Interpersonal Relationships in the Modern Age Thesis Submitted in Partial Fulfillment of the Requirements of the Jay and Jeanie Schottenstein Honors Program Yeshiva College Yeshiva University May 2020.

Abstract

Mirroring refers to the conscious or unconscious adaptation of the mannerisms and behavioral patterns of a person's interaction partner. This study searched texting data for evidence of mirroring, and found a strong connection between emotion exhibited over text by one party, and the response of their peer. This was measured by looking at exclamation point usage and emoji use. Additionally, I explored trends in how communication changes depending on the gender of the two people conversing. When the two people texting were of opposite genders, formal language increased, whereas when the two individuals were of the same gender colloquialisms and slang phrases were used. Finally, an attempt was made at training a machine learning model to predict the relationship between the two parties conversing. Although the trained model outperformed a baseline model of random guesses, the model's success at distinguishing between different relationships was underwhelming. A framework for analyzing WhatsApp data was built with the hope of creating infrastructure that future research projects with access to increased data could utilize.