Schneier on Security
A blog covering security and security technology.
« Cheating at Chess |
| More on Chinese Cyberattacks »
February 21, 2013
Age Biases in Perceptions of Trust
Interesting research (full article):
Abstract: Older adults are disproportionately vulnerable to fraud, and federal agencies have speculated that excessive trust explains their greater vulnerability. Two studies, one behavioral and one using neuroimaging methodology, identified age differences in trust and their neural underpinnings. Older and younger adults rated faces high in trust cues similarly, but older adults perceived faces with cues to untrustworthiness to be significantly more trustworthy and approachable than younger adults. This age-related pattern was mirrored in neural activation to cues of trustworthiness. Whereas younger adults showed greater anterior insula activation to untrustworthy versus trustworthy faces, older adults showed muted activation of the anterior insula to untrustworthy faces. The insula has been shown to support interoceptive awareness that forms the basis of "gut feelings," which represent expected risk and predict risk-avoidant behavior. Thus, a diminished "gut" response to cues of untrustworthiness may partially underlie older adults' vulnerability to fraud.
EDITED TO ADD (3/12): I think this result reflects the fact that older people discount the future more than young ones, and therefore are more willing to gamble on a good outcome. It makes sense biologically; they have less future ahead of them. We see the same thing in pregnancy; older mothers have a higher threshold for spontaneous abortion of a risky embryo than younger mothers.
Posted on February 21, 2013 at 7:24 AM
• 18 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
1. No control or accounting for any generational differences. I.e., there's no indication that a particular person's trust increases with age, only that this particular generation's trust is higher than that generation's trust at this singular point in time.
2. 39 participants total. That's not really adequate for the sweeping claims they are making.
The study is not worthless, but it doesn't measure what they think it measures. Run this test a few more times with a 10-year difference between each run, or if you want it now, maybe run it in a few countries with a different socioeconomic and bad events history from ours, to see if the same trends hold up there.
This paper looks suspiciously like junk science and utter crap. The whole thing is predicated on the validity of trust assessments made simply by looking at a picture of a face.
1) is there any reputable work anywhere that would show that trust evaluations based on superficial physical appearance is valid?
2) isn't it more likely that such judgements reflect the most common prejudices within a society, rather than any true trustworthiness?
3) wouldn't scammers try hard to physically resemble the societal norms for trustworthy appearance, thus making judgements based on appearance less reliable - frankly this is what most 'business dress' is about - trying to look trustworthy.
4) Don't a lot of the frauds that seniors fall pray to involve communications with no visual component? (telemarketing frauds, advance fee fraud e-mails, etc.).
The paper gives no discussion on how the test images were assigned to trustworthy, neutral and untrustworthy categories, except to say they were previously assigned. Perhaps it's in the references cited.
The only thing that seems possibly valid about this paper is that seniors are less likely to discriminate against someone based on appearance alone. This might not be a defect, but perhaps is the result of hard learning that appearance is not a reliable guide to trustworthiness.
So do trusting people live longer and distrusting people die before they reach old age?
I think David nails it: younger people are more likely to fall for "cartoonish" ideas of who looks trustworthy or untrustworthy than older people, who've long since learned not to judge by appearances.
It's a lousy paper: a better methodology would be to give a couple of hundred subjects a couple of dozen images to rank in order of "trustworthiness" and look for ranking correlations with the age of the subjects (for extra fun, draw the images from the set of images of the subjects themselves - making sure no-one got their own, mind!)
Disclaimer: I'm over fifty
Another random snipe at the study: did they control for size of hometown as well as age?
This is one of those things that has generated lots of folk-wisdom, but may or may not be true. The assumption is that people who grew up in small towns are more trusting than people who grew up in a dense urban environment.
It's the kind of question that a careful researcher ought to ask.
BTW, what are the facial indicators of untrustworthiness?
I've heard that psychologists and cops have methods for assessing the truthfulness of a response. Basically, if the person answering a question is imagining their answer eyes shift up and in one direction. If the person is remembering the answer, the eyes shift up and in the other direction. (But some 5 to 10 percent of people exhibit an opposite pattern of eye-shift, which can make things really interesting for police investigators...)
Anyways, methods like eye-shift while answering questions are impossible to capture in photographs. So what did the researchers use?
A few things to note here:
1) The original study was almost certainly done with a sample biased toward younger people, so the labels should really be "faces younger people consider trustworthy/neutral/untrustworthy".
2) There's already a visible (although not 95% significant) difference in classification of neutral faces.
3) The older people do in fact classify the "untrustworthy" faces as untrustworthy, just not with the strength of negativity that the younger people exhibit, suggesting that some of this might have to do with differing levels of politeness in studies.
I'd love to see a study where a large population of faces is classified by older people first, and then younger people are tested on subsets.
Oh, and 4) although it's no doubt a matter of convenience, quite a large proportion of scams directed at older people are carried out by phone.
Yes further study is needed, but there could also be a evolutionary benefit to this behavior.
Assuming this is a change that happens with age, then an elderly individual of the population is more likely to take a risky behavior. The older generation becomes a higher risk taker as they age. This makes available the ability to find new success patterns by potentially sacrificing a member that now provides low contribution to the society.
