Does living in urban areas cause cognitive and linguistic changes that affect your mental health? Can these changes be detected through our online interactions? This project at UMH will take stock of changes in urban mental health via large online social media datasets (i.e. Twitter).
Online differences in the language and social networks of individuals that live in urban versus non-urban areas will be studied to get a better idea of how behavioral, cognitive, and social factors interact over time and lead to different mental health outcomes. The ultimate goal will be to support future analysis and develop new toolkits to better understand mental health in cities.
Johan Bollen is a professor of informatics at Indiana University. He was formerly a staff scientist at the Los Alamos National Laboratory from 2005-2009, and an Assistant Professor at the Department of Computer Science of Old Dominion University from 2002 to 2005. He obtained his PhD in Experimental Psychology from the Vrije Universiteit Brussel (VUB) in 2001.
Bollen’s work is situated at the intersection of computational social science and large-scale data analytics. His research has focused on the complex interactions between human behavior, emotions, and cognition, in particular in online environments, with applications in the psychological and mental health aspects of computational social science.
For the past 15 years he has studied the complex dynamics of human behavior and emotions interacting with large-scale techno-social systems, such as the internet, social media, financial markets, governance, and scholarly communication. He has published over 80 publications from research that has been funded by numerous agencies such as the NSF, DARPA, IARPA, NASA, and the Mellon Foundation, leading to innovations, in particular, on significant questions of how human cognition and social behavior interact, and increasingly how those bidirectional interactions shape mental health, public health, and human well-being.
More information about the UMH Fellowship:
As a IAS/UMH Fellow Professor Bollen will investigate whether living in urban areas causes cognitive and linguistic changes associated with mental health that can be detected from online language. Professor Bollen and his team will examine the content of a very large set of geolocated timelines and Twitter data sets for longitudinal indicators and language features associated with changes in mental health using a variety of tools that his team has developed over the past 5 years, such as sentiment lexicons, indicators of so-called distorted thinking (thoughts associated with a variety of internalizing disorders), and other features, including online social network parameters, discovered by a range of supervised and unsupervised machine learning techniques that can be trained to detect the differences in the language of individuals urban vs. non-urban areas. The results of this analysis will be compared to public health indicators such as infection rates, excess mortality, and population density. The analysis will not be solely focused on COVID-19-related changes, but general differences in how the mental health status of individuals in urban areas differs from those in non-urban areas and which factors drive such differences. The present toolkit of analytical tools is specifically focused on US English, but can be translated to the Dutch context which would be part of the proposed research activities as a fellow. Furthermore, the analysis can be performed for “snapshots” of the data, e.g. to test general differences, and longitudinally over time, possibly revealing the dynamics of the complex interactions of behavioral, cognitive, and social factors that are involved in these differences.