Happiness May Be Catching

chrb writes "The NY Times Magazine has an interesting article about research, based on the long-running Framingham Heart Study, modeling real world social networks. It seems that tendencies to be happy, not to smoke, and not to become obese are passed between nodes in a directed graph in a way that suggests such concepts are 'contagious.' Well-connected nodes in the graph (i.e., people with more friends) are more likely to be happier than less-connected nodes, even when the edges represent more distant friendships. Individuals quitting smoking, or becoming obese, influence not only their immediately connected friends but also friends of friends, with the effect sometimes skipping the intermediary node. The contagion effect is most noticeable when a tendency is passed from one person to another of the same sex — friends of the opposite sex, including spouses, are not as influential."

Comments

Duh

[Jurily]

This is like that $8m study that found out men think differently than women.

Re:Duh

[coastwalker]

What a load of cobblers this report is. The connections are real but they are not causal. Smokers tend to like other smokers and avoid the ranting anti smoking brigade. These researchers are not worth the food that has been wasted on them. Happiness is not a universal measurement in any case, there are different kinds of happiness.

Research like this may sound ridiculous...

[Xerfas]

but it's quite important to be able through research and testgroups to actually show that it's true. Not only on this subject but on almost all subjects. Most of us know this for a fact, but sometimes it's nice to know the reason why a certain feeling like happiness suddenly shows for no apparent reason more then that your friends are happy. I have a friend who just got out of a mental institution whom I have been worried about for quite some time, now that she is out in the real world and feels better I can honestly say that my days have improved a lot. Not having to worry and this has affected people around me because I'm a happier person again. Rant ends here..

Re:Research like this may sound ridiculous...

[Thanshin]

I have a friend who just got out of a mental institution whom I have been worried about for quite some time, now that she is out in the real world and feels better I can honestly say that my days have improved a lot.

As hollywood taught us, that story has only a discrete amount of possible endings:

- Your friend will get into your house tonight and kill you. With an axe.
- Your friend is actually you, as you'll discover waking up covered in dry red stains and possibly a dead animal next to you.
- Your friend is now a vampire.

Notice how all those plots can be intermingled seamlessly for the sequels; also, as hollywood taught us.

Gaming it for more sex

[Swizec]

Are you saying that if I have sex with my girlfriend's friend she'll have more sex with me? Seems like a fairly interesting notion.

What if I have sex with a bunch of my girlfriend's friends, will that make my girlfriend's whole social circle all want to have sex with me at the same time? 'Cause I could totally live with that.

Re:Gaming it for more sex

[RuBLed]

FTFS -

"The contagion effect is most noticeable when a tendency is passed from one person to another of the same sex â" friends of the opposite sex, including spouses, are not as influential."

Not that I am suggesting anything other than we are on Slashdot.

Re:Gaming it for more sex

[Anonymous Coward]

Are you saying that if I have sex with my girlfriend's friend she'll have more sex with me? Seems like a fairly interesting notion. What if I have sex with a bunch of my girlfriend's friends, will that make my girlfriend's whole social circle all want to have sex with me at the same time? 'Cause I could totally live with that.

Don't forget that behaviour isn't the only thing that's contagious... :-)

Re:Gaming it for more sex

[Denial93]

There certainly are some social groups where way more casual sex is going on than in others. Geekdom isn't one of them, in fact geeks are one of the most monogamous groups of people I know. Some of the music scenes tend to make much less of a fuss about casual sex, as do the hard political left, the art scene and the ecologically-bent.

These are just stereotypes of course, I don't know whether there is hard data on this. Would be interesting, though. I am continually amazed at how much sex average-looking people are getting out of being part of certain scenes.

Not a new phenomenon

[Sobrique]

Peer pressure isn't a new phenomenon. Groups mutually conform, as part of their group identity. Which can be mutually positive, and can be mutually destructive. Particularly drinking/drug use tends to increase in much the same way.
I've also run into the 'domino wave' of couples getting married as well - you seem to get several over the course of about a year, and the same with dropping sproglets.

