PROJECT SUMMARY / ABSTRACT
Many major life transitions, such as starting a new job or switching schools, involve learning new social
structures. Successfully learning these structures is critically important for feeling a sense of belonging,
improving academic or job performance, increasing well-being, and establishing relationships with others, but
failure can increase feelings of social isolation and loneliness. These feelings are strongly associated with a
greater risk for heart disease and other cardiovascular risk factors, such as hypertension and obesity. This
striking relationship between loneliness and negative cardiovascular health outcomes is an urgent public health
priority in the United States. While factors like social integration and adapting to novel social environments
influence both mental and physical health, relatively little is known about exactly how people learn new social
structures, and what factors make some individuals better social learners than others. Moreover, no empirical
research to date has examined the relationship between social-relational learning and loneliness using a
combination of behavioral, physiological, and neurocognitive methods. The proposed research aims to address
these gaps by conducting a longitudinal study during the real-time formation of a social network and measuring
biological and psychological signals that may explain individual differences in social learning success. In this
study, 300 individuals from the first-year Honors program cohort of undergraduate students – who live and take
smaller classes together – will complete a social-relational accuracy survey at three timepoints during the
academic year. Each participant will receive a roster of names for all cohort members and will be asked to identify
the strength of friendship with each name. Participants will additionally complete surveys assessing social
integration, loneliness, and social support. We will test the hypotheses that more socially integrated and
supported people will feel less lonely, and that social-relational accuracy is associated with individual differences
in loneliness. A random subset of 60 individuals will additionally participate in a neuroimaging experiment using
a previously-validated social-relational learning paradigm. Participants will view videos of interactions between
contestants on a reality television show and rate the degree to which contestants are friends or rivals with each
other. Resting heart rate variability (rHRV) will be measured continuously throughout the task. Using multivariate
analysis techniques, we will identify patterns of activity for individuals who are better or worse at identifying the
strength and valence of these viewed relationships. We will test the hypothesis that participants who are better
at identifying the quality of these relationships will show increased neural synchrony in the dorsomedial and
ventromedial prefrontal cortex, temporoparietal junction, and ventral striatum, will report feeling less lonely, show
higher social-relational accuracy, and have a higher rHRV. By illuminating a variety of mechanisms that support
social-relational learning and directly relating these mechanisms to individual outcomes, these findings may
support the development of strategies to reduce loneliness and the risk of cardiovascular diseases.
