Does crossing fields make your paper more disruptive, or just harder?

“Be more interdisciplinary.” Every funding call, every keynote, every senior colleague who has run out of specific advice tells you the same thing. Reach across fields. Borrow from the neighbors. It’s the safest thing you can say to a young researcher, because nobody will ever call it wrong. Safe advice and useful advice aren’t the same thing, though, and I’ve always wanted to know whether crossing fields actually makes your work land harder, or just makes it harder to do.

A 2025 paper by Alex J. Yang in the Journal of Information Science (doi.org/10.1177/01655515251330614) gives a straight answer, at least on average. Looking at 38 million papers across all fields from 1960 to 2020, Yang finds a clear positive link between how interdisciplinary a paper is and its “disruptive potential.” Papers that pull ideas from further apart tend to reshape their area more, not less. And the link has been getting stronger over time, more so in STEM than in the social sciences and humanities.

What the two words actually measure

Two words are doing a lot of work here, so it is worth being precise. Interdisciplinarity is an ex-ante input: what you drew on, read off from how varied your references are. Disruption is a post hoc score: whether later work cites your paper instead of the things it built on (you eclipsed your predecessors) or alongside them (you extended them). One is fixed the day you publish. The other only makes sense years later, once the citations arrive.

The slogan gets complicated

Now the part that undercuts the slogan. Over the same 60 years, interdisciplinarity went up and disruption went down. And bigger teams, which produce more interdisciplinary work, are less likely to be disruptive. So “be interdisciplinary” and “be disruptive” have been drifting in opposite directions across science as a whole, even while, paper by paper, the two move together. Yang also reports that the interdisciplinarity-to-disruption coupling is stronger inside larger teams, which does not obviously square with larger teams being less disruptive overall. I’m reading this from the abstract, so I would check the full text before leaning on that one.

To explain how interdisciplinarity feeds disruption, Yang points at a few candidates: how diverse the team is, how varied the references are, and delayed citation recognition. Those are the plausible machinery behind the correlation. They’re also, notice, three more things measured from the same paper trail, which is worth keeping in mind before you treat any of them as a recipe.

Where I get uneasy

Both quantities are built from citation data, and citation data flatters some things and hides others. “Interdisciplinary,” here, mostly means you cited far-apart fields. Stapling a neuroscience reference onto an economics paper counts, whether or not you did anything real with it. The disruption index is a clever proxy for “this changed the conversation,” but it’s a proxy: it’s sensitive to how many references you’ve, to a field’s citing habits, and to which database you trust. And “delayed citation recognition,” one of the mechanisms Yang points to, is a polite way of saying interdisciplinary work often sits ignored for years before it disrupts anything, if it ever does. The average is positive. The variance around that average is the part that lands on you personally.

What to do if you are the one deciding

So what does a researcher deciding how far to reach do with this? Not much, if you read it as “reach further, score higher.” A 38-million-paper average is a fact about the population, not a promise about your next paper, and the team-size paradox says the lever isn’t simply adding collaborators and citations from more fields. The honest reading is narrower. When a problem genuinely needs a second field, the evidence says crossing over tends to pay off in impact, and pays off more now than it used to. When the problem does not need it, importing a distant reference to look interdisciplinary is just noise that a citation counter happens to reward. And the payoff is not evenly spread: the effect is stronger in STEM than in the social sciences and humanities, so the same reach across fields is not worth the same everywhere. If you work outside the hard sciences, the average that this paper reports is quieter than the headline sounds.

I like this paper because it refuses the easy version of its own headline. Interdisciplinarity is linked to disruption, yes. Disruption is also falling while interdisciplinarity rises. And bigger teams do more of one and less of the other. That is messier than “cross fields, win bigger,” and messier is usually closer to true. Whether your particular bridge between two fields disrupts anything, the index will only tell you in a decade. Would you make the same bet if the payoff were that slow, and that uncertain?