Re:Not a new phenomenon

[cerberusss]

Peer pressure isn't a new phenomenon. Groups mutually conform, as part of their group identity.

I don't think it's just about peer pressure and groups.

I've read the book by Neil Strauss in which he becomes a "pick-up artist". One of his techniques for impressing girls is to have you and a friend go into a bar and act like you're having fun. Laughing and joking is contageous to the girls, but they are not in your group, and neither is peer pressure involved there.

Re:Not a new phenomenon

[smoker2]

That is peer pressure. You think you have to know an individual to be influenced by them ? Peers are people in the same social grouping, not social group. i.e. all 16 year olds who like certain types of music, all retired people who buy at certain stores. They do not have to personally know all the other members of the group. In your example, I doubt that the 2 guys were doing their act in front of a group of pensioners. They were doing it to impress members of their age / peer group who they wanted to attract.

Graph theory

[should_be_linear]

"Well-connected nodes in the graph (i.e., people with more friends) are more likely to be happier than less-connected nodes"

So /. must be saddest place on earth.

Go into panic. Now.

[Thanshin]

I told you. Stupidity is not only deadly but also contagious through any information transmission capable medium.

That's why the extra terrestrial visitors fly so fast and with their radio turned off.

They're playing chicken.

"Did you hear? GX-3-ThBlarg just did a low fly at merely three fongs per chronocycle! And he turned the wave receivers on for FIVE SECONDS!"
"No way! He'll end up idiotized, like his big brother. He must already be getting fatter and sad."

They're using FAT as the example...

[Vintermann]

We well-informed teetotallers have known this for years about alcohol. Attitudes aren't made in a vacuum. If the drug/alcohol use of your kids, or even the use in society, bothers you, the first thing you should do is cut back (or better yet, cut out) yourself.

It was the French demographer Sully Ledermann who first suggested that alcohol consumption appears to follow a log-normal distribution - he didn't provide much evidence for it, but it turned out later he was completely right. In principle, a single variable is enough to describe the variation in total alcohol consumption across cultures: The average amount consumed. As the number of moderate drinkers increase, the number of heavy drinkers increases with about the square.

I'll quote (and translate) a piece of an article from the journal of the Norwegian physician's association:

"The stable traits and connections that have been found in this are are not natural laws, they could all in principle have been different. The suprising thing, however, is that the connections are as stable as they are.

These connections and regularities were at the outset pure statistical descriptions of reality, without any understanding of the social mechanisms that generated them. Through the 1980s there came some studies where one tried to explain how these regularities appear and are kept stable (9, 11, 13). The original hypotheses were one that drinking habits are explained by a series of factors that appear to combine multiplicatively, and another that alcohol users are strongly influenced by the drinking habits in their social networks.

Both hypotheses have good empirical support. The first one can, by the so-called central limit theorem in statistical theory, explain that the distribution becomes approximately log-normal. The second hypothesis can, from theories of interaction and spread in social networks, explain why there is such a strong connection between average consumption and the prevalence of high consumers."

Emphasis mine. Original article with references here: http://www.tidsskriftet.no/?seks_id=649944

Interesting concept, but....

[Niubi]

Essentially man is a social animal and has an inbuilt desire to fit in with the society that surrounds him/her. I'm not quite sure why an expensive and pointless scientific paper needed to be written about what is essentially a psychological and societal issue. Take DubLi, for example - it's growing because those who have used it are reporting positively to friends, collegues etc positively. It's not exactly rocket science, just my 2 cents worth.

Re:Interesting concept, but....

[stewbee]

Interesting response. Not trying to troll here. My wife is a scientist, so she has partially warped my mind to think like this. You realize that you yourself have made a generalization that on the surface seems quite plausible. Do you have any direct proof in support of your hypothesis, which I will assume is "Essentially man is a social animal and has an inbuilt desire to fit in with the society that surrounds him/her" Have you found any quotable research showing that your hypothesis has already been proven?

