Schneier on Security
A blog covering security and security technology.
« Los Alamos Explains Their Security Problems |
| Computer Virus Epidemiology »
February 17, 2009
Difficult-to-Pronounce Things are Judged to Be More Risky
Do I have any readers left who think humans are rational about risks?
Low processing fluency fosters the impression that a stimulus is unfamiliar, which in turn results in perceptions of higher risk, independent of whether the risk is desirable or undesirable. In Studies 1 and 2, ostensible food additives were rated as more harmful when their names were difficult to pronounce than when their names were easy to pronounce; mediation analyses indicated that this effect was mediated by the perceived novelty of the substance. In Study 3, amusement-park rides were rated as more likely to make one sick (an undesirable risk) and also as more exciting and adventurous (a desirable risk) when their names were difficult to pronounce than when their names were easy to pronounce.
Posted on February 17, 2009 at 1:56 PM
• 48 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
This surprises nobody who remembers the Coalition to Ban Dihydrogen Monoxide...
Which is riskier to handle, hemlock or Conium maculatum?
I think that humans are somewhat rational.. but we delude ourselves that we are completely rational. In general, we have to work at being rational, we have to train ourselves to take the time to let that part of our brain work... instead we lie and say that whatever decision we came to was the only rational one we could make at the time :).
At the risk of seeming irrational...
Can we read the actual article (or paper)?
Or are we limited to the abstract?
Or am I crazy to actually want to read TFA?
Randy -- okwhichisit
If you look at the list of worst food additives and they're not exactly words taught in a hooked-on-phonics class. The amusement ride study adds some validation though.
I'd like to know where acronyms land on our risk assessment meters. I've always viewed them with a raised eye-brow after I learning that MSM is mechanically separated meat!
And the fraudsters try to take advantage of this. There was one period of time while I was doing order verification that I had to stop my analysts from stopping orders with "foreign" sounding names, and start stopping orders with names like "Johnson", "Jones", and "Davis", despite flags that the latter were coming from unverified addresses. The fraudsters were counting on the "normal" sounding names passing muster and getting shipped for resale.
For a period of time, on close calls, I let go orders with strange names because of the bias of fraud orders with normal names...and did not get hit with credit card disputes on any that I recall. However, people who kept releasing orders to "Mr. John Smith" got burned a few times.
That's why Mohamet Wadi al-Hijarah is perceived as a greater security risk than Anthony Stone?
@: "In Study 3, amusement-park rides were rated as more likely to make one sick (an undesirable risk) and also as more exciting and adventurous (a desirable risk) when their names were difficult to pronounce than when their names were easy to pronounce."
I wasn't able to read the whole thing, just the abstract (I don't think the whole thing was accessible, if so I'm too impatient to find it).
I found the roller coaster parrallel interesting. I'd be curious what would happen if they named the same model two different things, and perhaps this was studied. But I suspect that the scarier, more sickening ride, may naturally have harder to pronounce names. People would probably be more likely to get sick on something called "Balthazar's Abyss" than "The Care Bear Coaster," and it probably wouldn't be due to the name.
Also couldnt read TFA. So, are they saying that the name caused them to be PERCEIVED as more likely to make one sick or it ACTUALLY did so?
If it is only perception then the amusement park should take a tamer coaster and name it "Poisonous Radioactive Violent Terrorist Death Machine" to increase ridership while decreasing sickness ( I assume sickness negatively affects the profitability of a park patron because they stop riding rides and buying food or junk).
@bob: "Poisonous Radioactive Violent Terrorist Death Machine"
I'd be less afraid to get on that than to endure "It's a Small World" ever again. Anyone who has had the misfortune knows of the intensity of the headache of that ride. So I guess there are exceptions to Study 3.
This finding is pretty easily boiled down to one small fact: ignorant people feel they're more at risk from things knowledgable people understand than they probably are.
I'd say this is the fundamental reason behind conspiracy theories: the people who hatch and maintain them are some of the most ignorant people in the world.
An example: Most people don't understand how the LHC works, and even fewer know anything but the very basics of particle physics (an extremity of ignorance). Ergo, they fear the first nutjob theory that comes along which suggests the experiment may create a black hole capable of destroying the Earth.
I think Bruce is dangerous because he has a difficult-to-pronounce last name....
