Smarter People are More Trusting

Interesting research.

Both vocabulary and question comprehension were positively correlated with generalized trust. Those with the highest vocab scores were 34 percent more likely to trust others than those with the lowest scores, and someone who had a good perceived understanding of the survey questions was 11 percent more likely to trust others than someone with a perceived poor understanding. The correlation stayed strong even when researchers controlled for socio-economic class.

This study, too, found a correlation between trust and self-reported health and happiness. The trusting were 6 percent more likely to say they were “very happy,” and 7 percent more likely to report good or excellent health.

Full study results.

Posted on March 27, 2014 at 6:52 AM38 Comments


GalupD March 27, 2014 8:14 AM

After the NSA revelations began, I’ve found it very difficult to trust use of the Internet. The lack of trust has had a substantially negative impact on my life and health.

Dave Null March 27, 2014 9:01 AM

Charles Darwin indicated about animal instincts of fear and protectionism as a form of survival method that was passed down. This study is purely irrelevant outlook of particular position.

Andy March 27, 2014 9:10 AM

This is a study among adults, so there could be strong elements of learned behavior influencing the results. I imagine that kids who are bullied or tricked quickly learn not to trust, even those who are not “intelligent”. Folks with strong self-confidence (from years of learning to see quickly through tricks) may be more willing to trust. The authors allude to this by their reference to “people [who] might be better at correctly evaluating whether people are trustworthy” — I suggest that trust evaluation process is a learned skill. I wonder what the results would be among (very) young children. I also wonder if the results vary with wealth or across countries with very different economic conditions (e.g., “third world” countries).

Rob March 27, 2014 9:10 AM

This is plausible:

The researchers posit that intelligent people might be better at correctly evaluating whether people are trustworthy, or whether a particular person is likely to act untrustworthily in a particular situation.

But it’s also plausible that by and large the ‘better educated’ (note: not necessarily smarter!) will live in wealthier, safer environments and therefore the less educated might have their lower level of trust tuned very nicely to the circumstances in which they live.

vas pup March 27, 2014 9:13 AM

‘It depends’ – this answer better correlates with not just intellectual wisdom(high score on IQ or like), but with social wisdom (EQ – where ‘E’ stands for emotional) as well. My point with trust and with any other social interaction: on first step of interaction treat other people as you want to be treated (trust included), but then adjust your treatment towards them based on feedback (applies to trust component as well as to commitment, loyalty, support, respect, etc.) meaning treat them on the second step of interaction as they treat you by reciprocity. NOBODY deserve more trust, respect, etc. than provided. Reciprocity should be the backbone on that – I guess.

dave null March 27, 2014 9:26 AM

This study is purely idiotic and completely bias in so many ways. I read many economic papers to understand correlation matching based on a number fudging. It is known as a statistics of correlation matching to change the conclusion. If we apply this logic, we are saying, “Liberals who trust their government are generally happier and wealthier than the Conservatives. The right wing is generally poor for not trusting their government.” We both know many CEOs are right wing and generally not trust their government. There are so many fallacies in this study. It is so laughable. It is very unscientific to not apply a behavioral approach of humanity. Revenue person earns is based on notions of friendliness and able to sell oneself. Trust and affections are major assets of acquiring it. It is part of sales tactics of analytical approach. The same tool utilized in social engineering. What a waste of time reading this.

Winter March 27, 2014 9:30 AM

Both vocabulary and question comprehension were positively correlated with generalized trust.

I think “vocabulary” and “question comprehension” are also positively correlated to reading history. Novels are all about other people’s actions and motives. Better understanding tends to lead to better trust.

Maybe, reading more novels leads both to an increased vocabulary and better trust?

In other words, there is a lot of work to do to disentangle all these correlations.

Peter Boughton March 27, 2014 9:34 AM

Our first measure of intelligence is a 10-word vocabulary test in which the respondent is asked to identify which of five phrases supplies the correct definition of a given word

Wow, that’s a great way to identify “smart” people. ¬_¬

paul March 27, 2014 9:43 AM

Note that the effects on health and happiness are both self-reported measures. To me, this says that people who are more willing to tell an interviewer that people can generally be trusted are also more willing to tell an interviewer that they are healthy or happy. And I think there’s little question that “generally, people can be trusted” is the socially expected response.

