Tag: science

Categories
Culture

Irrationality by Stuart Sutherland

The general message of Irrationality is that, if there has ever been a rational decision taken in the history of humankind, it was a fluke.

Sutherland goes through various different kinds of bias and error that are present in the way people think – obedience to authority, conformity to the group, a poor grasp of probability and statistics, being influenced by whatever you heard most recently, placing too much emphasis on unusual cases, looking for evidence that confirms your hypothesis and not evidence that contradicts it, being influenced by the order in which information is provided, placing too much confidence in intuition, being unwilling to cut your losses, and so on and so on – and for each of them he provides examples of psychology experiments that demonstrate that people systematically and repeatedly make the same stupid mistakes.

Rorschach inkblot

It’s a reminder that the scientific method is, in the end, just a whole series of elaborate ways to resist the tendency of the human mind to leap to the wrong conclusion. Not that science always gets it right first time as a result, but least at its best there’s a cultural understanding within science that it’s very easy to be wrong in lots of different ways and that you have to be very careful and methodical to try to avoid error.

It also tends to suggest that anyone who has to make complicated and important decisions – politicians, doctors, judges, engineers – could usefully take similar care to carefully and methodically eliminate systemic biases in the way they decide things, because they’re almost certainly less good at it than they think they are. That’s true of all of us, of course, but most of the decisions most of us make aren’t actually going to have particularly serious consequences.

Anyway, the book. It’s mainly made up of lots and lots of examples – often with several experiments described in a single paragraph – so it’s somewhat dense, and I should probably read it again if I want to take it all in, but it’s well written, which helps. And always interesting.

Categories
Nature

How noble in reason! How infinite in faculties!

I’ve just started a book called ‘Irrationality’, about the irrational behaviour of human beings. So far, much of the general drift has been fairly familiar, but no matter how many times you get told about the untrustworthy tendencies of the human mind, the specific experiments are still startling. Three that happened to jump out at me:

‘In one study, a telephone call was made to a nurse by someone claiming to be a doctor in the hospital whom she had never met. he told her to give a patient a 20 mg dose of a drug called Aspoten (in reality a placebo), adding that she must give it immediately because he wanted the drug to take effect before he saw the patient, when he came to the ward. He added that he would sign the prescription then. Despite the fact that he had ordered twice the maximum dose set out on the label and that there was a rule that no one should administer a drug before the doctor had signed the prescription, 95 per cent of nurses approached complied.’

‘Subjects were encouraged to give (sham) electric shocks to a stooge. When they were dressed like nurses they became less aggressive than those normally dressed, while wearing Ku Klux Klan outfits made them very much more aggressive.’

‘In a simple experiment, four short lines were each labelled ‘A’ and four slightly longer ones ‘B’. People saw a bigger difference in the average length of the two sets of lines when they were labelled in this way than when no labels were attached.’

Categories
Nature

Moths and meteorites

With National Moth Night and the Perseids, it should have been a good weekend for night-time stuff.

I didn’t have a lot of luck on either front. Really of course you need a moth trap to count moths effectively. I had a go at treacling—spreading a mix of treacle, brown sugar and rum on tree trunks to attract the moths—but nothing came. In the end my total count was three species; Jersey Tiger, seen earlier in the day, a Marbled Beauty attracted to the porch light, and a Double-striped Pug which came into my bedroom. Still, Jersey Tiger was one of the target species for NMN this year, so that’s good.

One thing becomes apparent walking around the garden at night; lots and lots of slugs.

leopard slugs

And here’s some hot slug-on-slug action.

slug sex

I didn’t try very hard with the meteorites, I must admit. And didn’t see any. But little white flashes of light appearing in the sky because the orbit of our planet is rolling through the dust trail of a long-gone comet seems like a good enough reason to post this:

Categories
Culture Nature

The Beak of the Finch by Jonathan Weiner

This book was recommended to me when I was in the Galapagos; I finally got round to reading it and I’m really glad I did. It’s an account of Peter and Rosemary Grant’s long-term study to measure the effects of natural selection on finches in the Galapagos. When this book was published in 1994, the study had been going for twenty years, but it’s still ongoing.

The choice of Galapagos finches isn’t just because of their iconic status in history of evolution; they’re an isolated population, they’re particularly variable, and a few very similar competing species live together in a very simple environment — only a few species of food plant, and almost no other small birds.

