Spring 2022 PROGRAM (Dartmouth College)

Bio: Dr. Jiawei Sophia Fu (Ph.D., Northwestern University) is an Assistant Professor at the School of Communication and Information of Rutgers University. Her research interests revolve around social networks, digital technology, organizational communication, and social innovation. Fu is dedicated to applying mixed methods to answering one question: How can organizations more effectively tackle grand societal challenges.

Abstract: Due to the significance of organizational innovation for human, social, and economic development, research has long sought to understand how existing knowledge and information may be recombined in innovation efforts. In this presentation, I focus on two distinct, yet interrelated factors for organizational and social innovation. In Study 1, I investigate how social networks influence public health innovation adoption in an interorganizational system in India. In Study 2, I examine the paradoxical effects of communication visibility in organizational social media use for social entrepreneurship organizing in China.Taken together, the findings of the two studies underscore the impact of network and technological factors on organizational innovation. This research offers important theoretical contributions for research in social and interorganizational networks, technology management, hybrid organizing, and global social impact organizing. It also has important practical implications for organizational leaders, social entrepreneurs, and network managers to adopt or develop social innovations to more effectively tackle grand challenges, such as public health crises and social exclusion.

Paper: https://academic.oup.com/joc/article/70/4/497/5857779?guestAccessKey=efb2ae78-7c35-4b8b-b39c-546af44be069 

Website: https://jiaweisophiafu.com

Bio: Eve Marder is the Victor and Beinfield University Professor at Brandeis University.  She studies the dynamics of small circuits.  She is a Past President of the Society for Neuroscience, a member of the American Academy of Arts and Sciences, the National Academies of Science and Medicine, and recipient of numerous honors and awards.  She has a distinguished record of mentoring trainees at all levels, and has published both scientific papers and numerous opinion pieces relevant to the lives and training of scientists.

Abstract: In our new format "Pathways in Network Science", we aim to shed light on the exciting, diverse, and sometimes challenging career paths of women and nonbinary researchers in network science. In today's seminar, Eve Marder (Victor and Beinfield University Professor, Brandeis University) reflects on her path that led her to and through the research of neural circuits.

Bio: Mia Jovanova is a PhD candidate at the Communication Neuroscience lab and the Annenberg School for Communication at the University of Pennsylvania. Broadly, she uses tools from computational social science, neuroscience and network science to study how people change their health behavior. Her current work focuses on the neural and social bases of susceptibility to peer influence. 

Abstract: Physical inactivity and sedentary behavior, are important determinants of health, and health messages are important tools to influence these behaviors. While activity in regions of the brain’s default mode and salience systems are independently associated with attending to information, it remains unclear how these brain systems interact during exposure to health messages. Here, we examine how the integration between brain systems while viewing health messages relates to changes in behavior outside the lab. Using wrist-worn accelerometers, we logged physical activity among 150 participants continuously for 10 days. Next, participants viewed messages encouraging physical activity while undergoing functional magnetic resonance imaging (fMRI) and completed an additional month where physical activity levels were logged and the messages were reinforced with daily text reminders. We find that individuals with higher default mode and salience system integration during message exposure were more likely to decrease their sedentary behavior and increase light physical activity in the month following fMRI than participants with lower brain integration. Interactions between the salience and default mode systems are associated with message receptivity and subsequent behavior change, highlighting the value of expanding the focus from the role of single brain regions to larger-scale brain connectivity in studying health behavior change.

Paper: Brain System Integration and Message Consistent Health Behavior Change

Website: https://jovanovamia.com/

Bio: Termeh Shafie holds a PhD in Statistics and is currently working as a lecturer (assistant professor) in Social Statistics at the University of Manchester. She has previously worked as a postdoctoral researcher at ETH Zurich and the University of Konstanz. Her research is mainly focused on the development of statistical methods and models for analysing multivariate social networks. She loves working interdisciplinary and doing research that moves across scientific borders. As such, she has created R packages in which novel statistical network analytic tools are implemented and made accessible for fellow researchers to use.