(i still haven't quite figured out how to pronounce it)
I always figured Bruce's last name was pronounced "Sh-nye-er", to rhyme with "buyer" or "flier".
With regard to food additives, I suspect there is a tendency (not entirely irrational) to assume that unfamiliar, complex names are probably some sort of synthetic chemical, probably a preservative or artificial flavoring or coloring, which is therefore more suspicious than common traditional food ingredients such as water, sugar, or flour. Sure, you could list water as "dihydrogen monoxide", but no real food product package does that.
So, germans are less likely to fear Bruce (because they can easily pronounce his name)?
In other news, Welsh is the riskiest language on earth....
My take is that human decision-making is actually rational in a very sophisticated way. We tend to limit our idea of rationality to artificially simplistic rules. (That's why economics is such a shambles, but I digress). The real world is infinitely more complex, and our brains take a creditable shot at compensating for all the fluctuating variables out there.
Everyone wants to ride the renal-ubfuilsticaztor.
Regarding roller-coasters: perhaps people have learnt to "reverse-engineer" marketing? We know roller-coaster rides are meant to be scary or thrilling, and the better the ride, the more scary (or vice-versa). So we expect the most thrilling rides to have the most stand-out names.
I wonder of that's why a Lotus Espirit Turbo is harder to drive than my AMC Hornet was.
@HJohn: _thank_ you for awakening that particular earwig. B*stard.
The real question is, should we fear consumption of synthetic chemicals? Water may be produced synthetically by burning pure hydrogen and oxygen together. I would rather consume that than I would drink water from an unknown source.
Rationally, I know bacteria reside in most places water is obtained. I certainly would not drink untreated water up in the mountains - I have no desire to fight giardia lamblia or other nasties.
High Fructose Corn Syrup (HFCS) is a synthetic sugar, but very similar to traditional sucrose (cane sugar), which most of us are familiar with. There's 10% more fructose in it and the fructose and glucose molecules aren't held together by a (weak) chemical bond (which the body disposes of by necessity during digestion anyway). Every year, charges are made by people who fear HFCS causes obesity (with no science to back these claims up).
Do we fear it irrationally, knowing that some very smart people designed a process to extract it from corn, simply because it doesn't come directly from a plant we intrinsically trust (sugar cane)?
Do we really believe that companies put poison into our food just to save a few cents, knowing that putting poison into our food will reduce our lifespans (and, therefore, the amount of sales they can conceivably make to us)? Do we really believe that they feel they're so clever that they can hide poison in plain sight from people by describing said poison in chemistry terms?
Or is the simplest explanation the best - that they change all of the ingredients in the list to chemistry terms in order to make it more difficult for the layperson to duplicate their product?
I was exposed to that particular gag by Penn & Teller when they got a ton of people at an environmental rally to sign a petition to ban it - not a single one ever asked what it was.
There is a subtle distinction, which I hope wasn't lost, between synthetic water (produced as a byproduct of chemical reactions) and "natural" water (the kind you'd find.... everywhere). I trust synthetic water over natural water, hands down. I'd rather drink the water obtained from urine or sweat which is distilled by the new machine on board the space station than I would water from a mountain stream.
It's strange how far we'll take our fears though. Most people would not drink water produced from urine. Then again, most of us aren't aware that - according to the laws of probability - it may be taken as certainty that the water in our bodies has previously been a component of urine.
Regarding names and risks ... it's widely suggested that children who are taught the proper names of private body parts are less vulnerable to abuse.
Hmm ... familiarity makes something less like to be be perceived as "exciting and adventurous".
This would seem to echo ongoing suggestions that security depends on knowledgable observers. (!)
Interestingly, one of the most common sources of this information is the Catholic church.
(typical sample link ... more to be found if you go looking)
When we drink a glass of water we are normally drinking a lot more than H2O. It also typically contains loads of dissolved minerals which our body needs. If your water source comes from combining pure hydrogen and oxygen you will miss out on these other goodies.
Here's what the WHO says about drinking demineralised water:
Bruce, and what makes you believe that our "rational" risk analyses are any better than throwing dice? As I see it, most boil down to Bayesian analysis that doesn't account for informational limits, or undetected correlations.
You may be better off doing something random like being afraid of strange words, than consistently choosing a wrong rational analysis.
What if you combine the names in unexpected ways: "The Poisonous Radioactive Care Bear Coaster".
