Collective sensemaking, trust, and credibility in unfolding information environments.
I study how people make sense of unfolding situations and decide which interpretations are credible enough to rely on when authority is unsettled.
My primary empirical setting is severe-weather livestreaming. Weatherfluencers interpret radar, official warnings, storm reports, and other evidence for large public audiences while events are still developing. Their professional standing is often contested, which makes these livestreams a useful setting for studying how credibility and reliance are worked out in real time rather than settled in advance.
Across this work, I focus on the relationship between visible interpretive work and ongoing audience evaluation: what creators make available for inspection, how audiences question, compare, corroborate, and track revisions, and how those interactions allow collective sensemaking to continue under uncertainty.
Severe-weather livestreams are public, synchronous, and safety-critical. Audiences may need to use an interpretation while the event itself, the available evidence, and the interpreter's credibility are all still in motion.
I call these creators weatherfluencers: platform-native creators who provide real-time severe-weather interpretation for public audiences, often through livestreaming platforms such as YouTube.
My research does not suggest that formal credentials stop mattering. Broadcast meteorologists retained important advantages in trustworthiness, accuracy, and legitimacy. The puzzle is that weatherfluencers can still become consequential interpreters, giving audiences reasons to keep evaluating and sometimes relying on them even when institutional standing does not settle the question in advance.
Doctoral research in Human Centered Design & Engineering at the University of Washington. Advised by Mark Zachry and David W. McDonald (co-chairs), with committee member Kate Starbird.
The dissertation develops a set of linked concepts for explaining how collective sensemaking proceeds when credentials and institutional position do not fully pre-resolve trust in an interpreter.
Weatherfluencers occupy an unsettled position relative to the weather enterprise. They may perform interpretive work that resembles broadcast meteorology without the same institutional role, accountability structure, or publicly legible professional standing.
Creators make parts of the interpretive process publicly visible through source triangulation, narrated uncertainty, visual annotation, comparison, and revision. Audiences can inspect how an interpretation is being produced rather than seeing only the conclusion.
Audiences evaluate interpretation continuously through questioning, corroboration, comparison, correction, and revision tracking. Trust is therefore observable as ongoing monitoring and calibration, rather than only as a one-time judgment.
Livestream chat is part of the interpretive environment. Audiences contribute local observations, questions, confirmations, repairs, and other information that helps people work through an unfolding event together.
In an unresolved event, people do not always need final certainty about the source before interpretation can continue. A more immediate question is whether the interpretation remains credible enough to keep using while new evidence arrives.
I am extending this work beyond individual livestreams to study how relationships with platform-native interpreters develop across repeated severe-weather events.
That means looking more closely at the community and information ecology around severe-weather livestreaming: how audiences return to particular creators, how trust accumulates or erodes, and how local observation, ground-truth reporting, moderation, vouching, and correction become part of collective interpretation over time.
Weather is the empirical center of this research program, but the underlying mechanisms may matter in other settings where people must evaluate consequential interpretations before source authority or accuracy can be fully settled.
I am interested in testing where those mechanisms travel and where they break. I do not assume that a framework developed in severe-weather livestreaming automatically generalizes to other domains.
Full publication list, talks, and media: Publications & Talks →