Social networks--that is, collections of individuals linked by inter-relationships--are by now well-known important factors in understanding social and behavioral phenomena. Missing from prior considerations is the fact that these relationships represent sequences of dynamic short-term interactions, where each interaction reflects the concerns of a particular environment. Present survey-based research methods are not able to capture accurate, detailed data on interactions within social networks over long timescales, and yet, such data would likely lead to significant new insights into the contextualized behavior of individuals, the way in which communities emerge, and the flow of information/transformation within human societies. This project brings together a multi-disciplinary team from sociology, anthropology, criminology, psychology, public health, and education, to identify critical research problems whose resolution would be advanced by the availability of dynamic interaction data. The research requirements of these diverse disciplines will be synthesized by computer scientists, leading to the design of a new cellphone-based system which will be capable of revealing the form and evolution of dynamic interaction networks in a privacy-preserving manner. To evaluate the design, a small-scale prototype will be developed, and applied in a case study exploring the impact of dynamic student interactions on individual academic performance. Potential future iterations of this tool will ensure the confirmed willing participation of study participants by the inclusion of a component requirement to ''opt in'' in order to take part in the automated data-collection process. By leveraging recent advances in proximity-based network technology and data science to yield the design of a general-purpose cellphone-based system that can provide much-needed access to dynamic interaction network data, this project furthers a deeper understanding of human societies by enhancing long-term research capabilities across a range of social, behavioral and economic sciences. The case study will validate the system design, while also providing new insights on patterns of interaction in social networks, and their potential impact on student achievement.