My point here is that until you actually do the research, you can generalize all you want, but that doesn't make it right. Everyone has some sort of anecdotal evidence which could seem to invalidate some research, but does that evidence fall outside of 3 standard deviations for example? Does your anecdotal evidence even have any relation to the original experiment

I am reminded of a recent Daily Show where John Oliver interviewed two different scientists about which primates humans most resemble. (I would link to this, but I am at work and can't get to comedy central). One scientist was arguing that humans were more closely related to Orangutans whereas the other scientist was going with the generally accepted Chimpanzee relationship. John Oliver was trying to get the 'Chimp' scientist to put down the other scientist's research with a 'yo momma' joke. John Oliver Gave the lead in "Yo research is so whack...". The 'Chimp' scientist said, "that it fails to verify the hypothesis".

This is a long way of saying that science is done to find things that seem possibly painfully obvious, and to validate it through experimentation.

Good article.

[AbRASiON]

I've always noted the point mentioned towards the end, discussing how a 'social hub' kind of person can leave their element (place of living, workplace) go somewhere and within a few short weeks become a social hub again, these people fascinate me (and probably most of us) often interesting, social, active and often fun.
I'm by far not one of them sadly - infact I'm the loner in the article likely to die fat and speaking to no one however doesn't change that I mostly agree with what the article says, despite being difficult to proove it of course.

Re:Good article.

[Sobrique]

Thing is, it's not actually all that hard to do. It just requires a bit of overcoming of the initial 'I don't want to interact' antipathy. If you're anything like me, you've been introverted for a lot of your life, because ... well, people just suck. It's true, the do. Everyone is in some degree an arsehole. That doesn't mean you can't like them, nor does it mean you can't appreciate the positive parts of them. There's relatively few who are outright poison in terms of relationships.
To become a social hub, all you really need is to be able to take an interest in everyone else. Start off by faking it, but once you've done that a bit, you've already got the level of background knowledge that you don't need to any more - it's basically the same as 'geeking' only this time the subject of your study is people and social dynamics. Accept the idiosyncracies of people without passing judgement, much like you would with a hardware platform. Take the time to figure out what they're good and bad at, and keep up to date with their revision history. From there, all it takes is a bit of spreading of invites when you choose to do something - e.g. if you feel like going to the cinema, circulate the notion - include time, venue and film, and invite people to turn up if they're interested. People will, and suddenly you're a social hub, and that's something that'll take fairly minimal effort to maintain.

Re:Everyone already knew this.

[LKM]

I find it amazing that you call the study a waste because everybody already knew what the results would be, yet then immediately contradict the results of the study.

Red Dwarf positive virus :)

[delibes]

http://en.wikipedia.org/wiki/Quarantine_(Red_Dwarf) for the unenlightened. Mr Flibble agrees with me, don't you Mr Flibble?

Nodes connected BECAUSE of attributes

[Memroid]

tendencies to be happy, not to smoke, and not to become obese are passed between nodes in a directed graph

Wouldn't it be more likely that these people that are happy, athletic, and don't smoke tend to make friends with other people like them, as opposed to this suggestion of viral happiness? I mean it seems pretty obvious that people who don't smoke are going to have a higher percentage of friends that don't smoke than those who do smoke. It's called a "lifestyle."

Re:Nodes connected BECAUSE of attributes

[PeterBrett]

tendencies to be happy, not to smoke, and not to become obese are passed between nodes in a directed graph

Wouldn't it be more likely that these people that are happy, athletic, and don't smoke tend to make friends with other people like them, as opposed to this suggestion of viral happiness? I mean it seems pretty obvious that people who don't smoke are going to have a higher percentage of friends that don't smoke than those who do smoke. It's called a "lifestyle."

