Schneier on Security
A blog covering security and security technology.
« The Doghouse: Crypteto |
| Nice Use of Diversion During a Robbery »
September 30, 2009
Immediacy Affects Risk Assessments
New experiment demonstrates what we already knew:
That's because people tend to view their immediate emotions, such as their perceptions of threats or risks, as more intense and important than their previous emotions.
In one part of the study focusing on terrorist threats, using materials adapted from the U.S. Department of Homeland Security, Van Boven and his research colleagues presented two scenarios to people in a college laboratory depicting warnings about traveling abroad to two countries.
Participants were then asked to report which country seemed to have greater terrorist threats. Many of them reported that the country they last read about was more dangerous.
"What our study has shown is that when people learn about risks, even in very rapid succession where the information is presented to them in a very clear and vivid way, they still respond more strongly to what is right in front of them," Van Boven said.
Human emotions stem from a very old system in the brain, Van Boven says. When it comes to reacting to threats, real or exaggerated, it goes against the grain of thousands of years of evolution to just turn off that emotional reaction. It's not something most people can do, he said.
"And that's a problem, because people's emotions are fundamental to their judgments and decisions in everyday life," Van Boven said. "When people are constantly being bombarded by new threats or things to be fearful of, they can forget about the genuinely big problems, like global warming, which really need to be dealt with on a large scale with public support."
In today's 24-hour society, talk radio, the Internet and extensive media coverage of the "threat of the day" only exacerbate the trait of focusing on our immediate emotions, he said.
"One of the things we know about how emotional reactions work is they are not very objective, so people can get outraged or become fearful of what might actually be a relatively minor threat," Van Boven said. "One worry is some people are aware of these kinds of effects and can use them to manipulate our actions in ways that we may prefer to avoid."
"If you're interested in having an informed citizenry you tell people about all the relevant risks, but what our research shows is that is not sufficient because those things still happen in sequence and people will still respond immediately to whatever happens to be in front of them," he said. "In order to make good decisions and craft good policies we need to know how people are going to respond."
Posted on September 30, 2009 at 1:17 PM
• 10 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
"Many of them reported that the country they last read about was more dangerous."
This is interesting because, in his book, Kevin Mitnick explains what it always did when performing social engineering on a victim: after obtaining the valuable info, he made sure to always end the conversation asking for some inane information. He knew that the victim would remember the inane last questions -- and not the crucial ones.
Thanks for bringing this to our attention, Bruce! People's tendency to pay excessive attention to the latest trivial threat obviously poses a serious risk to our entire... Oh, forget it, I have to get out of here - a wasp just flew in my office.
There's an element missing -- these are threats that you have no power over.
In threats you can take action for -- it makes perfect sense to act on the most recent, the most salient, threat. In that case, our emotions are a fairly good approximations of a good risk assessment.
This is not an ancient and decrepit brain system -- but a vibrant system relevant to taking action. You first take action on immediate problems, whether or not they are "more important". Your emotions lead you to respond where your response is most efficacious -- not at the "greatest" risk, but the "current greatest" risk.
The problem is when unactionable data comes in. That 'information' displaces real data -- it's just distracting noise.
So it is not the first impression that counts but the last?
"a wasp just flew in my office"
It is not just the immediacy of a threat but also the immediacy of a response, that effects peoples judgment.
Minor problems with quick and easy solutions tend to get dealt with in prefrence to major problems with either slow or difficult solutions.
Infact this even over rides common sense on many occasions.
For instance you are sitting outside with your chosen after work bevarage and a wasp flies towards you, what do you do?
Some people try to frighten the wasp away which is actually more likley to get them stung.
The real solution is not to drink bevarages that attract wasps where wasps are likley to be, or to a lesser extent put something more attractive to wasps down wind of you.
Maybe I never have really been hit by wasps, but I have always got away with "just staying where I am, drinking said beverage a little more aware to not swallow a wasp, and telling that bugger to get off".
Well, in the light of this topic this probably is the best thing to do facing general threats as well.
Will it directly and outright harm me?
* if "no": get on with whatever you are doing
* if "yes": shift focus to the imminent threat
Wasps, for example, are generally not interested in stinging me. They might be interested in my beverage, well, but that is something to be dealt with. (Put something on the drink and the wasp needs a little more force than it can muster to get to the drink.)
"Stay calm", there is a reason why this is the first directive in every emergency instruction.
been working through Science of Fear (by Daniel Gardner and really worth your time) and Gardner describes our fear and risk measurement responses as a conflict between gut and head.
Only it's not really a conflict. Gut usually wins. This is an evolutionary shortcut that allows our limbic systems to react and save us. Frontal lobe doesn't get a say.
I used to fear bees, they moved fast and erratically I know they would cause pain. As a child pain was bad.
After driving cows for a while I learned to do what cows do, that is, a casual switch of the tail...(the hard part was growing the tail).
I now live in the woods with bees, wasps, hornets, woodpeckers and bears. I drink my beer on my deck and yes it attracts bees and wasps. It's thier world and a casual wave discourages them. They don't want the beer ... what are they going to do, pollinate it?
Point is it took years to overcome the gut "It's a bug! Kill it!" reaction and let head "calmly" react.
This study doesn't explain clowns though. That is missing from a lot of these studies.
``"What our study has shown is that when people learn about risks, even in very rapid succession where the information is presented to them in a very clear and vivid way, they still respond more strongly to what is right in front of them," Van Boven said.''
OK, but *how much* stronger is the response? Not very much at all, actually. Van Boven et alia's actual results for this study are given are given on page 10 (numbered 377) in study 5 in this paper:
(PDF, 465 kB)
Participants were asked to score the perceived risk from 1 to 7; the second risk had a mean score of 4.59 with a s.d. of 1.24, while the first risk had a mean score of 4.32 with a s.d. of 1.35. This tells us two interesting things:
1. The differences *within* these groups are around 5 times larger than the differences *between* these groups. In other words, while immediacy has an effect on risk perception, it is a very small effect that is dwarfed by other, non-assessed factors. Similar results apply for all the other statistical tests performed in this study: on a 7 point scale, a fraction of a point difference is seen in samples with spreads of well over a point.
2. Despite the fact that both of the countries used in the study had actually suffered multiple and on-going serious terrorist attacks, in some cases causing hundreds of deaths, very few participants ranked the countries as "very dangerous" and most put them close to the middle of the scale, with quite a lot putting them lower.
I remember from Psych 101 that memory tends to favor the first and last items heard in a sequence.
I'm not sure how well this translates from listening to a sequence of numbers to reading descriptions of dangerous places, but it seems to me that if the texts started out with a boring intro paragraph ("Eastasia is a largely agrarian nation...") before delving into the gory details, that might well be a factor as people would tend to remember the boring intro to country A and the gory details of country B.
Imagine what a skilled influencer like Derren Brown would be able to do with a Risk Comittee. He would have them all dancing around swatting wasps (probably imaginary ones at that)
Schneier.com is a personal website. Opinions expressed are not necessarily those of BT.