The other thing here is that we’re talking about a much wider range of intelligence than most of us hanging out on the intertubes encounter in day-to-day life: only about a third of the people at the bottom end of the scale were assessed by interviewers as being able to understand the survey questions well. So if you’re at or near the bottom end of the scale, lack of trust would be something grounded in experience: first, you’d be the kind of person that others tend to take advantage of; second, a fair number of interactions in life would be hard to comprehend, leading to a perception that people were violating your trust even if they were actually just violating your (mistaken) expectations.

I do wish, though, that the experimenters hadn’t lumped “no” and “it depends” together. The results for all three would be interesting, as would the results for “it depends” lumped with “yes”. (Although a really smart person would, of course, recognize that “it depends” was potentially subsumed by “in general”.)

Matt from CT March 27, 2014 10:06 AM

There may be social factors beyond just “socio-economic.”

Colin Woodward in his 11 Nations of America book

Lays out that the heritage of which “nation” your area was settled by is a stronger predictor of political affiliation than socio-economic factors.

It happens that areas that are very individualistic and more prone to violence, such as Appalachia and the Deep South, which probably directly impacts trust in others (Appalachia in particular tends to be “clannish” — think Hatfield and McCoys) are also low in education.

Places like Yankee and Midlands — think blue states and purple states — have more faith in communal organizations (trust) and also value higher education for themselves and their community leaders more.

Hitchcock March 27, 2014 10:33 AM

Maybe this study is just part of some US government plan to get people to trust them.

Anura March 27, 2014 11:28 AM

I imagine if you are poor, live in a high crime area, and constantly harassed by police, you are significantly less likely to trust people than if you are wealthy and live in a good area. Wealthy people are also more likely to have a good education.

dave null March 27, 2014 11:48 AM

No, Anura, maybe, most people here need to understand correlation and statistics. It is different than studying calculus. Parallel events aren’t necessary found to be true. That is known as a situational event or positioning. I can easily numerically prove, tall people will die more often than shorter people, then someone can create a random thesis from it.

keiner March 27, 2014 1:33 PM

Highly sophisticated nonsense.

Denmark is the happiest country in the world. At the same time they have the second highest use of antidepressants in the world.

Winter March 27, 2014 1:57 PM

“Denmark is the happiest country in the world. At the same time they have the second highest use of antidepressants in the world.”

I assume that being unhappy among so many happy people can make you depressed. 😉

On the other hand, maybe these antidepressants might actually work.

Never been in Denmark? I can totally understand that they are the world happiest people (I am from number four, the Netherlands).

dragonfrog March 27, 2014 2:02 PM

Given that it seems to hinge specifically on comprehension and vocabulary, rather than other forms of “smarts”, that makes a lot of sense. Most of the time, our sense of betrayal arises from miscommunications, not intentional backstabbing.

The more easily you’re able to understand when people tell you what they’re going to do, and to be understood when you request actions from them, the less likely their behaviour will surprise or disappoint you and feel like a betrayal – leading to more trust in people’s honesty.

dave null March 27, 2014 2:10 PM

No, Winter. That means we get into various political arenas. Does socialism truly reduce people’s worries compare to capitalism that pushes people to survive? Does a party model truly work when both socialism and capitalism models both have historically failed many times on various countries and succeed on few? What we have now, are these replications of party models or adoption of governmental changes based on cultures and history? How important is for politics to adopt around historical cultures?

Back to the topic, a collation of pattern matching in economic models to AI have been discussed so many times… Pop psychologists masturbation are often laughed at, even some economists and scientific behaviorists claims should be laughed at too. We can even get into various probability model discussion in finance why so many of them failed. End of the day, every thing is pushed by a perception model. Altering perception is proven to work in results. Numerically not well represented…

dave null March 27, 2014 2:16 PM

Dragonfrog, that means we get into your personal history and experience. Your observations might be true to yourself and that’s what has happened to you, but that might not necessary be true for others. Personal experience isn’t necessary scientific. Now, we are getting subset of philosophical discussion which isn’t directly applied.