If you'd RTFA (and no, I'm not new here) you'd know that this effect is called homophily, and that one of the criticisms of the study is that the researchers efforts to account for it were insufficient.

Re:Nodes connected BECAUSE of attributes

[chrb]

Wouldn't it be more likely that these people that are happy, athletic, and don't smoke tend to make friends with other people like them, as opposed to this suggestion of viral happiness?

Your point is brought up in the article: One is âoehomophily,â the tendency of people to gravitate toward others who are like them. People who are gaining weight might well prefer to hang out with others who are also gaining weight, just as people who are happy might seek out others who are happy.Christakis and Fowler argue that they have stripped out the confounding effect of homophily from their statistics, although some other researchers have disagreed.

I mean it seems pretty obvious that people who don't smoke are going to have a higher percentage of friends that don't smoke than those who do smoke. It's called a "lifestyle."

Why is it obvious? At one time it was "obvious" that smokers were the cool socialites that everyone wanted to emulate.

Re:Nodes connected BECAUSE of attributes

[darthflo]

Why is it obvious? At one time it was "obvious" that smokers were the cool socialites that everyone wanted to emulate.

And how would one emulate a smoker? By smoking yourself, maybe? So you start smoking. You take smoke breaks whenever your role models take them and happily light a fag or two, taking good care of that nicotine addiction. Already, your smoke breaks will probably be somewhat social. There's going to be more smokers around, they're going to be smoking (just like you!) and every now and then they're going to be asking you for a light. Some talking might ensue, names be exchanged and friend requests be sent back and forth. Congratulations, the percentage of smokers among your friends is likely to increase.

The great thing about smoking is it's addictingness. After all those breaks with your idols and perhaps a few new people, you'll probably run into a situation where you don't know any cool socialite in the vicinity. Doesn't matter, you'll still be taking a smoke break, it's not just about aspiring to the cool guys, it's also about getting that nicotine fix. Others will probably be in the same situation. Just like you, they'll be used to talking to others while smoking. Asking for a light or fag is a great conversation starter; and starting from your mutual love of processed tobacco, a conversation is easily started. Conversation leads to more smoke break leads to more conversations, leads to friend requests. And like that, two lonely smokers may find each other, get together, gift the world with a bunch of newborn (future) smokers. Shortly thereafter he dies of lung cancer, her next child is stillborn, pulling her into a deep depression during which she abuses her children. Finally, she takes her own life. Days later, her dead body and starved children are found by the landlord. While retrieving the bodies, the police accidentally rips open the wallpaper, freeing a large patch of old asbestos-containing isolation. The landlord proceeds to patch it all up with another layer of wallpaper, but having breathed asbestos nanoshrapnel for hours finishes his (due to smoking) already damaged lung off. He manages to call 911 and an ambulance is immediately dispatched. Unfortunately, on it's way to the landlord's apartment, the driver carelessy drops his cigarette. He looks down for a split second to localize the still glowing stub. While he's grabbing it, an unnamed Federal Agent Closely Resembling Jack Bauer sprints across the street in a vain attempt to stop the ticking countdown of a (novel, extremely deadly for the whole continental U.S., Hawaii and Alaska) bomb located a few blocks down. The ambulance slams into our facrjb, killing him on the spot (in a painful, slow way!). Only seconds thereafter, the countdown of the discussed explosive device hits 0:00:00. Smoking kills. But I seem to be digressing a tiny bit, so back to topic:

Doing anything (e.g. being hapy, eating meat, smoking) makes you more likely to spend time with people of similar interest and less likely to spend time with diametrically opposed people (emos, vegans, non-smokers) because the former will approve of your actions, the latter condone them and people, being social animals, tend to favour approval over condemnation. Simple as that.