Kermit the bog,
All of those minerals may also be obtained from synthetic processes (in the form of dietary supplements). I'm making a case that "synthetic" may be better for us in many cases than "natural." Clearly, the inventor of the machine which now processes water onboard the space station is aware of this and the machine produces water which contains needed minerals (or the minerals come from other sources in the astronauts' diets).
Would that knowledge change the general "ickyness" people feel when asked if they'd drink that water? I imagine they'd probably think that was worse, not better, if the minerals are derived from the constituents of urine as well.
"Books? What are books?" Weena asked the Traveler.
Without reading the paper, it is likely a variation of fear of the unknown. An IUPAC chemical designation is complicated, obscure, and unknown to anyone who doesn't understand the syntax. Likely this is why pharmaceuticals name their drugs with simpler names, though even this doesn't always work to ease fears. The opposite effect is used by bureaucracies to make bad things sound better. Post-Traumatic Stress Syndrome sounds better than shell shock and lessens the emotional impact of the truth (Carlin pointed this out.). Death by multiple traumatic amputation sounds better than saying someone was blown apart. Chemotherapy sounds better than poisoning which it is. You are poisoning a patient to the point where you kill the cancer before you kill the patient, hopefully.
random drug namez
if your doc prescribed one of these, it means you're gonna die eventually!
> What if you combine the names in unexpected ways: "The Poisonous Radioactive Care Bear Coaster".
What happens is American Greetings will sue you.
> All of those minerals may also be obtained from synthetic processes (in the form of dietary supplements). I'm making a case that "synthetic" may be better for us in many cases than "natural."
Perhaps. On the other hand, there is a certain scientific arrogance in thinking that one has identified ALL of the components of [insert naturally grown foodstuff here] that the body has figured out some use for during its evolution. Chemical analysis of the natural stuff will only tell you about that which you already know you should be looking for.
@Ward S. Denker
High Fructose Corn Syrup is found mainly in American junk food. Americans are obese. QED.
Really, most diet studies are that good. Its too expensive to get a decent sample size and account for geographical, ethnic, age and sex variation. When you try correcting for these things your statistical test loses all power and its hard to publish a nill result.
This is not limited to diet studies either. A study on weather gore in video games was done on a sample of 1/3 males and 2/3 females. Despite the fact that the sales statistics indicate that the game they used in the study was played by over 90% males.
But here is a question. Why do we need some of these quite artificial flavoring etc?
Just because we don't know they are unsafe is not a proof that they are safe. And just because some dude believes they are safe does not make it so either. I mean why do we still have a tobacco industry if there weren't companies/people that are prepaid to make a buck even if thats shortens your life.
Incedently I drink water from mountain streams all the time. I have yet to get sick from it.
Aha! That explains why Blackwater changed their name to Xe: now they sound even scarier.
I'm not sure why this is even a study. We spent hundreds of thousands of generations evolving in a world where the unknown can kill you. It's a positive evolutionary adaptation to be afraid of or at least wary of things that are unfamiliar.
If something has a name or function that is familiar ("John Smith" or "Roller coaster") we're likely to be less apprehensive than if it has an unfamiliar name ("Ahmed Mujaswamifredistan" or "Large Hadron Collider")..
This is why I'll be walking past the swarthy street vendor with his foreign sounding food (felafel?) and having crunch balls in pocket bread with flavor sauce for lunch.
@John Hardin: "HJohn: _thank_ you for awakening that particular earwig. B*stard."
My apologies. That ride is definitely brutal. I'd suggest we move "It's a Small World" to Gitmo, but I think it would be too cruel.
Seriously, I'm still interested in the unknown dynamics of these studies. I'm not fully convinced the name itself makes people more afraid, or if more complex names are given to more dangerous/frightening things. Harmless things tend to be given simpler names like water, dirt, sugar, etc as opposed to "radioactive." But I do think there is a natural fear of something one cannot pronounce, being as it is probably less understood.
But no, I don't find people rational in their fears generally.
I'd like to see some results on how far such simple guidelines as "if you've never heard of it, be skeptical" take you astray under various circumstances. It's obviously fairly easy for marketers and scammers (and pharma companies, but I repeat myself) to hack this heuristic, but as a first approximation it may have served people well for a long time.
I have a hard time accepting that pronunciation of amusement park ride names is a reliable measure of risk.