DB March 27, 2014 2:36 PM

@ kashmarek they are talking about end-to-end encryption of data, where data is not just encrypted between browser and server as in the case of usual “https”.. but it’s encrypted in the browser, then it stays encrypted all the way until it reaches a new final destination (which could be your browser again for example). So the server (or the company you’re dealing with) never has access to the data in unencrypted form. The principle of what they’re doing is good, but of course the devil is in the details, as to whether their implementation is good or not, and I didn’t look into that (yet). What they’re doing itself isn’t very newsworthy, by the way, but that more and more people are thinking about it and doing this is… It’s a good thing.

Trust your neighbors but lock your doors March 27, 2014 3:34 PM

These trust studies are interesting, if perhaps flawed, but my gut reaction is the cliche “smart in school, dumb on the bus…”

I think Bruce’s last book mentioned the idea that most people in a civil society are generally trustworthy depending on the stakes, as in if you pick a (pseudo)random stranger to watch your bag while you use the bathroom, odds are it won’t get stolen, as most “strangers” would be willing to help someone out. But if a random person offers to watch your bag, you’re better served not trusting them. And if you pick a (pseudo)random stranger to hold your winning powerball ticket, you are apt to lose it.

Not that it is scientific, but don’t con artists have their greatest success when they use “intelligent” people’s self-confidence against them? I just don’t know if a great vocabulary is as useful as watching people’s eyes, observing body language, gut feelings from past experience, knowing you’re a target and what the stakes are, etc. when it comes to identifying opposition.

Winter March 28, 2014 3:23 AM

@dave null
“Does socialism truly reduce people’s worries compare to capitalism that pushes people to survive?”

I am not sure what you are commenting on. If it is about Denmark, it seems the Danes are mostly happy. You do not need the research, you can ask them. And they tend to agree.

And contrary to reports from people in the USA who have never been in there, Denmark is a nice country where you can totally imagine to live a happy life. Clean, good education, good health, safe, and people are wealthy too.

Evan March 28, 2014 5:46 AM

Postulate 1: More intelligent people are more capable of dreaming up and getting away with schemes for revenge or retribution.
Postulate 2: Everyone knows Postulate 1.
Conclusion: Intelligent people can be more trusting because they know they can make the net outcome of betrayal worse for the betraying person than maintaining trust, and count on others to realize this.

kronos March 28, 2014 8:17 AM

@ unsure: “I don’t even trust myself!”

I was sure I was completely trustworthy, until last week when I brought home an especially yummy bag of chocolate candy. I promised myself that I would ration the bag over a two week period but by bedtime I had already eaten half of it. Kronos not trusted any more!!

David Thornley March 28, 2014 5:28 PM

As somebody who’s intelligent and well-off, I can afford misplaced trust. If I lose a few thousand dollars by trusting somebody I shouldn’t, I can afford that without serious consequences. I have friends who would be seriously hurt by losing a couple hundred to a con artist. Trusting somebody where I live is unlikely to cause personal injury, which is not true in all neighborhoods around mine. I find that I’m happier trusting people, and can indulge that.

Wesley Parish March 29, 2014 5:24 AM

Back in the eighties and nineties of last century, there was an extremely rich New Zealander named Bob Jones writing opinion pieces for the newspapers. In one opinion piece I remember rather well, or rather the piece of it that I remember well, he was talking about an investor or some such creature that he knew, an open and trusting individual. He characterized him by saying that his peers in London or New York would eat him alive (so to speak) if he ever got amongst them.

At that time it was possible to invest in New Zealand businesses, but because of regulation, it was impossible to make big gains or suffer big losses.

It was a feature of both Bob Jones and this credulous critter, that all and any of whatever losses they had suffered, were from their own stupidity, and not from the market. And Bob Jones, who was white-hot on deregulation like it was some sort of apocalyptic religion, never made the connection, that the only reason his friend was so trusting was that the New Zealand stock exchange was regulated so as to take the threat out of it.

