backtop


Print 51 comment(s) - last by MrBlastman.. on Jun 23 at 1:44 PM

Results suggests trolls aren't seeking praise or wanting to manipulate; other studies indicate they may breed more

While it's important to preface what is to follow with a reminder that correlation doth not causation make, a new study has found seemingly clear ties between internet trolling and a person's likelihood of being predisposed to the "Dark Tetrad".  But more interestingly it finds that only one particular negative personality trait is stroked by trollish behavior -- and its findings suggest trolls may be breeding.

I. Wait, What's a "Dark Tetrad"?

Psychologists have long sought to quantify what personality traits make a person unlikeable and viewed by their peers as a "bad person".  

Perhaps the best-accepted term to describes these traits originates in a 2002 paper by graduate researcher Kevin M. Williams, and Dr. Delroy L. Paulhus of the Univ. of British Columbia.  Published in the J. Research on Personality, the paper identifies what it calls the "Dark Triad" of traits.  These personality variables -- commonly considered undesirable -- include Machiavellianism, subclinical narcissism, and subclinical psychopathy.
The dark triad
The paper's critical finding was that the "Dark Triad" of personality traits were often correlated, or in other words someone who was narcissistic would be more likely to be psychopathic, and so on, statistically speaking.

The "subclinical" tag attached to some of these symptoms means that they are not directly indicative of an underlying disease.  Unlike more serious symptoms of clinical psychopathy and/or narcissism, which can be used to form a diagonosis, some would argue that these symptoms may simply be a person's natural personality and not symptomatic of a true mental illness.

What do these personality variables mean?  Here's a brief definition from Dr. Paulhus's original work
  • Machiavellianism:
    • "the manipulative personality"
    • cold
    • manipulative
  • subclinical or ‘normal’ narcissism
    • grandiosity
    • entitlement
    • dominance
    • superiority
  • subclinical psychopathy
    • high impulsivity
    • thrill-seeking
    • low empathy
    • anxiety
    • anti-social behavior
Since the 2002 paper by Dr. Paulhus, the buzzword "Dark Triad" has been embraced as research shorthand for these undesirable personality variables.  Dr. Natalène Séjourné, a psychology professor and researcher at the Université de Toulouse (University of Toulouse), France in a 2009 paper suggested the sadism be added to the so-called "Dark Triad".  She and her colleagues wrote:

As in the studies by Paulhus and Williams’ study (2002) and Vaughn et al. (2008), psychopathic, narcissistic, and Machiavellian traits were moderately correlated. Furthermore, they correlated moderately with sadistic traits in our study. This suggests that all four of these constructs are overlapping but distinct. We propose calling the association of psychopathic, narcissistic, Machiavellian and sadistic traits the ‘Dark Tetrad’ of personality traits.

50 shades handcuffed
Recent work ties sadism to the Dark Triad. [Image Source: After-Set]

The term "Dark Tetrad" has also gained growing acceptance as subclinical sadism has been confirmed to be correlated to the members of the Dark Triad.

II. Send in the Trolls

The new work -- published in the peer-reviewed psychology journal Personality and Individual Differences -- examines how so-called internet "trolls" -- as well as other internet denizens -- rank when it comes to the Dark Tetrad.

Dr. Paulhus is the paper's senior author, while Erin E. Buckels -- a graduate researcher at the University of Manitoba -- gets the first author credit.  Also on the paper is Dr. Paul Trapnell, a personality researcher at the University of Winnipeg (Winnipeg is the capital of Manitoba, Canada).

Professor Del Roy, et al.
Professor Delroy Paulhus (right, front) is seen here with his lab in 2009.

The authors first surveyed 418 people (42 percent female) via an online survey target only U.S. users.  The survey participants were paid 50 cents via Amazon.com, Inc.'s (AMZN) Mechanical Turk (MTurk) service, which matches up people willing to answer surveys for pay to those looking to ask questions.  The first survey gave people a basic personality test to determine whether they showed strong signs of any of the Dark Tetrad traits, then they were asked to pick their favorite activity on sites from: non-commenter, chatting, debating, trolling, making friends/other.

