Expanding the Utility of Social Network Analysis for Multilevel Health Outcomes

社会网络理论有可能在多个层面上提高我们对人类健康问题的理解和治疗，但我们目前缺乏关于网络的空间和数学关系如何与关系的内容和质量相关以及这种变化如何影响健康结果的基本信息。我们建议使用非人灵长类动物来确定内部（例如，性格和气质、遗传倾向）和外部因素（例如，环境和社会压力）在多个人相互作用，影响网络结构和动态，以及这些，反过来，如何影响社会社区的健康结果。我们相信非人类灵长类动物模型为社会网络理论在人类健康方面的发展提供了几个优势，因为猴子为人类提供了认知和社会模拟，可以通过直接观察多个社区（提供统计复制）来收集数据，并且所有个体的遗传和社会历史都是完全已知的。我们有四个具体目标：（1）先进的理论和方法，在与健康背景相关的多个层面上评估网络动态和鲁棒性，（2）表征作用于个体的内部和外部因素如何共同影响网络结构和鲁棒性，（3） 量化网络结构和鲁棒性对作为健康结果的压力度量的影响，以及（4）评估网络组成的实验扰动对网络结构和鲁棒性以及健康结果的影响。将收集三个主要类别的数据：（1）亲和和攻击性互动的行为观察，（2） 个体内部因素，包括人格/气质的生物行为评估和5-HTTPLR和MAO-A基因的基因分型，和（3）健康结果的行为、身体和生理测量，包括Rhadinovirus脱落、C-反应蛋白水平、态度、水合作用、身体状况和创伤。八个社会群体中的每一个将在两年中观察78周。观察员将使用事件抽样设计记录个人之间的附属，侵略性和顺从性互动。将通过对每只动物的50种性格特征进行评级来评估性格/气质。每天和例行围捕期间将测量几项健康结果。行为数据将用于构建各种社交网络，其结构，动态和鲁棒性将被测量，随后使用多层次广义线性模型分析内部因素，健康结果和行为测量。 公共卫生相关性：这项研究将推进目前对社会网络结构和动态的理解，并开发新的网络措施和技术，以进一步了解社会网络理论如何成功地应用于对健康结果的理解，并最终改善人类健康。

DESCRIPTION (provided by applicant): Social network theory has the potential to improve our understanding and treatment of human health issues on multiple levels, but we currently lack the basic information on how the spatial and mathematical relations of networks relate to the content and quality of relationships and how such variation influences health outcomes. We propose to determine, using a nonhuman primate, how internal (e.g., personality and temperament, genetic predispositions) and external factors (e.g., environmental and social stressors) in multiple individuals interact to affect network structure and dynamics and how these, in turn, influence health outcomes in social communities. We believe a nonhuman primate model offers several advantages to the advancement of social network theory with regard to human health because monkeys provide a cognitive and social analog for humans, data can be collected by direct observation of multiple communities (providing statistical replication), and the genetic and social history of all individuals is fully known.)We have four specific aims: (1) advance theory and methodologies assessing the network dynamics and robustness at multiple levels pertinent to the health context, (2) characterize how internal and external factors acting on individuals collectively influence network structure and robustness, (3) quantify the influence of network structure and robustness on metrics of stress as health outcomes and (4) assess the effects of experimental perturbation of network composition on network structure and robustness and health outcomes. Three main categories of data will be collected: (1) behavioral observation of affiliative and aggressive interactions, (2) assessment of individual internal factors including biobehavioral assessment of personality/ temperament and genotyping of the 5-HTTPLR and MAO-A genes, and (3) behavioral, physical and physiological measurement of health outcomes, including Rhadinovirus shedding, C-reactive protein levels, attitude, hydration, body condition, and trauma. Each of eight social groups will be observed 78 weeks across two years. Observers will record affiliate, aggressive, and submissive interactions among individuals using an event sampling design. Personality/temperament will be assessed by rating each animal on a list of 50 personality traits. Several health outcomes will be measured daily and during routine roundups. Behavioral data will be used to construct various social networks whose structure, dynamics, and robustness will be measured, and subsequently analyzed with respect to internal factors, health outcomes, and behavioral measures using multi-level generalized linear models.) PUBLIC HEALTH RELEVANCE: This study will advance the current understanding of social network structure and dynamics and develop new network measures and techniques to further understand how social network theory can be successfully applied to the understanding of health outcomes, and ultimately to the improvement of human health.