An aside: It strikes me that the US has suffered one of its worst defeats, economically, by the repeal of various financial regulations. After being sold on the benefits and joys of stock market investments by various self-interested individuals, the average American citizen has seen him- or herself taken for a ride and comprehensively robbed. So the US financial system is an inverted pyramid, everything depending on the apex not to crumble. And likewise dependent on propaganda and coercion to keep itself from coming under investigation. It really depends now on the US financial “industry” self-regulating in order to maintain trust; but that is asking for the moon.

Getting back on topic: trust is the result of the generalization of previous experiences. Do smarter people do a better job at generalizing? Or do they have better experiences to generalize from? Until that question is answered, this sort of thing will just be “pop psychology”.

Coyne Tibbets March 29, 2014 9:15 AM

I think the article incorrectly conflates trust (recognition of the benefits of cooperation) with trust (naïveté); these are not the same thing.

Humanity has achieved some of its greatest accomplishments through cooperation, but the fact that someone is willing to cooperate to their own benefit and the benefit of another, doesn’t mean they are unaware of the risks of “defection” (a la “Prisoner’s Dilemma).

It is interesting that another study, “Brain scans link concern for justice with reason, not emotion“, came out at the same time, that associates rationality with justice. (I realize the association beween being rational and being intelligent is loose, but it seems to tie somewhat.)

Consider these together: Intelligence/rationality leads to a sense of justice (fair play) and trust (cooperation). In that sense, the studies seem to dovetail nicely, with each aspect reinforcing the other.

I have to say that it also makes unfortunate implications for those who do not cooperate and are not just.

MikeW_CA March 29, 2014 7:31 PM

I wonder if it could be due to the effort some media and political organizations put into trying to make less intelligent people distrust others with whom they might otherwise find common cause?

Matt from CT March 31, 2014 1:33 PM

Consider these together: Intelligence/rationality leads to a sense of justice
(fair play) and trust (cooperation).

Except it’s not true.

I would say societal norms (culture) will affect trust and fairness much more.

On a very high view, are Russians less intelligent than Americans? Or Chinese? From what I see in the news and on the web, they have less of a sense of justice or trust than in the U.S. There’s a reason for so many webcam videos out of Russia and the Great Firewall of China.

You can start to break down the U.S. much more finely though.

Trust and fairness are valued most in areas originally settled (after disease destroyed Indian institutions and war took care of the survivors) by folks who followed a “Public Protestant” tradition of seeing salvation as predestined and the world as something that could be improved. And improving this world takes education, and accepting science, and acting as a community.

“Private Protestants” that dominate the nexus of religion and politics in the U.S. today saw the world as irredeemably wicked and individual salvation (did you accept Jesus as your savior?) as all that matters. It doesn’t matter whether or not you understand this world, or protect it, or whether your community is moving forward or not, only whether you accept Jesus.

They are capable of being just as intelligent and rational, but their cultural heritage profoundly affect how they chose to use that intelligence and rationality, and how they view justice and fairness.

Vermont didn’t have the lowest crime rate in the nation because they had the most lax gun laws up until a few states became even more libertarian on that issue in the last decade; they had the lowest crime and least restrictive gun laws because those Yankees fundamentally trusted and respected each other not because they were smarter but because their forefathers had. That cultural inertia of keeping your house painted, your lawn mowed, and not pulling out a gun to settle fights continues on.

The U.S. I would say values justice and trust more than Russia or China; within the U.S. how much those are valued then varies by region primarily based on the cultural norms of their settlers (and I’m sure in China and Russia similar variations are found.)

Snoglydox June 11, 2015 9:10 PM

“The authors say one explanation could be that more intelligent individuals are better at judging character and so they tend to form relationships with people who are less likely to betray them.”

In other words…they don’t trust people.

But if they don’t trust people, they must not be “smarter,” but if they are not “smarter,” then…

It’s on the internet, so it must be true!


Leave a comment


Allowed HTML <a href="URL"> • <em> <cite> <i> • <strong> <b> • <sub> <sup> • <ul> <ol> <li> • <blockquote> <pre> Markdown Extra syntax via

Sidebar photo of Bruce Schneier by Joe MacInnis.