LTS are the first of its kind in longitudinal twin studies of cognitive ability. By linking childhood cognitive tests to adult tests, we will increase the significance and innovation of our findings by specifying the lifespan growth functions of ability at the subtest and latent factor levels to model individual twins? true intraindividual cognitive ability trajectories.
One of the aims of NIH Grant # R01AG063949 is to test the genetic and environmental mechanisms through which child cognitive development influences cognitive functioning at midlife. In the 62-year history of the Louisville Twin Study, only age-standardized intelligence scores have been studied despite having access to all twins? raw item-level cognitive data at each assessment. Age-standardized ability scores systematically remove age-related intraindividual variability that is of greatest interest to answering questions as to how cognitive development contributes to cognitive aging. This administrative supplement will support archiving item-level cognitive data so that individual twins? true intraindividual cognitive ability trajectories can be developed from childhood through middle age.