Abstract: In multivariate statistics, there is an abundance of different measures of centrality and spread, many of which cannot be applied to variables measured on a nominal or ordinal scale. Since network data in majority comprises such variables, alternative measures for analysing spread, flatness and association are needed. This is also of particular relevance given the special feature of interdependent observations in networks. In this presentation,  multivariate entropy analysis is introduced and demonstrated as a general statistical method for finding, analysing and testing complicated dependence structures such as partial and conditional independencies, redundancies and functional dependencies.  For example, consider the joint entropies of all pairs of variables which are used to construct a sequence of association graphs that represent variables by nodes and pairwise dependences above decreasing thresholds by links (cf. graphical models).  By successively lowering the threshold from the maximum joint entropy to smaller occurring values, the sequence of graphs gets more and more links. Connected components that are cliques represent dependent subsets of variables, and different components represent independent subsets of variables. Conditional independence between subsets of variables can be identified by omitting the subset corresponding to the conditioning variables. By comparing such graphs given different thresholds and with different components and cliques, specific structural models of multivariate dependence can be suggested and tested by divergence measures of goodness of fit. These and other entropy-based measures are highlighted and illustrated during the presentation by applications on social network data.  The applications show that important social phenomena and processes are often identified using these tools. The proposed framework is implemented in the R package 'entropy' and examples of using functions from this package are also presented.

Bio: Gursharn is currently a postdoc research associate at the Biocomplexity Institute. She completed PhD in statistics from Indian Statistical Institute and worked as a postdoctoral fellow at the National University of Singapore. She holds a bachelor’s degree in mathematics and statistics from the Banasthali University and a master’s degree in statistics from the Indian Statistical Institute. Her PhD thesis work is focused on the reinforcement models with applications in load balancing problems, and modelling opinion dynamics. Her work at NUS has been focused on developing fluctuation results for the subgraph counts in large random networks. She is also interested in studying different shapes for random phylogenetic trees. At the Biocomplexity Institute, she is working on epidemiological modelling and analysing missing infections.

Abstract: We introduce a general two-colour interacting urn model on a finite directed graph, where each urn at a node reinforces all the urns in its out-neighbours according to a fixed, non-negative and balanced reinforcement matrix. We show that the fraction of balls of either colour converges almost surely to a deterministic limit if either the reinforcement is not of Polya type or if the graph is such that every vertex with non-zero in-degree can be reached from some vertex with zero in-degree. We also obtain joint central limit theorems with appropriate scaling. Further, in the remaining case when there are no vertices with zero in-degree and the reinforcement is of Polya type, we restrict our analysis to a regular graph and show that the fraction of balls of either colour converges almost surely to a finite random limit, which is the same across all the urns.

This is a joint work with Neeraja Sahasrabudhe.

Paper: Interacting Urns on a Finite Directed Graph

Bio: Larissa Doroshenko is a postdoc in the Department of Communication Studies and an affiliate faculty in Network Science Institute at Northeastern University. Her current computational research projects explore how marginalized groups, ranging from racial minorities and women politicians to pro-democratic citizens in authoritarian regimes, use emerging media to gain power. Her previous research projects focused on “the dark side” of online media: populism, nationalism, and disinformation campaigns. Larissa received her doctoral degree in Communication Arts with a minor in Political Science from the University of Wisconsin-Madison.  

Abstract: Telegram, a cloud-based instant messenger, has become popular due to its encryption features and VPN accessibility. This messenger played an important role during 2020 Belarusian pro-democratic protests by organizing activists into local chats for various apartment buildings, neighborhoods, schools, university departments, and demographic groups, where participants discussed and exchanged the news, as well as coordinated future actions. Yet another feature of Telegram—information channels—has been widely used for spreading both verified information by traditional news sources and disinformation by semi-anonymous users.