Using the first personality test, the study looked at correlation among various preferred activities and the four traits of the Dark Tetrad, dividing subclinical sadism into two categories -- "vicarious sadism" (those who enjoy observing others suffer or experience discomfort at the hands of others) and "direct sadism" (those who looks to personally make others suffer or experience discomfort).

Mechanical Turk
Both surveys use Amazon's Mechanical Turk.

A second survey used 609 American participants via MTurk and 188 Canadian psychology student.  One key difference in the second study was that it differentiated direct sadism into two categories -- verbal and physical.  The study also allowed users to rank how much they like each of the aforementioned activities (e.g. lurking on a site without commenting, chatting, etc.) versus the first study which only allowed a single "favorite" response.

III. What a Troll Wants, What a Troll Needs

Unsurprisingly, both studies found that older users spent less time commenting on the internet, generally than young folks.  Men spent, on average 52 minutes a day commenting on the internet, while women spent only 29 minutes a day commenting, on average.

One good piece of news is that the study suggests trolls are in the minority on the internet, as noisy as they are.  In the first study, only about 1 in 20 individuals (5.6 percent) identified trolling others as their preferred internet activitity.  The most popular active activities were debating issues (23.8 percent) and chatting with other users (21.3 percent) -- each of which appealed most to roughly 1 in 5 users.  And of course the very most popular activity of all was passive -- internet lurking.  Roughly 2 in 5 users (41.3 percent) reported not commenting on internet articles and forums.

Trolling v. Dark Tetrad
Trolls show the highest rates of sadism, much higher than those who prefer other activities. [Image Source: Personality and Individual Differences]

Generally non-commenters and those who used the net for "other" reasons (e.g. friendship) had the lowest rates of negative personality traits, but they were more predisposed to narcissism.

The most common personality flaw among the chatty Cathys (those who preferred chatting) was vicarious sadism, but perhaps predictably those who debate issues saw much higher scores for most of the personality flaws.  While their vicarious sadism correlation was roughly the same as the chatters, their most common Dark Tetrad trait was Machiavellianism.  Both groups saw relatively low rates of narcissism.

When it comes to narcissism of commenters versus non-commenters, non-commenters with a correlation to at least one Dark Tetrad trait were more likely to be narcissitic than the chatters/debaters with at least one of the negative traits.  However, when you include those with no correlation to any of the Dark Tetrad traits, you see more narcissistic chatters/debaters than lurkers.  So you could make a compelling argument that either group is the bigger egotist.

Troll Doll
[Image Source: Word of the Nerd]

Turning finally to the trolls, they were by far the most out of control, showing dramatically higher rates of sadism and the rest of the Dark Tetrad (even narcissism).  Interestingly, they narrowly preferred watching others flame each other (vicarious sadism) than engaging in flaming themself (direct sadism).  In other words, trolls love to see trolling even more than they like to do it themselves (although they also like to do it themselves).

IV. Pleasure-Seeking Sadists

The second study found some additional information via first looking at the user's ranking of how they felt about trolling and then comparing this with survey quesitions which would seemingly identify a troll -- a so-called Global Assessment of Internet Trolling (GAIT).

Most internet narcisissist are actually non-trolls.  Of all the Dark Tetrad, it was the weakest correlated (-0.09 w/ those who like trolling on the enjoying rating scale; 0.18 w/ GAIT).  By contrast, subclinical sadism (0.52 w/ enjoying rating 0.68 w/ GAIT) followed by psychopathy (0.38 w/ enjoying rating, but 0.55 w/ GAIT) were the strongest correlations.  The correlation between Machavellianism and trolling was also rather weak (0.37 w/ enjoying rating; 0.34  w/ GAIT).
Troll typing

Perhaps most interestingly the enjoyment ratings showed that those who enjoy trolling somewhat, but also enjoy other activities (e.g. chatting or debating) were skewing those groups to appear more strongly correlated to the Dark Tetrad than they really were.  And when it came to the Dark Tetrad the correlation to other personality flaws -- psychopathy, Machavellianism, and narcissism -- appeared to be driven by the fact that people with sadism tend to be some of those other things.  However, analysis indicated that trolling only appeared to be an enjoyable outlet for one Dark Tetrad trait -- sadism.

