Computational Social Science Laboratory

The Computational Social Science Laboratory (CSSL) investigates properties of cognition by using documents of the social discourse, such as stories, narratives, social media, transcriptions of speeches and news articles, in conjunction with behavioral and fMRI studies. CSSL is an interdisciplinary group, where researchers and students from psychology and computer science tackle similar research problems from different perspectives. We use machine-learning and natural language processing techniques guided by psychological theories to identify various social and cognitive properties of human psychology. Our focus is on theory-based natural language processing and social media analysis to examine the structure of human values without relying on preset questionnaires. Further, we are interested in how text analysis and machine learning algorithms can be used to examine neuro-semantic representations, and how these representations, in turn, can advance natural language processing techniques. The work we conduct plays important roles in the study of values and group dynamics, in addition to the general study of human language in Psychology, and in sentiment analysis and machine learning.

Investigators

Morteza Dehghani, PhD

Marlon Mooijman, PhD

Eunkyung Kim

Justin Garten

Reihane Boghrati

Joe Hoover

Kate Johnson

Katelyn Horton