Research

My Background

I am Mengxin (AvA) Ran | 冉孟馨, currently a lab manager in Prof. Feilong Ma's lab at the University of South Carolina. I was fortunate to be trained by Prof. Julie Golomb at Ohio State and Prof. Oriel FeldmanHall at Brown University.

What Drives Me

One Thousand and One Nights is my first sparkle: the clever vizier's daughter, the greedy and brutal king, curiosity against power, a fight between life and death. That was enough to get me hooked on the power of narrative. Then What Remains of Edith Finch showed me something else entirely: how tightly visual experience and meaning are woven together in storytelling, and how that weaving shapes what we feel and remember.

These experiences turned a childhood fascination into a scientific question. The same episode, the same conversation, the same reality, yet entirely different stories constructed inside different minds. How does the human brain parse continuous experience into structured narratives, and why does this process diverge so profoundly across individuals? That is the question my research lives with.

Three questions guide my work:

1. From experience to narrative: how does the brain segment continuous sensory input into discrete events and organize them into causally and temporally structured narrative representations?
2. Neural alignment and the limits of shared understanding: when two people experience the same event, where do their neural responses converge and where do they diverge—and why is consensus in creative production and interpretation so difficult to achieve?
3. Decoding idiosyncratic neural mechanisms from narrative: if different brains construct different narratives, can we reverse-engineer the cognitive mechanisms that produce them from individual-specific neural representations?

What I Study

I work at the intersection of narrative cognition, individual differences, and computational neuroimaging. To prepare for the questions I care about, I have built experience across several projects along the way.

It started at Ohio State, where I demonstrated that a moving object's location influences object identity judgments. This grounded me in basic vision science and taught me how the human brain binds visual features to spatial and temporal context. At Brown, I designed experiments showing that identical social cues produce opposite behavioral responses across individuals, which taught me how to measure the divergence in narrative construction at the behavioral level. Now at USC, I develop and evaluate hyperalignment algorithms on naturalistic fMRI data, separating genuine neural alignment from smoothing artifacts. This gives me the computational tools to compare how different brains represent the same narrative.

How I Build It

To compare how different brains represent the same narrative, I build computational pipelines that work at the intersection of geometry, statistics, and cognitive theory.

Methods: Hyperalignment · RSA · Searchlight Analysis · Mixed-effects Models · Bayesian Analysis · Machine Learning

Modalities: fMRI · EEG · Eye-tracking

Languages: Python · R · MATLAB