In this talk I will first explain how the network structure of local Telegram chats helped to organize and sustain the protests in Belarus, specifically in the capital city of Minsk, which became an epicenter of this movement. I will discuss how these findings can be useful for other countries and social movements. Second, I will sketch out disinformation strategies employed on Telegram during the ongoing war in Ukraine and suggest the ways for exposing, studying, and fighting these strategies as users, citizens, and scholars.

Website: https://www.networkscienceinstitute.org/people/larissa-doroshenko 

Bio: Miranda Lubbers is an Associate Professor at the Department of Social and Cultural Anthropology of the Autonomous University of Barcelona (UAB), Spain. She directs the Research Group of Fundamentally Oriented Anthropology (GRAFO), a consolidated group recognized by the Catalan Government (2017-SGR-1325). Her research addresses social cohesion and social inclusion. In particular, she analyzes the role that formal and informal social relationships and settings have in the production, mitigation, or exacerbation of exclusion and segregation. In the area of social inclusion, she focuses on the embedding processes of migrants and the relational dimensions of poverty. She applies personal network analysis to obtain a detailed, micro-level understanding of processes of integration, cohesion, and exclusion. 

Miranda received her Ph.D. in Social Sciences from the University of Groningen, the Netherlands. After being a postdoctoral research fellow at the Groningen Institute for Educational Research at the University of Groningen from 2004-2006, she was awarded a Rubicon fellowship for young, talented researchers by the Netherlands Organisation for Scientific Research, in 2008, the Beatriu de Pinos fellowship by the Catalan government, and in 2010 the Ramón y Cajal senior research fellowship of the Spanish Ministry of Education.

In 2021, she received the prestigious ICREA Acadèmia award, which helps her intensify research in the coming five years, as well as the ERC Advanced Grant for the project "A Network Science Approach to Social Cohesion in European Societies" (PATCHWORK), which she will develop between fall 2021 and 2026. Earlier, she received the 2010 Award of Excellence in Research from her university, in 2014 the Award for Outstanding Research Trajectory to the effects of the I3 Program of the Spanish Ministery of Science and Innovation, and in 2017, her accreditation for full professor.

Abstract: In our new format "Pathways in Network Science", we aim to shed light on the exciting, diverse, and sometimes challenging career paths of women and nonbinary researchers in network science. In today's seminar, Miranda Lubbers (Associate Professor in Anthropology at the Autonomous University of Barcelona) reflects on her path that led her to and through network science.

Bio: Jennifer Watling Neal is an Associate Professor of Psychology at Michigan State University. Intersecting the fields of developmental psychology, education, and social networks, her research aims to understand how social networks are associated with health and well-being in public schools. Specifically, she has four interrelated lines of research that address: (1) contextual influences on child and adolescent behaviors and traits (2) social networks and the dissemination and implementation of interventions for children and adolescents, (3) child and teacher perceptions of classroom networks, and (4) the advancement of social network theory and methods.

Abstract: Early childhood is an important developmental period for social network formation. In the U.S. and in other countries, it is common for young children to enroll in preschool, providing one of the first opportunities to form positive and negative ties with other same-age peers. In this talk, I will present results from a systematic review of the developmental literature on social networks, demonstrating that few papers focus on preschoolers and even fewer consider negative ties. Next, I will introduce two key challenges to measuring social networks in preschool. First, traditional self-report methods of collecting social network data are difficult to use with young children. I will illustrate how observational data can be used as a solution to this data collection challenge. Second, positive and negative ties are difficult to infer from observational data. I will illustrate how bipartite projection backbone models can be used as a solution to this data analytic challenge. Using observational data from two preschool classrooms, I will demonstrate that bipartite projection backbone models yield signed networks with patterns of homophily, triadic closure, and balance that are consistent with theoretical and empirical expectations from the early childhood literature. This demonstration offers initial evidence of the validity of bipartite projection backbone models for inferring signed networks in preschool.

Website: jennawneal.com