Are you a budding science sleuth? Take our plagiarism quiz to test your skills

Close-up side view of a laptop computer keyboard being violently smashed with a metal hammer, throwing lettered keys into the air.

These days, I spend a lot of my time looking through published papers that I suspect are fraudulent. It is a gruelling task. But amid the tedium, there is some light relief, thanks to some hilarious phrases used by authors seeking to avoid detection by plagiarism checkers. It’s hard not to smile when ‘breast cancer’ becomes ‘bosom peril’, or a citation for ‘Baker and Logue’ become ‘Pastry Specialist and Logue’.

I retired in 2022 from the University of Oxford, UK, where I worked as a developmental neuropsychologist. I now track down research fraud as part of a community of scientific sleuths. We work together to identify papers that we consider to be problematic in some way, then send the evidence to publishers or post it on PubPeer, a post-publication peer-review platform. Our aim is for the problematic papers to be investigated and, if appropriate, retracted by their publishers. But, often, publishers are slower and more recalcitrant than many sleuths would like them to be.

Many people blame academia’s ‘publish or perish’ research culture for the rise in problematic papers. Without a large number of publications and citations, prospects for employment, promotion and glittering prizes evaporate. Some researchers cut corners, manipulate or fabricate data, or copy the work of other scientists without citing them.

The demand for desperate researchers to get their name on a publication in any journal has spawned semi-professional organizations known as paper mills. These sell authorship status, with prices that vary depending on the impact factor of the journal. Articles are typically generated using artificial-intelligence tools or plagiarized, and tend to find homes in journals that lack sufficient editorial scrutiny.

One of the few amusing consequences of these developments is the phenomenon of tortured phrases, first described in 2021 by computer scientists Guillaume Cabanac and Cyril Labbé and software engineer Alexander Magazinov1. They analysed huge volumes of research literature to identify patterns. Tortured phrases arise when someone — or, in most cases, a computer — replaces words in a published article with synonyms to evade plagiarism detectors and ultimately publish the result as their own work.

Cabanac and colleagues developed a free website called the Problematic Paper Screener, which documents tortured phrases and other red flags in publications. Detection of tortured phrases has been a communal effort by research sleuths, who document specific articles with tortured phrases on PubPeer.

Enjoying our latest content?
Log in or create an account to continue

  • Access the most recent journalism from Nature's award-winning team
  • Explore the latest features & opinion covering groundbreaking research

Original source Are you a budding science sleuth? Take our plagiarism quiz to test your skills

Back to home