It seemed that trolls most enjoyed direct physical sadism, followed by direct verbal sadism.  Performing other forms of analysis, researchers concluded that those who identify as trolls appear to do so largely because it offers them a pleasurable outlet for the sadistic side -- a premise that led to the paper's title -- "Trolls Just Want to Have Fun".

The authors write:

In fact, the associations between sadism and GAIT scores were so strong that it might be said that online trolls are prototypical everyday sadists.... Our research suggests that, for those with sadistic personalities, that ideal self may be a villain of chaos and mayhem – the online Trickster we fear, envy, and love to hate: the cyber-troll.

They add that when you remove those who enjoy a mix of trolling and other activities, virtually none of the people involved in other activities were inclined to sadism.  In other words, the study indicates that if you're a troll you're probably a mild sadist (at least) and if you're a mild sadist (at least) you'd probably become a troll, if given the opportunity.

The Sadist
[Image Source: VHS Movies]

On the other hand, the studies indicate that while those enjoy trolling tend to be predisposed to manipulative behavior (Machavellianism) and subclinical psychopathy, they don't really enjoy doing those things when trolling.  Trolls appear disinterested in manipulating their marks (Machavellianism), winning their adoration (narcissism), or creating chaos for chaos's sake (psychopathy).  They troll to express one Dark Tetrad trait and one alone -- sadism, the desire to hurt others and/or make them uncomfortable.

V. Other Studies, Combined With This Work Suggest Trolls May be Breeding at an Alarming Rate

The paper said that one major question left unresolved was whether the internet was making people score higher on the Dark Tetrad (and hence more anti-social) or whether it was simply letting people be themselves -- including those inclined to sadism (the trolls) or other forms of antisocial behavior.

The good news if you're a troll, and the bad news for the generally internet audience who suffers their presence is that it turns out the Dark Tetrad isn't wholly negative.  Other studies -- including a 2009 study by Dr. Peter K. Jonason (was then at New Mexico State University) and Dr. David P. Schmitt (Bradley University) shows that those with the Dark Triad of traits tend to be more likely to breed.  

A 2006 analysis by Victor Nell suggested similar advantages might be confered upon the sadist, writing:

Cruelty has fitness beneits in solving problems of survival and reproduction in forager, pastoral, and urban societies.

Troll baby
Trolls' sadistic cruelty may be genetics: studies suggest cruelty and sadism improve reproduction chances. And that means more baby trolls. [Image Source: NeithTK/DeviantArt]

In other words, trolls will survive, as long as they know how to troll that's how they will survive, and they got all their life to live, and all their love to give (and will likely get more opportunities to give it).

While they may be less likely to be able to form stable relationships, trolls' potential advantage in achieving temporary, permiscuous one night stands leads to a startling potential conclusion -- the trolls are breeding, and may be favored over non-trolls by natural selection.

Source: Personality and Individual Differences



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

RE: "trolling" -- can be a subjective call
By Motoman on 6/21/2014 11:47:35 AM , Rating: 3
I disagree. I think if you were able to study Tony in actuality, you'd probably find he ranks pretty high on the 3 criteria noted at the beginning of the article.

And while it's not a "true" measure of virtue, the rating system on a forums like this I think lets the general population self-determine who the trolls are. On DT, anyone with a rating notably under 2 is almost without a doubt a troll.

To be sure, the rating system is really more of a popularity contest than it is a measure of how "right" you necessarily are - because people as a group frequently are wrong about a given topic anyway, or lash out at people who are presenting a *true* viewpoint that they just don't want to listen to. But in the long run, I think it probably serves a pretty good role as a proxy for identifying trolls.


RE: "trolling" -- can be a subjective call
By Reclaimer77 on 6/21/14, Rating: -1
RE: "trolling" -- can be a subjective call
By bug77 on 6/23/2014 3:47:27 AM , Rating: 2
Well, the study says that trolls are a minority, so they wouldn't change the ratings significantly. But yes, I can see why that hurts (Reclaimer77 has posted a total of 17201 comments at DailyTech, the average comment rating was 1.59).


