20 thoughts on “Looks like a job for the Amazing Randi

  1. Didn’t they have an episode on the TV show _House_ on this same subject a while ago?
    Dr. House disposed of the subject handily, of course.

  2. I think I heard about this cat some time ago. It was on Nature or maybe Nova.
    There must be something (the ketones mentioned?) that the cat smells or senses.
    I love Romeo’s comment at 1.

    I do know my cats acted differently when I was pregnant. I didn’t know I was, but they did. It could have been something unrelated (this observation was not falsifiable), but when I learned of my pregnancy (at 9 weeks–I wasn’t expecting to be expecting!) their behaviour made sense. I think their behaviour changed when I would have been about 6 weeks pregnant. It’s been nearly 7 years, hard to recall details.
    My dog did nothing special.

  3. Shrugs. Maybe there’s something to it. Warning! Anecdote ahead! My mom died at home under hospice care. Her dog and cat went about their business as usual until one day they both curled up at her feet (a place they never laid) and would not budge. She died that evening. Maybe there’s a smell emminent death gives off, noise humans can’t pick up. Maybe the animals are just picking their victims and sending brain waves that silently kill…bwa-ha-ha ha. Again….just wondering

  4. I don’t have much problem believing this. Obviously I’m a researcher and evidence is everything, so I would need more proof than just anecdote but there doesn’t seem to be anything particularly strange here.

    There seems to be some weak evidence that a small number of dogs are able to tell when a person is about to have a seizure. Although admittedly not as much evidence as the press would have us believe, it does seem to be a rare but real effect. It makes sense to me that they could be detecting a change in the person’s scent brought about by changing biochemistry.

    Similarly, a person close to death sometimes has a very distinctive smell. I can smell it and I’ve met others who say the same. It seems logical that organs entering the final stages of failure would lead to quite distinct and large changes in biochemistry and scent which animals could pick up without too much difficulty.

    The only odd thing to me here is that the cat chooses to curl up with people so close to death. Perhaps, far from being unsociable, it has learned that that is where it gets most attention?

    Basically, if I were James Randi, I’d be a bit circumspect about gambling my $1million on this being false.

  5. I thought either Randi or Ben Goldacre already addressed that one. At any rate, it’s bunkum as usual and was used as an example of confirmation bias. I think it might be Goldacre who wrote about it.

  6. Agreeing with #6 and #8, sounds like a severe case of confirmation bias here. However it is interesting to think that some cats can detect some sort of “death scent”. Either way, I’m certainly never going to own a cat – I’d become paranoid if it started getting too close…

  7. IanW: Oscar has been in the business for a long time, but the research that is reported here (via the link) is new. Not that I subscribe to the rule that if it is more than 9 days old it isn’t bloggable.

  8. I think rather than talk about confirmation bias straightaway, this might be an interesting opportunity to explore something unknown about animal physiology. I mean, it might be nothing more than confirmation bias, but they do seem to address that in the article, and the doctor they interviewed sounds like he’s a pretty sharp guy (he is also trying to sell a book, though.) Anyway, I think when you see something like this, it’s important to remember that there are plenty of scientific discoveries that begin with someone looking at something and saying “isn’t that odd?” and then doing the appropriate follow-up. Who knows, perhaps dying people really do release an odor that cats can pick up.

  9. Who knows, perhaps dying people really do release an odor that cats can pick up.

    Perhaps they do, but the first step in figuring out what’s going on should be ruling out explanations like confirmation bias.

  10. Yes, that’s part of what I meant by “appropriate follow-up.” I just don’t like to see people dismissing interesting observations out-of-hand.

  11. I just don’t like to see people dismissing interesting observations out-of-hand.

    I hope that’s not how I came across. I don’t see anything blatantly wrong with the hypothesis that something occurs shortly before death that is detectable by animals. What raised a flag for me was that the three articles I’ve seen on this cat in the past couple of days have all cited 50 hits without any mention of a rate.

  12. The New England Journal of medicine article from 2007 is a quick anecdotal account of this cat; it is not a scientific study. The reason Oscar’s in the news again is that the author of the article has now written a book about death and dying.

    Judging by the account, there is probably not a lot of confirmation bias here; the cat is unsociable and only hangs out with dying patients. Since the facility where this cat lives is for geriatric dementia patients, his “talent” mostly comes in handy so the nursing staff can give families a heads-up that their loved one is near death. Evidently, Oscar has not made a lot of wrong calls.

    I agree with pworthen’s post; this is an interesting observation, perhaps worthy of follow-up. There’s no reason to call it bunk, particularly. (I was not able to find the Ben Goldacre article, perhaps Madscientist could link to it)? In the meantime, file this under the same heading as the dogs who can smell impending seizures. Maybe Oscar just likes the way people smell when they die….

  13. confirmation bias my ass.

    c’mon people! what are the odds that the cat just happened to be around 50 people right when they died?

    obviously the cat killed them.

  14. So, it’s a real phenomenon? Interesting. But when I read about Oscar being ordinarily “unsociable”, I wonder if it’s not a case of the cat looking for a spot that is both warm (like a human body) and very quiet (like somebody very feeble and close to death). My two cents, anyway.

  15. You know, if a human were coincidentally present at over 50 deaths, there would be some kind of police investigation in progress. But evidently, just because they’re cute and fuzzy, domestic felines get a free pass.

Leave a Reply

Your email address will not be published.