Re:Nodes connected BECAUSE of attributes

[Biogenesis]

I've never studied social networking, but there's a chance that the viral model is still useful mathematically even though it's causal relationship is flawed. For example in semiconductor physics it is often useful to model electron holes as positive charge carriers even though only electrons are actually moving. Basically the idea I'm trying to put forward is that if a model has limits (and every physical model does) it can still be useful if these limits are well understood.

Not just fat

[chrb]

They also studied drinking: When it came to drinking, Christakis and Fowler found a different kind of gender effect. Framingham women were considerably more influential than Framingham men. A woman who began drinking heavily increased the heavy-drinking risk of those around her, whereas heavy-drinking men had less effect on other people. Why? In the age of frat-party binge drinking, you might imagine that hard-partying men are the most risky people to be around. But Fowler says he suspects women are more influential precisely because they tend to drink less. When a woman starts drinking heavily, he says, it sends a strong signal to those around her that it's O.K. to start boozing too.

Re:Hmmn.

[chrb]

From the previous story on male juvenile delinquent behaviour being contagious people made the same "correlation is not causation" statement. I think that at this point the majority of Slashdot readers are well aware that correlation is not causation, so I'll just copy/paste an appropriate response:

"correlationdoesnnotnecessarilymeancausation

Indeed, which is why the vast majority of studies that get tagged by the moronic "correlationisnotcausation" involve some application of Mill's Methods and/or statistical and theoretical inference to demonstrate causation based on the observed correlations.

What gets reported is the correlation, because reporters are even dumber than /. taggers, but the researchers generally have thought a little bit about elementary logical errors somewhere along the path of their experiment design.

The tag is particularly idiotic when you consider that every correlation is caused by something, so the OP here is absolutely correct: if you really believe that there is no relationship whatsoever between correlation and causation, such that you can reflexively dismiss every reported correlation with this little snippet of nonsense, then you're pretty much committed to nothing being caused by anything.

Tagging stories this way is completely vacuous. All it tells us is that you haven't read the study or considered whether the usual methods have been employed to properly infer causation from correlation. It would be as useful and relevant to tag all stories with "theskyisblue", which is true in one sense (although the sky happens to be overcast where I am right now) but is only true in a way that is a) known by everyone and b) adds nothing of value to the discussion."

Re:Hmmm

[azior]

monkey C, monkey sudo!

Re:Correlation does not equal Causation

[MyLongNickName]

Yeah, yeah, yeah... the knee jerk correlation is not causation. And I am sure you read the study to see how they accounted for this, right? You looked at their methodolgy and made sure that they were not looking at how habits changed over time (for example in the article: At the time, her cigarette habit didn't seem like a problem; most of her friends also smoked socially. But in the late 1980s, a few of them began to quit, and pretty soon Eileen felt awkward holding a cigarette off to one side when out at a restaurant. She quit, too, and within a few years nobody she knew smoked anymore.
) , and other factors that could explain this. And I am sure that at the end of your research you found that a grad student just plunked some nmbers into an Excel spreasheet and used the built in statistical function.

Yup, a long-term study spends significant time and resources researching something to come to a conclusion. But with your keen perception and research skills, you have totally debunked it. And the slashtards mod it up to +5.
 

Re:Correlation does not equal Causation

[chrb]

From the previous story on male juvenile delinquent behaviour being contagious people made the same "correlation is not causation" statement. I think that at this point the majority of Slashdot readers are well aware that correlation is not causation, so I'll just copy/paste (again) an appropriate response:

" correlationdoesnnotnecessarilymeancausation

Indeed, which is why the vast majority of studies that get tagged by the moronic "correlationisnotcausation" involve some application of Mill's Methods and/or statistical and theoretical inference to demonstrate causation based on the observed correlations.

What gets reported is the correlation, because reporters are even dumber than /. taggers, but the researchers generally have thought a little bit about elementary logical errors somewhere along the path of their experiment design.

The tag is particularly idiotic when you consider that every correlation is caused by something, so the OP here is absolutely correct: if you really believe that there is no relationship whatsoever between correlation and causation, such that you can reflexively dismiss every reported correlation with this little snippet of nonsense, then you're pretty much committed to nothing being caused by anything.