RE: "trolling" -- can be a subjective call
By Reclaimer77 on 6/23/14, Rating: 0
RE: "trolling" -- can be a subjective call
By bug77 on 6/23/14, Rating: 0
RE: "trolling" -- can be a subjective call
By Reclaimer77 on 6/23/2014 10:05:34 AM , Rating: 2
I seriously doubt there is any correlation between this study and Daily Tech's massively flawed rating system. That was my point.

Also my post had NOTHING at all to do with this study or it's findings. Where do you get that? I was disputing Moto's arbitrary statement that ratings 2 and higher are not trolls, and those below are.

Tony, who is none of these things, gets rated into oblivion simply because his stance is unpopular. In a different venue, a pro-Apple forum, his posts would be rated UP highly. So which rating system is correct? Who decides who the "trolls" are?

quote:
This one places you squarely in the area of the trolls


Coming from you that doesn't mean a whole lot. You can be quite the "troll" yourself :)


RE: "trolling" -- can be a subjective call
By bug77 on 6/23/2014 10:45:36 AM , Rating: 1
quote:
You can be quite the "troll" yourself

Of course I can, anyone can. Trolling requires no special skills/training. But my average is still above 2 :D


By Spuke on 6/23/2014 12:08:01 PM , Rating: 2
quote:
But my average is still above 2 :D
LOL! Bug made a funny.


RE: "trolling" -- can be a subjective call
By retrospooty on 6/21/2014 3:40:56 PM , Rating: 2
"To be sure, the rating system is really more of a popularity contest than it is a measure of how "right" you necessarily are"

That is true, and why it isnt related to trolling, mostly popular opinion, and sometimes demeanor.


RE: "trolling" -- can be a subjective call
By Motoman on 6/22/2014 11:27:05 AM , Rating: 1
It's related to trolling because people hate trolls, and they punish them.

There's only one way to punish a troll. You down vote them.

It is, in fact, a function of popular opinion to identify the trolls. If you don't believe me, go and look at the scores of people you would identify as trolls. The correlation is there.


RE: "trolling" -- can be a subjective call
By retrospooty on 6/22/2014 12:36:23 PM , Rating: 3
Yes, they do down vote trolls, but they also downvote anything they disagree with, so people with the constant -1 might be trollish, but also might just have an unpopular opinion. It's not an automatic system.

For example, someone might come on and say "I actually really like Blackberry OS10 and I prefer it to IOS, Android, and WP" That in itself isn't trolling, but they would be voted down to -1 right away because everyone think they are a moron, even though they were polite and not trolling.


RE: "trolling" -- can be a subjective call
By Motoman on 6/22/2014 3:43:35 PM , Rating: 2
But just having an unpopular opinion (which by definition is in relation to a given topic) won't get you permanently relegated to troll status (well under 2.0 over all). You might get wasted on that specific topic, but overall if you're not otherwise a troll and/or abject moron, your total will be fine.

Your example about someone preferring BB to anything else is a perfect example of what I just described - sure, that user will probably get downrated on that thread like crazy. But if he's otherwise a normal person, his total isn't going to be hit much.

To have a total that's significantly under 2.0, and assuming more than a handful of posts, you have to essentially *always* be downrated. It's actually pretty hard to maintain a total that's much less than 2.0 - because then even if you just post a normal comment that stays at 2, you're raising your own average.

You have to be well and truly wacko to get much under 2 and stay there.


RE: "trolling" -- can be a subjective call
By Cheesew1z69 on 6/23/2014 10:14:00 AM , Rating: 2
quote:
You have to be well and truly wacko to get much under 2 and stay there.
So your saying that someone with say, 1.87 overall is a troll? I think you need to re-evaluate that number 2 because the majority of the people on this site are under 2 and by your definition, they are trolls. Not quite.


RE: "trolling" -- can be a subjective call
By Motoman on 6/23/2014 10:51:29 AM , Rating: 2
...says the guy with a 1.87 over 3470 comments.

So. Yeah. That means that you're *very* consistently getting rated down. That can't be just because of your opinion on one or two topics. It's endemic.

I'm not sure exactly what the "cutoff" is for true trollness...maybe it should be 1.80 for at least 1,000 comments or something.