Can the risk "judgement" in the mind of the average amusement park patron really be compared directly and without caveat to someone trying to cope with real risk environments?
They might as well have studied people attending horror/scare events. Does the average 14 yr old judge Ghoulandiapolis more "risky" than Hall of Zombies?
Bruce has a very easy-to-pronounce name.
Not as easy as "Chuck Norris," but still easy!
That would make the artist formerly known as prince the scariest thing in the world.
If Bruce's last name is of German origin, then it is pronounced to rhyme with flier. In the letter combinations "ei" and "ie", it is the second letter that governs.
However, the correct pronunciation for anyone's name is the way the owner says it.
This blog has done quite a few articles on people over-estimating risks. Interestingly, I have been doing a lot of reading recently on the opposite phenomenon.
It arose out of the eternal project management question: "why are people so lousy at estimating how long they will take to complete a simple task they've done a dozen times before? (And why are their errors far more often underestimates rather than overestimates?)" There are many aspects to this question, but one of them is that most of the time, most people underestimate the risk of things going wrong, so their estimate has an implicit "assuming everything goes to plan!"
Because of the substantial economic repercussions, there has been a lot of research in this area, and it is pretty well demonstrated that most of the time, most people overestimate the likelihood and magnitude of positive outcomes, and underestimate the likelihood and severity of negative outcomes. The phenomenon is called "optimism bias." The exceptions are the clinically depressed, who do not overestimate negative outcomes; rather, their estimates are on average fairly accurate, at least far more so than "normal" people.
Many interesting examples can be given. For example, if you say to a typical modern citizen of USA that they will very likely end up with a broken marriage, with themselves or a closely loved one a victim of rape or violent assault, and dying slowly and in agony from cancer, you might be told that your outlook is very pessimistic and gloomy. Yet in fact, all these outcomes have quite a large life-time probability.
We might wonder why humanity seems to have built-in optimism. Surely, evolution would favour a realistic appraisal of risks? But if risky strategies can have high pay-offs, then there is a situation in which it pays to gamble a little, even if the mean expected outcome is in the "House's" favour. I use gambling terminology here because that situation where this works is when you can play a martingale: if the population is constantly increasing, then the wins from the occasional spectacular success can more than compensate for the steady losses, even if they cannot do so in a static game, because every time you lose you increase your bet.
Like a martingale, though, the problem is that you cannot keep it up. If this hypothesis is true, it means that humanity's risk perception strategy is hard-wired to a perpetually expanding population, and becomes flat out wrong when population growth has to slow. In that scenario, our perceptions of risk must become gloomier if we are to make good decisions.
If a rational analysis leaves you worse off than rolling dice, it's not very rational, is it? Solomonoff induction may help if you're having trouble establishing priors and estimating odds.
I missed this before. Craig and Frances are right--it rhymes with "flier."
It seems these researchers have a penchant for this kind of study. They are from U of Michigan, and if you to an online search for them, you'll find that they've done studies about such things as fonts and certain tasks...specifically, that certain fonts are easier to read, and thus the readers think tasks/instructions written in those fonts are easier to complete.
Given the number of bad software systems we inflict on end users, and being an end user on many "enterprise" system... @DebW I'm all in favor of all the usability studies researchers can give us.
Next step - getting development teams to recognize there are more to an it system than a database, a webfront end and some cool python coding.
One of my favourite food additive labellings, obviously chosen to avoid customer panic, is found on the American "Morton's" brand of table salt (the one with the girl and umbrella).
The anti-caking agent (presumably present in small amounts) is listed using the 19th century name "yellow prussiate of soda" whose systematic name is the decidedly more scary "sodium cyanide".
As often observed here, many things are done in the interests of managing public perception, rather than addressing factual reality (cue TSA).
Re: Food additives. Nothing new here. Check this rather old Breyer's Ice Cream commercial: http://www.youtube.com/watch?v=Q-GlBK82nuA
@Ward S. Denker: "Do we really believe that companies put poison into our food just to save a few cents, knowing that putting poison into our food will reduce our lifespans (and, therefore, the amount of sales they can conceivably make to us)?"
Been to China lately?
Or, have you bought any peanut products from Georgia lately?
Schneier.com is a personal website. Opinions expressed are not necessarily those of Co3 Systems, Inc.