As an example. The old antelope will stay at the edge of the herd, or lag behind the heard. While this increases the likelihood of predation for that individual, it also serves the byproduct that the more reproductive ages face less predations.
Another example. The old antelope can try to eat a plant that normally isn't eaten by the population. If consumption doesn't cause an issue the heard will find an alternate food source.
In both cases having the aged portion of the population become higher risk takers provides an evolutionary value to the society.
So while the study is a good prilimary examination and more needs to be done, I wouldn't classify it as junk straight away. It seems to establish the basis for further study of a potentially benefical evolutionary traight.
You call this "research"? (Everyone has already said what I was thinking.)
this reeks of pseudoscience. at best it is very sloppy "research".
they might be measuring a real difference in behavior, but their conclusions are based on an obviously flawed premise. Physiognomy has been long and thoroughly debunked.
I find this article/top quite interesting, as I'm employed by a firm whose sole line of work is investigating fraud. The vast majority of victims in our cases are elderly.
This study unfortunately misses the mark. Judging the trust-worthiness of someone or a business is more than just looks. At least from what I've seen, as people age, they regress until they eventually become like children again if they live long enough. Senility becomes naivety.
One of the worst scams we investigated was a securities fraud where the perpetrators would place ads in the newspaper for high interest CDs, and would drive the old people from their nursing homes to their offices. Once there, they would "make friends" with these people, gain their trust, and then sell them a bogus investment. All of the elderly people we spoke with had high regard for these people when we first met them. Some of the elderly accused US of being the bad guys, claiming that these nice gentlemen who sold them the bogus investments were honest, nice people and it's only the government's fault that they're not getting their money back.
Simple test for gullibility: do they watch a lot of television?
As @mrfox points out,
Physiognomy has been long and thoroughly debunked
Even in it's heyday in the Victorian era it was regarded by most who thought about it as "bunkum". But it unfortunatly lasted in "gentalmen researchers" and others with unscientific socio political views. It unfortunatly evolved into eugenics or "closed stud book breeding" in Europe in early 1900's right through untill well after WWII with humans, and can still be found alive and flourishing in the likes of the UK's "Kennel Club" with dog breading.
That said the problem with age related trust can be shown equally as well to late onset illness as it can to other factors as the early stagess of dementia etc are almost impossible to diagnose in the conventional ways. So the small number of test subjects would need to be checked over a period of time to rule late onset illnesses out.
But that aside one factor that needs to be considered is fairly well known and harks back to tribal times, and can be sumed up by the question,
"Why do humans out live their breeding life?"
In a tribal situation prior to agrarian activities you have the normal "hunter / gatherer" split. Males were usually hunters and died long before they ceased being reproductivly active. Women on the other hand were more often gatherers primarily because they were encombered by children, however their effectivness was improved if the younger women passed over child minding to older less fit women.
Now consider the difference in outlook, a young woman is placing her genetic future into the hands of another person, she has to be very wary of whom she trusts so in general is more distrustful.
The older women are beyond child bearing and their genetic future is viewed differently, she is not particularly interested in the trustworthyness of the mother giving her the child to look after, as long as she gets a share of the food the mother collects. What she does care about is her genetic future in her grand children and great grand children and ensuring they get to child bearing age.
Thus in a tribal situation the bidirectional trust equation of childbearing age to non childbearing age is not equal infact it is very unequal.
The trust a non childbearing older woman needs to have in a childbearing woman who is a mother is further reduced in slightly larger tribes, where they are not fed by the mother but as part of the tribes colllected hunting and gathering.
All the older woman realy has to trust is that the mother is not going to injure or kill her if an unfortunate event happens to the mothers child.
This risk is actually low because it's not in the mothers genetic future interest to engage in a dangerous activity if the child is dead or injured beyond genetic viability.
So in a tribal situation the older women don't have to trust the mothers very much if at all, where as the mothers do have to trust the older women a lot.
Similar reasoning applies to older men in tribal and other social groups.
So does this report investigate this fairly well known phenomenon? and does it take steps to rule it out?....
What's a trustworthy face? You had a bad experience with someone called Cathy and if someone else reminds you of Cathy, she will probably seem to be untrustworthy.
@trustworthy face: Even worse, you think that anyone named Cathy is untrustworthy.
I agree with David and others that this is junk science. The photos of faces methodology is complete nonsense. The neuroimaging had no value because no control studies were performed on the subjects by imaging them during expsores to objective untrustworthy and trustworthy situations.
So . . . the "children of the Sixties" are more trusting than the "children of Reaganism and austerity" because they're older, and for no other reason?
Perhaps the elderly learned by wisdom and experience that facial cues relating to trust and actual trustworthiness have no correlation at all?
It's not clear at all to me whether the "cues to untrustworthiness" actually have any correlation at all to actual untrustworthiness of the brain behind the face, or if they're just faces that previous test subjects have felt were less trustworthy.
It could just be that older people have correctly learned that "facial cues to trustworthiness" may be useful in casting plays and movies, but they're no use in deciding whom to trust in real life.
Schneier.com is a personal website. Opinions expressed are not necessarily those of BT.