New Twitter account flags article on research conducted in mice but implied to be in humans

We all know how bad popular science reporting has become, and here’s yet more evidence. This new Twitter account, justsaysinmice, is useful as it indicts not scientific studies but journalists who report scientific studies in mice as if their results were in humans. It’s not just that they say “this could apply to humans”, but they flat-out mislead the readers into thinking that they ARE IN HUMANS or ARE KNOWN TO APPLY to HUMANS.

Here are a few examples that I checked, and yes, the studies reported were all in mice. I love the moderator’s all-caps notation “IN MICE”.

Mice don’t eat bacon.

That is reported here. Note the picture of the pregnant woman, clearly implying this is in humans.

Are female mice “women”? Report on this study is here.

Original study here (IN MICE).

If you’re interested in science journalism and follow Twitter, this will be a site to watch. And maybe it’ll help hold journalists’ feet to the fire.

h/t: Grania



  1. pablo
    Posted April 14, 2019 at 11:45 am | Permalink

    “Are female mice women?” It depends on whether they identify as women, or are you a trans-speciesphobe?

    • Jenny Haniver
      Posted April 14, 2019 at 1:10 pm | Permalink

      I was thinking the same thing.

  2. Diana MacPherson
    Posted April 14, 2019 at 12:16 pm | Permalink

    Good grief this is bad.

  3. Caldwell
    Posted April 14, 2019 at 12:24 pm | Permalink

    The worst part is that mice never read those articles.

  4. Nicolaas Stempels
    Posted April 14, 2019 at 12:56 pm | Permalink

    Although there is no reason to extrapolate to humans )i agree 100% there), physiological findings in mice should give us food for thought, and -I hate this trope- some more research.

    • Posted April 14, 2019 at 3:44 pm | Permalink

      I can immediately say that in humans, exercise during pregnancy raises the risk of miscarriage.

  5. rickflick
    Posted April 14, 2019 at 1:16 pm | Permalink

    I met a “science writer” once. I wonder if she does this kind of thing. It’s unfortunate, but after this I’ll always be suspicious of the trade.

    • Diana MacPherson
      Posted April 14, 2019 at 1:22 pm | Permalink

      There are good science writers – for example Jennifer Ouellette, Carl Zimmer – but these here are probably not really science writers but journalists tasked with writing about science things.

      • Grania Spingies
        Posted April 14, 2019 at 1:40 pm | Permalink

        The other possibility is that there is nothing wrong with the writers, the fault lies with the sub-editors who write the headlines. (To be fair, many of the articles make it clear that the study was run on mice). A good headline is the headline that generates maximum clicks.

        • Diana MacPherson
          Posted April 14, 2019 at 1:42 pm | Permalink

          Yes, that’s likely as well as the people who select the images for the article.

          • infiniteimprobabilit
            Posted April 14, 2019 at 6:49 pm | Permalink

            That is so true. I have a permanent frustration with newspaper editors over any technical article concerning trains, planes or automobiles; you can be at least 90% sure they will insert a picture of the wrong train/plane/car, and I don’t mean just hair-splitting tiny differences but spectacularly wrong to the point of being misleading.

            The article would be better off with no picture, but they just grab the first picture of a train/etc that comes to hand and think their job is done.


        • tjeales
          Posted April 14, 2019 at 8:48 pm | Permalink

          I disagree. If the result is just in mice then what is the news value of the piece at all? I shouldn’t modify my behaviour based on a result that’s just in mice. I have no power to do the necessary follow up research to see if it applies to humans. I can see no public interest in outlets like Business Insider or Tech News Today publishing stories about preliminary mouse research week after week. I seems like lazy, thoughtless repackaging of University Press releases

  6. Posted April 14, 2019 at 1:48 pm | Permalink

    I spent years trying to get my mother to stop smoking. She continued until she died—at age ninety. At one point I showed her the evidence of tobacco on the mortality of mice. She told me “All that proves is that mice shouldn’t smoke.”

  7. Posted April 14, 2019 at 3:19 pm | Permalink

    It would give the game away if an article began with a line like ‘Scientists find that mothers that eat their newborn babies…’.

    • rickflick
      Posted April 14, 2019 at 5:06 pm | Permalink

      That sounds fascinating. Tell us more. *click* *click* *click* *click*…*click* *click*…*click* *click*

  8. CAS
    Posted April 14, 2019 at 3:31 pm | Permalink

    Nice Twitter site!

  9. Ken Kukec
    Posted April 14, 2019 at 6:03 pm | Permalink

    Mice don’t eat bacon.

    So they say.

    Don’t eat a lotta swine myself. But every so often I give in to the irresistible force of bacon or pulled pork BBQ.

  10. Posted April 15, 2019 at 4:37 am | Permalink

    To slightly defend the writers of the second and third articles (the second piece especially), the erroneous inference was arrived at by the researchers carrying out the experiment, not the journalists.

    Jun Seok Son, a doctoral student at Washington State University, who carried out the study, said: “Based on our findings, we recommend that women – whether or not they are obese or have diabetes – exercise regularly during pregnancy because it benefits their children’s metabolic health.

    Accepted, it’s still a dereliction of duty not to ask the question “how do you know what works for mice also work for humans?”

%d bloggers like this: