Kimberly Saudino, Ph.D. - US grants

Parent ratings are the most widely-used measures of temperament in the literature. Many conclusions regarding child behavior and developmental outcomes (e.g., behavior problems, sibling relationships) have been based on studies utilizing such measures. However, research using twins suggests that parent ratings may exaggerate the differences between sibling temperaments. Most temperament research examines only one child per family, therefore, such effects will go undetected. The proposed project is a pilot study exploring parental perceptions of temperament in non-twin siblings. The research focuses on two questions: 1) Do parents exaggerate the temperamental differences between their siblings? 2) If so, are the previously reported relations between sibling differences in temperament and child outcome a result of parental expectations rather than actual child behavior? The sample will include 100 families with two children between the ages of 3 and 8 years. Parent ratings of temperament will be obtained for each sibling. For one-half of the families, parents will rate the temperaments of either children contemporaneously. The remaining families will have a one- month interval between the ratings of temperament assessed for a 48-hour period using mechanical motion recorders. Shyness will be assessed via behavioral observations within the laboratory. Parents' ratings of maternal differential treatment (affection and control) and positivity within the sibling relationship will be obtained. Sibling resemblance, as indexed by intraclass correlations, for parent-rated temperament will be examined. Parents perceptions of sibling differences for activity and shyness will be validated against objective measures. The association between sibling differences in temperament and outcome measures (quality of sibling relationship and differential maternal treatment) will be explored using both parent-report and objective measures of activity and shyness. This pilot study is a first step in understanding parents' perceptions of temperament in non-twin siblings. The results of this research could have significant implications for the study of temperament, not only methodologically, but substantively. For example, it is found that parents exaggerate behavioral differences between siblings, it would be important to understand how and why these perceptions are formed, whether they are enduring throughout development, and the consequences that these parental perceptions may have on children's behaviors and self perceptions.

DESCRIPTION (provided by applicant): Activity level (AL), a core dimension of nearly every temperament theory, is a highly salient feature of child behavior that has been linked to various facets of child health and development. Although there is considerable evidence of genetic influences on individual differences in AL, there are many unanswered questions about this important temperament dimension. First, little is known about the role that genetic factors play in he development of AL. Second, short-term, situational changes in AL have rarely been considered from a behavioral genetic perspective. Third, there is ample evidence of a relation between AL and behavior problems, but the question of how the relation arises has not been examined. Finally, we do not know which genes are involved in AL. The goal of the proposed research is to use a multi-method approach to explore these unanswered questions in a sample of 300 toddler twin pairs. At 2 and 3 years of age, twins will wear mechanical motion recorders for 48-hours in the home and in laboratory activity episodes. Observational ratings of AL and parent ratings of AL and behavior problems will be obtained at each age. DNA will be obtained from cheek scrapings collected from twins and parents. The specific aims of this project are to: 1) Investigate the role of genetic influences on developmental change in AL; 2) Investigate genetic influences on situational change on AL within each age; 3) Determine the extent to which the agreement between mechanically-assessed AL and parent and observer ratings of AL is genetically mediated; 4) Examine genetic and environmental contributions to the relation between AL and behavior problems; and 5) Examine candidate gene associations with AL. The proposed longitudinal study of activity level will go beyond the basic nature-nurture question which asks whether genetic influence is important and will begin to address the etiological processes by which change in AL takes place The molecular genetic component to this research will complement the quantitative genetic analyses by providing the opportunity to identify specific genes for AL. Our multi-method, multi-situation approach will allow a more complete picture of child behavior, moreover the finding of similar results across methods should inform about the robustness of the results.

DESCRIPTION (provided by applicant): The behavioral and emotional competencies that emerge during the preschool period are predictive of school readiness and long term socio-emotional adjustment. Consequently, this period has been acknowledged as a critical window for intervention. There is an abundance of research indicating that specific temperament characteristics can function as risk or protective factors in child development. Much of this work assumes that temperament represents stable and enduring characteristics and overlooks the possibility that there may be age-related changes in temperament that impact development in other domains. Moreover, there are a number of methodological problems that have hampered research in this area (i.e., reliance on a single rater; measures insensitive to change). The proposed longitudinal, multi-method twin study will address these weaknesses. Latent Growth Models will be used in a genetically-informative twin sample to examine patterns of temperament change across the preschool period and their relations to behavior problems, prosocial behaviors and academic readiness-three behavioral areas that have substantial long-term significance. We will also explore the mechanisms responsible for developmental change, both in terms of parenting effects, and genetic and environmental influences more generally. The temperament dimensions of negative emotionality, positive emotionality, activity level, attention, persistence, shyness and inhibitory control will be longitudinally accessed via multiple methods (e.g., behavioral tests, observations, actigraphs and parent ratings) in a sample of 300 twin pairs at 3, 4, and 5 years of age. Parent report and observational measures of parenting behavior will also be obtained at each age. Developmental outcome measures will include externalizing and internalizing behavior problems, prosocial behavior, and academic readiness. The specific aims of the study are to: 1) Explore individual differences in developmental change across the preschool period using observational measures of temperament in addition to parent ratings; 2) Assess links between child temperament and parenting trajectories. 3) Explore relations between temperament trajectories and developmental outcomes at age 5; and 4) Examine genetic and environmental influences on individual differences in temperament trajectories. The knowledge gained from this study will enhance understanding of temperament development and its relation to children's mental health outcomes and will contribute significantly to early childhood interventions through aiding in the early identification of high risk children, and providing insight on the early malleability of risky versus protective temperament characteristics. PUBLIC HEALTH RELEVANCE: Temperament has been linked to a wide variety of developmental outcomes. Knowledge of the factors influencing temperament development and of developmental pathways linking temperament to outcomes would reveal potential targets of prevention and intervention, and provide insight on the early malleability of risky versus protective temperament characteristics. The present study focuses on temperament during the preschool period because behavioral and emotional competencies that emerge during this period are predictive of school readiness and socio-emotional adjustment through the school years.

PROJECT SUMMARY/ABSTRACT The goal of this project is to understand how quantity and quality of the language input directed to children by their caregivers is influenced by genetically-influenced child characteristics (language ability as measured by standard assessment as well as children's productions during the interactions, and temperament as measured by standard assessments). The overarching activity of this research project is transcription and analysis of naturalistic parent-child interactions that have already been collected as part of a longitudinal study of over 300 same-sex twin pairs and their parents. The data set consists of one-on-one play interactions between a parent with each twin separately, at each of three time points (ages 3, 4, and 5 years). This unique sample allows investigation of how the parent's linguistic input to each child differs. Because half of the sample consists of monozygotic (genetically identical) twins and the other half of dizygotic twins (sharing approximately 50% of their segregating genes), we can make inferences about genetic influences on traits by the degree of similarity between members of a monozygotic and dizygotic twin pair. In this investigation we focus on how genetically-influenced traits affect the properties of the parent's linguistic input. Quantitative genetic model- fitting analyses will reveal bidirectional relations between parental linguistic input and child characteristics at the level of etiology by informing about genetic and environmental influences on parental linguistic input at age 3, 4, and 5; stability and change in parental linguistic input across age; and sources of covariance between parental linguistic input and child characteristics both within and across age. This is a timely investigation: much attention is currently being given to the study of environmental or parental traits that affect linguistic input, such as the parent's level of education, and how interventions might train parents within high-risk groups to improve the quality of the input they provide. The proposed work adds another dimension to these efforts by investigating child-specific features and how these may influence parental linguistic input above and beyond environmental and parental traits. The knowledge gained will reveal how these child factors contribute to the parental linguistic input children receive and the longitudinal consequences of this relationship.