Research

My work utilizes naturalistic experiments, computational methods, and qualitative conversation analysis to explore the multifaceted nature of human interactions, encompassing its neural, behavioral, social, and linguistic dimensions.

At the UCLA Social Cognitive Neuroscience (SCN) Lab led by Prof. Matthew Lieberman, I conduct experiments using functional near-infrared spectroscopy (fNIRS) to understand what happens in the brain between two people who are getting to know each other for the first time. In this Conversations and Connection (ConvoConnect) project, some participants are asked to talk about a variety of shallow topics (eg. the weather today), whereas other participants talk about deeper issues (eg. the last time you cried in front of someone else). We found that discussing deeper topics can promote more interpersonal connections compared to shallow topics. Our findings related to neural synchrony, brain states, and neural prediction are in preparation.

At the Communicative Mind (Co-Mind) Lab led by Prof. Rick Dale, I use computational methods to process experiment data and simulated data. Collaborating with Joyce Jiang, we utilized a deep neural network (DNN) approach for integrating neural activities and facial expressions. We developed metrics like synchrony, clustering, and volume to evaluate the representation space. Our methods pipeline has been published as a CogSci Proceeding. Our findings related to multimodality vs. unimodality are in preparation.

Contributing to the NSF grant Identifying Multimodal Signatures of Coordination to Understand Joint Performance in Diverse Tasks led by Dr. Alexia Galati, we developed a computational framework to capture the interpersonal dynamics during collaboration. A key feature of this framework is the inclusion of a task context that mediates interactions. We also included states of communication—active, inactive, and inhibitory—in the context. Simulation results show that these task constraints can be a robust predictor of simulated agents’ behaviors over time. We also found that, for turn-taking patterns to appear, the context matrix contained at least one inhibitory parameter 99.8% of the time. Our paper has been published in Nature’s Scientific Reports.

Working with Prof. Tanya Stivers, I conduct qualitative research using Conversation Analysis. People tend to agree with one another in conversations to build rapport. However, the ability to “be blunt” seems to be an important defining characteristic of true friends. In collaboration with Keith Cox, I analyze the assessments participants make in the ConvoConnect dataset to understand how risk-taking practices manifest in getting-to-know-you conversations.

As a passion-driven researcher with a love for meaningful connections, I strive to bridge the longstanding divide between qualitative and quantitative methods to decode the mystery behind human connections, vibes, and relationships–ultimately addressing the loneliness epidemic. To learn more about this endeavor, please check out the DIMS Conference I host, which brings together experts to explore dynamic interactions and methodologies in this field.