Over that period, they and their students have collected a staggering amount of data; detailed measurements of every finch on the island of Daphne Major, and records of who breeds with who, where their territories are, what songs they sing, what they eat, which territories are most productive, how the food supply varies from year to year and so on. That data has enabled them to show not just that tiny variations (in this case, particularly beak size) can have a measurable effect on the survival and breeding prospects of a bird, but that a change to the environment — a very wet year or a drought — can select for different physical characteristics to the extent of having a measurable impact on the average measurements of the population.

In effect, they have showed that you can observe evolution in action and that in the right circumstances it can happen extremely fast.

I really thought this was an excellent book. The detailed account of a single large research study sets it apart from all the other popular accounts of evolution I’ve read. There’s easily enough material to sustain a whole book and Weiner does an excellent job of communicating all the details with enough human interest to keep the book getting bogged down.

Categories
Nature Other

Fucking bootiful

It takes a lot to make me have sympathy for Bernard Matthews, whose company represents everything that’s worst in industrial food production, both the way they rear the turkeys and the revolting processed foods that they make from them. But I did get a twinge of sympathy when bird flu started killing all their turkeys.

I’ve got over that now.

The breeding ground for new strains of flu seems to be mass poultry production facilities around the world where huge numbers of birds are reared together in close contact with people. Ironically though, this outbreak isn’t going to do the free-range chicken industry any good. A chicken free to roam is a chicken free to expose itself to, and spread, infection. Perhaps if all the world’s chickens were reared in low-density, free-range conditions, the problem wouldn’t have started, but it’s too late for that now. And as far as I can gather from what they’re saying on the news, if H5N1 gets established here, free range will be thing of the past.

Categories
Nature Other

Crappy journalism in action

I nearly posted a link to this story on the BBC website about cows having regional accents because I thought it was mildly interesting. But the internet linguistics police quickly debunked it.

The BBC story starts by saying “Cows have regional accents like humans, language specialists have suggested.” What actually happened was that a PR firm working for a cheese manufacturer had called a couple of linguists and asked them whether there was any possibility that cows’ moos varied geographically. The answer was something like “well, it seems very unlikely, but it’s not completely impossible, because regional ‘accents’ have been observed in birds”.

Now I don’t particularly blame the cheese people’s people. They’re a PR firm. Spinning the truth is what they do. Trading on other people’s professional authority while misrepresenting what they actually said isn’t exactly attractive behaviour, but they’re salespeople and they are open about the fact that they’re selling you something. And to be fair, the original press release clearly bases its claims only on what the farmers have said. It’s not claiming to be any more than anecdotal.

But I do blame the BBC. They are the ones who reported this as a news story on their science pages, and who failed to call the linguists in question for a bit of fact-checking. They’ve actually made it worse by cutting out all the references to the West Country Farmhouse Cheesemakers. That’s presumably because they aren’t supposed to be providing advertising for anyone, but the result is that they’ve cut out any indication that the story is based on a press release put out by the PR firm for a dairy company, rather than, oh I don’t know, a paper in a scientific journal.

You’re probably thinking “chill out, Harry, it’s only a silly story about cows having accents”. And that’s probably what the BBC would say in their defence. Well, fuck that. I know that the fate of nations doesn’t hang on it, but if something is reported as news, I want them to have made the basic minimum of effort to report it correctly. Otherwise, why bother?

It’s typical of the media’s approach to reporting science. Or indeed just about any subject outside politics. They get all up their own arses about the importance of their role as protectors of democracy and speakers of truth to power, and the seriousness and integrity of their political journalism. And on their better days, all that stuff is true. But the moment they report on science, there’s a feeling that well, no-one can be expected to understand the technical details, so it’s alright to provide a watered-down and simplistic version; and anyway, it’s not very important like the political stuff (as though most political journalism was any more than gossip), so as long as it’s mildly entertaining, who cares if it’s really accurate? And then because that’s their attitude, they get all surprised when people like me get annoyed by it, because it’s ‘just’ a silly season story about cows, and surely people are media-savvy enough to know that it may not be held to the same standards as their political reporting?

Well, no. I actually care about the truth of these stories. Even the cow story; if it’s true, it’s interesting. If it’s not true, it’s just a waste of my time. I really feel quite strongly that if they’re going to do science and health reporting, they should do it properly. At the most basic level: if they get a press release about a piece of scientific research, they should call the scientists involved and make sure they don’t misrepresent them. And if it appears to be making an outlandish or controversial claim, call someone who can be expected to know about the subject and check with them. Otherwise just stop it. Stop reporting about science altogether if you can’t be bothered to get it right.

And if you think they’re more reliable when the subject, instead of cow accents, is something vitally important like vaccinations causing autism: *hollow laugh*