People’s creditworthiness, it seems, can be seen in their looks | Hacker News Reader

People’s creditworthiness, it seems, can be seen in their looks

economist.com

32 pointstyn22 comments

Other factors to control for include piercings, haircut, facial hair, glasses, makeup, any clothes in the picture (shirt/hat), hygiene, and tattoos.

However, it really wouldn't surprise me if some aspects of personality can, on average, be discerned from facial appearances. In hunter-gatherer days, being able to make an accurate snap judgment of a stranger's trustworthiness would be a big evolutionary advantage. And it wouldn't surprise me either if genes that affect one's personality (e.g. those relating to testosterone) affect one's appearance as well.

but if this became widespread it would then become advantageous to be able to fool other people's sense of trustworthiness by displaying the trustworthy physical characteristics while actually being untrustworthy. ad infinitum...

Excellent point! The old evolutionary arms race.

I think in the end, the set of genes that wins this game of cat and mouse will be the one that has the strongest selection pressure. For example, being able to detect untrustworthy people might increase my survival chances by x%, and being able to fool people might increase my survival chances by y%. If x >> y, then one would expect that most modern-day humans would be good at sniffing out untrustworthy people; if y >> x, then modern-day humans would have difficulty doing this, since the untrustworthy people have evolved an extremely strong ability to escape detection.

> if some aspects of personality can, on average, be discerned from facial appearances

Testosterone and estrogen levels are quite apparent in facial structure. People with male jaw lines and brows are more assertive.

Preventing emotion from showing on your face is nearly impossible. If you try to hold a smile you don't actually feel, it is not difficult to others to notice intuitively. They may not realize exactly what aspects of your face didn't match up to the smile, but often they'll be able to observe that the eyes didn't look right.

So I'm not all that surprised, really. Ability to recognize and read faces, along with ability to recognize vocal stress, are both important skills for a human.

Most pictures probably aren't taken with the express purpose of showing them on the site.

It might be a couple of other things: first aspects of his background that are discernable through a picture, and second that person's idea of what constitutes a "proper" picture for this site.

The most interesting thing I learned is that there are these kind of sites...

Ah this is truly a business opportunity - a nice supplement to your FICO score forget mechanical turk:

Step (1): Get access to frontal photographs of a large random sample of borrowers along with their financial history and records of any defaults.

Step (2): Extract a vector of facial features from each photo. Similar to the feature vectors used to judge attractiveness in the work of Cohen-Or et al. http://www.cs.tau.ac.il/~dcor/

Step (3): Have users apply for a loan by submitting a mugshot. Use support vector regression or some other similar technique to predict the probability of default.

Step (4): Profit

Or maybe just the people with worse scores have the most need for money and fewer options, and thus look more desperate.

Their test subjects came from Mechanical Turk. Is that considered reliable? I'm skeptical. And if they were using Mechanical Turk, why didn't they get 1000 subjects instead of just 25?

I agree that Mechanical Turk is a huge red flag as far as quality goes, but 1000 subjects looking at 6,821 loan applications could get fairly expensive. That is 6,821,000 *(number of questions per loan app and it looks like seven factors just from the article) HITS. This could be fairly costly depending on the price that the researchers set per question.

I think you got it backward, read it again. He used the Mechanical Turk to put a number on each face rather than just yes/no. He then compared the results with actual loan applications, and found that it correlated pretty well.

Oh, I see now. Makes sense. Thanks.

There's some clarification on how the Turk was used at http://www.econsteve.com/?p=166 .

Nice find!

Uh yeah, the blogger is my son.

Oh, cool. I was wondering if there was a connection.

But do we react to the underlying physiognomy or the muscle tone (posture and expressions on their faces)? It would be hard to separate these two factors, but you could start by generating 3D faces with different physiognomy but the same neutral expression.

Makes sense: * If I'm usually hiding something, my muscle memory will keep my face in that shape * If I'm usually open, vice versa Of course if I really believe stealing is good, my face will be open. So it's basically showing others what I believe about myself.

Can you really come to a conclusion based on inputs of just 25 MTurkers? 25 is less than 30 (the minimum required numbers to assume normal distribution) and also all being MTurkers is also not a good sign of unbiasedness.

Sorry, can't pay attention to The Economist for science/tech writing. They have far too many misses vs. hits.

Unrelated, but the article had one of the best photos I've seen for one in a long time.

Reply on news.ycombinator.com