Tagging stories this way is completely vacuous. All it tells us is that you haven't read the study or considered whether the usual methods have been employed to properly infer causation from correlation. It would be as useful and relevant to tag all stories with "theskyisblue", which is true in one sense (although the sky happens to be overcast where I am right now) but is only true in a way that is a) known by everyone and b) adds nothing of value to the discussion."

Re:Correlation does not equal Causation

[MyLongNickName]

No, dipshit. I am making fun of your research skills. You didn't even bother reading the article, much less the original research, yet you see yourself fit to "debunk" it.

You can't even read the criticism against you correctly. How do you think you are fit to judge this study.

Go read this comment that was pointed out by another reader: http://science.slashdot.org/comments.pl?sid=1306647&cid=28734109

It does a better job than I did of why your post is intellectually void.

Re:Correlation does not equal Causation

[kno3]

You completely misunderstood his post.
His point was that grad students conducting a long term research project probably would have thought about this and would have designed their experiments and analysis accordingly. Simply writing off the study by saying "Correlation does not equal Causation" is unfair and, unfortunately for Chapter80, demonstrates a certain amount idiocy.

Re:Everyone already knew this.

[MyLongNickName]

Hi Seumas,

This is NOT a waste of money. This study began in 1948 to discover causes of cardiovascular disease. The data was very broad and included health habits, diet, and sociological information. This "study" simply poured through the already existing data to find other interesting bit of information.

So, while some money might have been spent, this was more of an anlysis of existing information. In some ways, it is a money savings as no new study needed to be conducted to glean this information.

If you are interested, Google the study. There is a lot information out there, and the study added a lot to our body of knowledge.

Re:Correlation does not equal Causation

[lapsed]

His criticism is valid. Despite researchers' methodological rigour, social network analysis can identify causation that just doesn't exist. One study, using the same design that had previously identified obesity as being contagious or caused by an individual's social network, found that height, headaches and acne were similarly contagious. Height could be a good predictor of friends' height but your height won't be changed by your friends' heights. Granted, I haven't read the article and I'm not qualified to know whether the authors used the appropriate controls in the right ways, but it bears mentioning that even an ostensibly solid design can produce misleading results when trying to establish causation.

Re:Correlation does not equal Causation

[MyLongNickName]

Let's recap

1) Data from a study which has been going on for 60 years is analyzed. Researchers reach certain conclusions.
2) Poster who admittedly did not even read the article, much less the study says this is a prime example of correlation is not causation.
3) You support #2 as some studies are flawed.
4) You admit you have not looked at the study

So, my question... what study could possibly EVER make it past your lack of rigor? Seriously. Please answer this question. No matter how well designed, you and other like you will criticize it. There is no way to defend as you won't take the time to properly analyze the methodology. Yet, despite not taking the time, you feel qualified to discount it.

"Hi, I don't believe in global warming cuz last winter was so cold (or maybe it was the year before). No, I don't care about global data for 50 years, cuz those guys are pointy haired, librul, academics".

Re:Finite amount of Happiness

[Sobrique]

You would seem to be correct. But I think that's less about happiness, than about how one defines success. Success is how much you exceed expectations by. Your expectations are set by looking at your 'peers'. Therefore to be 'successful' you need to be doing better than your peers are. Successful in turn, tends to promote feelings of contentment and happiness, because people feel that 'things could be worse'. More enlightened will realise what utter hogwash this is, but most will still go to work tomorrow, to work for a crust, to support their family/buy their house anyway.

Re:Correlation does not equal Causation

[Belial6]

Given the amount of money to be made by both sides of the Global Warming debate, as well as the huge amount of obviously and verifiably inaccurate information spewed by both sides of the Global warming debate, I don't think that it is the subject you really want to use as a way to mock your opponents.