Abstract

Every day, people strike up conversations—some remain surface-level, while others spark instant connection. What makes certain interactions click? Social interactions rely on subtle, moment-by-moment cues—a shift in word choice, a nod, a frown—that reflect internal psychological and neurocognitive processes. While classic interpersonal communication theories like social penetration theory (Altman & Taylor, 1973) describe how relationships deepen through reciprocal self-disclosure, they offer limited insight into how these dynamics unfold in real time.

To address this gap, I conducted a between-subjects experiment with 70 dyads of strangers engaging in "get-to-know-you" conversations on either shallow or deep topics. Brain activity was measured using functional near-infrared spectroscopy (fNIRS), a portable neuroimaging technology, and behaviors were recorded on video.

In this talk, I present a multi-method approach to studying how connections are formed. I will cover: (1) the neural signatures of stranger bonding, (2) linking the brain with verbal and nonverbal signals using an Information Theory framework and deep neural network approach, (3) theoretical modeling: a simple dynamic framework to describe interpersonal synchrony and complementarity, and (4) the Dynamic Interaction and Multimodal Signals (DIMS) Dashboard, a novel tool for combining qualitative and quantitative analyses of interaction.

Together, this work bridges neuroscience, machine learning, and qualitative analysis to better understand the initial formation of meaningful connections.

About the speaker

Grace Qiyuan Miao is a Ph.D. Candidate at UCLA specializing in multimodal communication and social neuroscience. She studies the neuroscience of interacting minds, investigating how neural, linguistic, and nonverbal signals dynamically intertwine to shape meaningful human connection. Her research asks: How do people form deep versus shallow connections—both with other humans and with artificial intelligence? How does the brain distinguish different kinds of interpersonal experiences? And how can we better understand dynamic human behavior to design practical interventions that foster meaningful human connection at scale?

To explore these questions, Grace designs naturalistic experiments using portable neuroimaging technology functional near-infrared spectroscopy (fNIRS), integrating brain data with verbal and nonverbal behavior through computational tools and qualitative conversation analysis. Her work bridges neuroscience, communication, and machine learning to capture the complex, multimodal nature of human interaction. More information about her projects can be found at gracemiao.com/research.

Please note that this talk will be given remotely and streamed at the Interacting Minds Centre

Free of charge - All are welcome