The Leak in Language

Researchers at the University of Fribourg set out to test a clean hypothesis: do people with conspiratorial thinking tendencies spontaneously construct conspiracy theories when confronted with ambiguous situations? They showed 385 university students the film "Leave the World Behind" — an apocalyptic thriller with a deliberately open-ended plot — and asked them to write essays interpreting what they'd seen.

The prediction was that people who score high on measures of conspiratorial thinking would fill the ambiguity with conspiratorial narratives. They didn't. Participants with strong conspiracy-belief tendencies were no more likely to write actual conspiracy theories than anyone else.

But when the researchers stopped looking at overall narrative and started looking at specific words, something surfaced. The conspiratorial frame leaked at the vocabulary level even when it didn't show up at the story level. People higher in conspiratorial thinking used words like "deception," "government," "elite," and "world" significantly more often — the vocabulary of conspiracy websites, embedded in essays that didn't actually tell conspiracy stories.

And there was something else the researchers called megalalia: a tendency to drop unusually sophisticated words into otherwise simple text. Not because the participants' overall vocabulary was broader — it wasn't, it was measurably narrower — but seemingly to signal authority, precision, unique access to knowledge. Complex sentence structures alongside suspicion-laden vocabulary, with occasional outsized words scattered throughout. The stylistic footprint of a particular kind of mind.

Here's what's worth sitting with: the researchers were measuring a gap between conscious content (the story you're telling) and the unconscious structure doing the telling (the words you're reaching for). These students weren't trying to write conspiracy theories. Many probably would have denied thinking conspiratorially if you'd asked directly. But the vocabulary they naturally reached for under interpretive pressure revealed the frame they were using anyway.

This is the structure of all worldview leakage. You curate the narrative — the explicit argument, the conclusion you present. You cannot as easily curate the vocabulary that feels natural to you under pressure. Words that seem neutral to you because they match your background model of how the world works are read as loaded by someone whose model differs. The language you reach for when interpreting something ambiguous is a more accurate map of your assumptions than your considered assertions are.

The researchers were careful to flag scope: the effect was small and specific. Conspiratorial thinking doesn't produce conspiratorial narratives on demand — the mindset showed up in word choice, not in grand plots. But that specificity is part of what makes it worth attention. If the belief only showed up in fully formed conspiracy theories, you could spot it easily. Because it shows up in vocabulary, it's harder to see in others and nearly invisible in yourself.

Consider this the next time you're interpreting something ambiguous — a news story with multiple plausible readings, an unexplained event, a conversation with a charged outcome. Notice what vocabulary feels natural. What words you reach for before you've consciously chosen a conclusion. What assumptions the language is already carrying.

The narrative is what you're saying. The vocabulary is what you actually think.

Both are showing.

i · sources

• A conspiratorial mindset subtly reveals itself in the words people choose — PsyPost, 2026-05-25