Abstract: A Biometric Latent Curve Analysis of Visual Memory Development Using Data from the Colorado Adoption Project

A Biometric Latent Curve Analysis of Visual Memory Development Using Data from the Colorado Adoption Project Yan Zhou and John J. McArdle University of Southern California This study examined genetic influences in the development of visual recognition memory using longitudinal and sibling data from the Colorado Adoption Project (CAP). Scores from the Names and Faces Association Test, including an immediate measure and a delayed measure of associative memory, have been obtained for 247 adoptive and 246 biological sibling pairs who have been repeatedly tested up to 6 times from age 9 to 21 years, among which 3 testing sessions were conducted via telephone. The analyses consisted of three steps: (1) longitudinal measurement invariance was examined; (2) the curve-of-factors model was applied assuming strong factorial invariance, and the effects of telephone tests were modeled with an additional latent variable; and (3) the variances of the latent intercept and the latent slope were decomposed within a behavior genetic model (McArdle, 1986). The results showed that associative memory continuously increased into early adult, and the scores were higher in the telephone testing sessions than in the faceto-face interviews. Implied variance components from the model estimates were plotted in Figure 1. The variance that was accounted for by the intercept and slope was largely genetic, and the genetic effects increased with age. Influence from the shared-family environment was found to be negligible. In contrast, two other memory measures in the CAP data—an immediate score and a delayed score from the Picture Memory Test, which were found to represent a functionally distinct construct, showed no genetic influence in the developmental course. McArdle, J. J. (1986). Latent variable growth within behavior genetic model. Behavior Genetics, 16(1), 163–200. The authors would like to thank Dr. Sally Wadsworth (University of Colorado) and Dr. Bruce Pennington (University of Denver) for sharing the CAP data. Correspondence concerning this abstract should be addressed to Yan Zhou, John L. Horn Psychometric Laboratory, Department of Psychology, SGM 501, 3620 S. McClintock Ave., University of Southern California, Los Angeles, CA 90089. E-mail: zhou5@usc.edu