You are, if nothing else, proving the point I made above...in order to maintain an average under 2.0 for significant numbers of posts, you have to get downrated *a lot.* And if nothing else, you get downrated *a lot.*


RE: "trolling" -- can be a subjective call
By Reclaimer77 on 6/23/2014 11:32:52 AM , Rating: 2
Says the guy with only a 2.34 rating over 6k comments.

If you're so high and mighty, you should have at least a 2.5-3 rating. This means you're very consistently NOT getting rated up, troll.

See how silly this can get? Who decided that 2 was the cutoff anyway? You did! Well that's not arbitrary at all...

quote:
in order to maintain an average under 2.0 for significant numbers of posts, you have to get downrated *a lot.*


Nobody is disputing that, duh. However there's more to being downrated than whether or not you're a troll. Which is the entire point!

Until you realize the rating system here is more of a blunt instrument, clearly abused, than an honest evaluation of the subject matter - your point is moot.


By MrBlastman on 6/23/2014 1:44:57 PM , Rating: 2
Very few people here have a rating above 2.5. Most that do generally comment facts, avoid "hot button" discussions and are generally, from my observations, quite intelligent and helpful.

They are rare. You don't see them often. I could probably name them on one hand. But that doesn't mean the people with a rating below 2.5 aren't intelligent, either. It just means they take risks or unpopular positions on a topic.

The rest of us that enjoy debate--I know I do, we are willing to engage in uncomfortable positions and take a risk to defend what we believe in. From my observations, more times than not, downrates are inevitable and typically revolve around who or who not might be reading a thread and the overall consensus view at the time be it more liberals or conservatives, to pick an example, commenting in the thread.

I'm quite sure, as you suspect, that alt accounts are brought out in force on occasion just to smash a user into the ground.


RE: "trolling" -- can be a subjective call
By Spuke on 6/23/2014 12:10:35 PM , Rating: 2
quote:
You have to be well and truly wacko to get much under 2 and stay there.
So you really don't think that unpopular opinions will get you downrated here?


By Cheesew1z69 on 6/23/2014 12:18:00 PM , Rating: 2
Unfortunately, that is what it sounds like he is claiming. Except, we all know better.


By Motoman on 6/23/2014 12:41:53 PM , Rating: 2
Of course they will. But consider the math...it's pretty simple.

If your posts, on average, don't get uprated or downrated, they stay at 2. That's the average - that's "normal" based simply on how DT set up their system. Someone with, say, 1,000 posts and a 2.0 rating is utterly average. Middle-of-the-road. Normal. That's your baseline.

Now, if that person expresses an unpopular opinion on a given topic, of course they're going to get downrated...on that topic. In the grand scheme of things, that's not going to have any real affect on his overall rating though. The rest of his average posts will accomodate that.

The point that has been made is that to get to significantly less than 2.0, at a significant number of posts (like 1,000, just for the sake of argument), you have to be getting downrated *all the time.* On essentially every topic - every post you make.

You don't get to be significantly under 2.0 because of one or two unpopular opinions. You get there because you're constantly getting downrated - on everything. And it is, well and truly, hard to get down there and stay down there. Because if your current average is less than 2.0, every "normal" post you make actually increases your overall average. For a troll (or at least, a very unpopular person) to make a post that simply *doesn't* get downrated is to improve their own average.

The inverse is true for someone with a very good rating. If your rating is significantly over 2.0, and you make a post that isn't uprated, you're actually decreasing your overall average. For this person, making a post that doesn't get uprated is actually detrimental to their score.

Hence, it is plain to see that it is actually relatively difficult to deviate significantly from the baseline of 2.0. Because trolls are constantly improving their average by making "normal" posts, and net-positive posters are constantly decreasing their average by making "normal" posts. The normal posts constantly pull you back towards the 2.0 baseline.

Which is why my OP bears out - the voting system on DT is very likely an excellent proxy for the members of this site to identify the trolls. Because of the simple fact that to have an average significantly under 2.0 is actually pretty hard. It's not indicative of one or two unpopular opinions - it's indicative of being endemically disliked by the overall DT membership.


"I'm an Internet expert too. It's all right to wire the industrial zone only, but there are many problems if other regions of the North are wired." -- North Korean Supreme Commander Kim Jong-il














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki