Decoding Academia 30: Sadistic Trolls love Dark Humour *Preview*
This is a preview episode to remind those who might be interested that we have a bonus Decoding Academia series, available at the Patreon at the Revolutionary Genius tier and above, which is now up to episode 30! On Decoding Academia we usually focus on specific papers and indulge our more nerdy and academic tendencies. It is almost like a journal club of two!In this episode, we take a look at a study exploring the connections between Dark Humour, online trolling, and Dark personality tendencies. This is right up the alley for two brooding Decoders with twisted dark mentalities. Expect shocking personality quizzes, dad jokes & dank memes, Bayesian sidetracking, and an inception-level discussion of alleged regressions. This is one for all the family!Paper examined: Voisey, S., & Heintz, S. (2024). Do Dark Humour Users Have Dark Tendencies? Relationships between Dark Humour, the Dark Tetrad, and Online Trolling. Behavioral Sciences, 14(6), 493.
Hello and welcome to Decoding Academia, a sub-production of Decoding the Gurus, with me, the anthropologist Christopher Kavanagh, and him, the psychologist Matthew Bryan.
Class is in session.
How was that?
Was that pretty professional?
That's good.
Class.
I've never said classes in session.
It sounds bad.
Imagine starting a lecture with that.
Class is now in session.
Yeah.
It sounds like...
Be reported.
Yeah, it sounds like an action hero's sort of little lines, you know?
Class dismissed.
Oh, yeah, after you killed someone at school.
Class dismissed.
Class dismissed.
Yeah, you know, decoding academia, part of the suite.
Of Dakota Gurus products.
High quality products.
We have a product for everything.
You don't need to listen to any other podcasts.
We've got you covered.
Don't listen to history podcasts, science podcasts.
You don't need to get updates on politics and stuff.
Dakota Gurus is all you need.
Get rid of your friends.
Get rid of your associates, your family.
Get those guys out of there.
Entertainment products, you don't need them.
They're all your one-stop shop.
For very niche internet content.
That's what we are.
We are here to please.
And we are here today.
What you've just heard is some examples of dark humor.
Humor with a dark edge to it.
And that is the topic of today's episode, Matt.
That's what we're looking at.
That's us.
Yeah, this is suitable to us.
We're dark people.
We're edgy.
We're the bad boys of the internet.
Yeah.
People often say, how dark you are, how brooding, how mysterious.
People ask me what I'm rebelling against.
I say, what have you got?
What have you got?
What have you got?
So this is an article which was recommended on the Patreon.
So see, we also interact with our audience.
We listen.
It's a two-way street.
It's Parasocial City.
And this is by Sophie Voise and Sonia Hanks.
School of Psychology at the University of Plymouth.
It's in the UK, Matt.
Plymouth Brethren.
That's a religious cult kind of thing.
Is it?
The Plymouth Brethren?
Are they in Plymouth UK or Plymouth?
I don't know what the connection is, but I just know a little bit of a severe religious group.
So just be careful.
I'm sure there's no relation to these resources.
Plymouth?
Well, the researchers, yes.
Yes, the researchers know.
Again, dark humor.
Dark humor.
The title is Do Dark Humor Users?
Have dark tendencies.
Relationships between dark humor, the dark tetrad, and online trolling.
And this is in a journal called Behavioral Sciences, but by a outlet called MDPI.
Now, Matt, do you have any thoughts about MDPI articles just before we begin?
My thought is that I've published at least once in an MDPI journal.
But, you know, we've all done things we're not proud of from time to time.
Yeah, look, MDPI, we're not here to besmirch people's good name, but I'll just say it has a little bit of a reputation like Frontiers in that I don't think it's purely like a predatory play-to-play journal,
but it's a little bit that The standards tend to be lower.
So sometimes when you see an article and it's MDPI, you think, ah, so they couldn't get that published or they were tricked into publishing it here.
That's the general feeling I have.
But it doesn't mean the studies are going to be bad.
I've seen good studies on MDPI, but I've also seen bad studies.
So just a note.
And there's a whole bunch of different journals.
It is a broad publishing...
Group, I guess, is the way to put it.
But it's on the fringe of predatory journals, I feel.
Yeah.
Look, there's a huge grey area, I think, between full-on predatory, pay-to-play type journals that will just literally publish anything.
Oh, that's a good point, by the way.
For those that aren't in academia, predatory journals are these journals which try to get you to publish in them, but you end up having to pay them.
And they will publish anything without quality control, basically.
So people have done various things where they've had their cat publish an article or just copy and pasted things over and over and so on.
So, yeah, they don't have a good reputation.
But we are not saying that's what MDPI is.
We're just saying there's a spectrum between top-tier journals with very good reputations, deserved ones, not the ones which just have prestige attached to them because of their age.
Like pure predatory journals.
There's a wide spectrum.
A controversial point of view, Chris, perhaps, but I think all journals are to some degree predatory.
This is not controversial in academia at all.
It's like one of the main habits of academics to complain about journals and their unpaid labor doing reviews for them.
So, yeah, you're just mainstream, man.
That's not very obvious.
Well, well.
Look, I also think it's wrong to call it predatory because they're not always taking advantage of the researchers.
The researchers or the academics could be, you know, they just want to have lines on their vitae.
They're very happy to pay $3,000.
They get a DOI and it looks like a citation.
And, you know, people on hiring committees and stuff are so lazy they don't even check the articles.
They just look at the citation list and go, well, look at that.
It's quite long.
So we'll give them the job.
All right.
Fair enough, fair enough.
It's all facets of the system.
Problems with the system, man.
But that's okay.
That is all beside the point.
What's this article about, Matt?
Tell me.
What's it about?
Okay, well, it's about this, you know, thing people have heard about before, the dark tetrad.
I used to always call it the dark triad, Chris.
I assume it's the same thing, right?
What?
The dark triad.
How well did you read this paper?
Of course, the dark triad is the...
The original three, Machiavellianism, narcissism, and Frank, what's the third one?
But they've added, anyway, they added in the fourth, sadism.
Psychopathy.
Psychopathy is the third one.
So when you add in the fourth one, it becomes a tetrad.
So, yeah.
As they outline in the paper, Matt.
Might have missed that bit.
Don't look at me like that.
I looked at the supplementary materials.
I did my bit.
Okay, so what's it about?
So, you know, it's a correlation analysis.
Look, before I explain it, why don't I just read the abstract quickly.
Humor and antisocial behavior on the internet are under-researched.
Online spaces have opened a gateway for new ways to express unrestrained humour, e.g.
dark humour, and ways to behave antisocially, e.g.
online trolling.
The tendencies and motivations of those engaging with such humour and behaviour are yet to be clearly established and understood.
That's right.
This present study aimed to fill this gap by exploring the interplay between, by that they mean calculate correlations, between dark humor, online trolling and dark personality traits.
Participants, N equals 160, take note Chris, completed an online survey consisting of trait scales to assess the dark tetrad, dark humor and online trolling, as well as two online trolling tasks, enjoyment and ability, and two dark humor meme tasks.
Enjoyment and ability.
The results confirmed relationships between the dark tetrad and the dark humor trait and several dark tetrad traits were related to the enjoyment of and ability to produce dark humor.
Furthermore, dark humor and online trolling were closely related.
The findings also revealed that online trolls did not enjoy being trolled but did enjoy trolling and this ability to troll is underpinned by sadism.
Sadism.
I know people are getting in trouble if I say things wrong.
These findings illustrate the potential dark psychological motivations for using dark humor, demonstrate that online trolling is infused with darker forms of humor, and provide deeper insights into online trolls.
So there you have it, Chris.
That's what it is about.
That's it.
So when you said correlational study, isn't there also a regression or two in here somewhere?
Oh yeah, big difference.
Yeah, well, look, we should mention again, don't assume that our audience are born academics like us.
So this kind of study, kind of the stalwart of the psychology and social science field where people take measures and take another...
set of measures and then look at how they're related and say, this is another research area.
So look, we find this relationship.
That is the bread and butter.
I've done it.
You've done it.
We've all done it.
We've all done it.
And we love it.
Look, Chris, I'd even be more specific than that.
A self-report questionnaire.
Oh, yeah.
Yeah, self-report measures.
This actually does include some things which are not just purely...
You know, self-reports.
There's some little activities, I suppose, in there.
And yeah, you're correlating some measures with some other measures in the same instrument, measured in the same instrument.
And, you know, you often do a regression as well.
Regressions help you look at the relationships when you are controlling for other relationships, like for multiple...
Relationships simultaneously, right?
This is how they're often used.
There are other ways, Matt, yes, before you say there are other things they do, but the way people often do it is like that, right?
I wasn't going to say anything.
You looked like you were going to correct me, so I had to cut you off before you got there.
But, yeah, so the other thing to note, as you hinted at there, and without even looking at other details, this is just something for people to bear in mind.
When you see an N equals 160 study, and in this case, it's going to be a sample from a specific country, right?
So what people mean when they say we've investigated this is we have looked at one small sample in one country and we find these relationships, right?
Now, maybe it will be the same in every other country.
Maybe it would happen if you did a much larger sample and stuff, but...
When you're creating quality of evidence, this would, before finding anything else out, even about highly measured things, how to go, you would automatically say, well, that is a relatively small sample.
So my conclusions that I'll draw will be equally relatively measured because I obviously wouldn't want to say that this relationship will hold throughout the whole world because that would be...
But this is normal.
And I'm not thanking them for that.
I just mean, internally, this is what I do.
Whenever somebody says, you know, I looked at the relationship between depression and social media usage, and it's like an N equals 80 study in the US.
I'm like, okay, you did look at that topic.
But this is very constrained to a particular context and a particular set of measures and whatnot.
So just bearing that in mind.
Just saying.
Well, look, why don't we approach this?
Why don't we jump around?
Like, I don't think we need to work through this.
No, no, no.
From the beginning.
Let's jump around.
So why don't we jump straight to that?
The participants were...
There's something weird in this.
I wonder if you noticed that, but go ahead.
Well, these are University of Plymouth students recruited through the participation pool and then given course credit for their participation.
So this is very much the classic.
Psychology department approach to getting students, which is you recruit your students.
Sorry, did I say recruiting students?
Recruiting participants, which is that you recruit students.
Yes, yes, yes, Kavanaugh at the back.
You're at the back.
Well, just may I note as well, Matt, 132 women.
I was going to get to that.
I was going to get to that.
Oh, sorry.
Sorry, I thought you were going to move on to the next topic.
That's fine.
I've got another point about participants, so before you move on.
No, that's a good point.
A lot of women, 132 and only 30 men, three other.
So that's 165.
That's because, you know why, Mark?
That's because psychology in general now is dominated by women in terms of like undergraduate students.
It's the same in Japan.
A lot more women studying psychology than men.
Yep.
The times, maybe they're changing.
Or maybe that was always the case.
I'm not sure.
It's been like that for a while.
Yeah, yeah.
So, you know, this is another little ingredient to pop into that little equation.
You've got sort of sub-government running in the back of your head, which is, yes, it's an association study.
It's essentially correlations between things that are measured in the same instrument.
The sample size is low.
There is a single sample.
And the sample is not a...
It's not a representative sample.
It's not a community sample.
It's not an online sample.
Convenient sample.
It's a convenient sample of, you know, the particular kind of people who tend to enroll in a psychology degree.
Yes.
So my one other point, Matt, was did you notice that there was one participant who was not a student?
No, I didn't see that.
I was a little bit...
Oh, yeah.
Other participants were recruited through an online forum.
One participant.
And did not receive it.
Yeah, so who was that?
There was only one?
Like, did nobody take part?
I was just like, it's a funny thing to include.
Most people would probably just have dropped it.
It was like a recruitment attempt there, which clearly was not successful.
But, you know, all credit to them.
I mean, it's just as good to include them as...
Not to, I suppose.
Yeah, you had to add an extra sentence for it.
Yeah, so those are some issues.
That's details about participants.
So we know you're talking about a specific sample.
And they go through things which I like, actually, Matt.
They do a bunch of quality control.
They remove people who always select the same option across multiple scales.
If people complete the study too quickly, they have standard attention.
I don't like them so much, but nonetheless, they don't drop out that many people from it.
Five participants from this.
And they also, shall I say, Matt, that they pre-registered their study.
So you can go and look at the data, you can look at their hypotheses, and you can see the study materials all up on the OSF, the Open Science Foundation, which I approve and applaud them from.
And that makes me slightly happier.
There's this thing that's in sample size that non-social science people might not know about power analyses.
And this is basically people saying, when they do a power analysis, what is the sample size they need to detect a certain size of effect?
And these are often, they should be informative, but they're often not really informative because people only report the needed sample for like one test.
And people are often running multiple tests and the minimum sample size is often woefully inadequate.
So it is good.
It is important that people include it.
But it is something that a skilled researcher can kind of pass their sample with a justification that may or may not get flagged up.
So they said they only need 67 participants to...
Detect the effects that they want.
And that seems rather low for me, but I do believe G-Power will have spat out that number.
Yeah.
I'm not a fan of Power tests, except in very specific circumstances.
Daniel Likens will have to talk to you.
He is a fan.
Is he?
Oh, God.
That's got me worried.
But like in situations like this, they always feel a little bit arbitrary to me, like a medium effect, right?
A medium effect is defined in terms of these like standardized correlations, you know, like between variables.
And so it's like 0.3, right?
So it's basically a 0.3 correlation when you've standardized the variables.
And it's just, I don't know, with 80% power and 5% error and you do get numbers like they've reported here.
Getting at least 153 participants would detect medium effects with 80% power and 5% error.
But I don't know.
I don't know.
Yeah, but so the thing is, I understand, Matt, and this is the realm for us to engage in academic minutiae chat.
But I think the issue was that people weren't considering power at all.
They were doing sample sizes that were woefully inadequate and they just didn't mention power.
So journals...
Because of the replication crisis and whatnot, came to put an emphasis on you have to justify your sample size, you have to provide a power calculation.
And now they employ that rule too stringently, like they don't consider if it's appropriate, as you and I have experienced in certain cases, where we have an extremely overpowered study, but they want a very specific statement about power that fits this format.
Journals can do it in a very robotic way, but the basic idea I think is good because I think what it's intended to do is stop people doing N equals 30 studies and not noticing that they're way underpowered to detect the effect size that they're anticipating.
So it's employed wrong, but the improvement, I think, is in the right direction.
Can we say that?
Okay.
Yeah, yeah, fine.
Yeah, I mean, there are so many things that are not tested by a power analysis, like restriction of range, for instance.
So when you get a sample of psychology students, right, then you're probably not going to have like, you know, a large number of very statistic people, for instance, because compared to...
Even the general population, but certainly compared to other subpopulations, psychology students, especially ones that are comprised of 132 women and 30 men, probably not very large.
So that's the kind of thing.
So restriction of range is going to suppress all kinds of things.
Now, if you took economic students, you would have like a huge problem with statistics.
Medical students.
Yeah, medical students.
Dentists.
Anyway, there are other issues.
But these are the...
Look, this is what this is for, Matt.
You're highlighting to people when you read a paper.
It's not enough to just be like, oh, that's that, right?
These are all the little calculations and comments that go into your mental calculus about like, okay, so the paper is like this and like that.
And it is nitpicky.
And it is like a thing.
But it is what...
Academics are taught to do and what you actually should do.
Yeah, yeah.
That's what we're doing here, by the way.
We're not...
See, I'm kind of worried that, like, one of these days...
Oh, these people will come across...
The authors might listen and go, well, this is just mean.
And I'm not being mean at all because these aren't really criticisms.
This is more just sharing the kind of...
Thought processes that go through your head when you look at any paper.
My own stuff that I've published has been guilty of all of the things that I'm pointing out as well.
Exactly.
But this is like the neurotic madness of academics.
In the meme podcast, we're talking about what gurus do and stuff.
This is an insight into the madness of academics where they're like, oh, why was there one podcast event that was recruited?
That's a bit odd.
And this is why in some cases when you see the gurus and the referencing papers and whatever, they tend to never ever do this.
They only pick out details that allow them to dismiss a study that they don't want or something like that.
Or something flashy that supports the thing that I was saying in the first place.
I liked the term you used.
Was it decorative?
Decorative citations?
Cosplay?
No, no, you're talking about, you know, the TED Talkie guys, you know, and you described them citing their research in that article, for instance, as being like just sort of decorative science.
Yeah, flourishes.
Flourishes, yeah.
So that's what they do.
We're just sharing what an actual academic does when they're sort of, when they're just kind of weighing what the study is telling you.
Because this is a study which has found a bunch of significant effects and which is interesting and good.
You just have to factor in a whole bunch of stuff in terms of what you then conclude from that.
I think it's probably good to get to the measures next, do you think, Chris?
Yeah, measures or...
Yeah, I guess.
I was going to say you can talk about the results because you'll have to talk about the measures to explain the results anyway.
But, you know, if you were doing it properly, you would talk about the measures.
Yeah, for me, the measures are really important, right?
Because you can't, like...
You can't understand a correlation until you understand what it is that has been measured.
And a lot of the time, you know, psychologists, we always give labels to things that we've measured, like sadism or...
Another pathology that you have.
Yes.
Clinical psychology is the worst for this.
Like, you know, like dark humor and online trolling, right?
So what you tend to do is you mentally go, oh, right, so we've found that online trolling...
Is associated with sadism or something like that.
And we forget that actually what we're talking about is that those names are just what we've given to the score that gets calculated by adding up by what people ticked off on these liquid items.
And really all it is is the item.
So you need to look at the measures and see what it is that you've actually asked people.
And it's often related to the thing that we've called it, like sadism or whatever.
But it's not necessarily the thing, right?
Yeah, this is very important.
I think this is a point that we reiterate almost constantly, but it is so significant.
And it's a mistake that often comes up.
There is the concept that you're interested in, and then there is how you chose to measure that.
And the relationship between the two things can be one.
Almost one-to-one.
It can be very precise, like height.
So you took the height in centimeters.
That's how you measured it.
Very good measure of height, I would think.
But when it comes to these psychological things, like online trolling or dark humor, your first question would be, how the F do you measure dark humor, right?
Or this kind of thing.
And that's a good question.
And that's why Method Section exists, because you can then go and have a little look.
So which measures did you want to talk about, Matt?
All of them.
All of them.
All of them?
There's a lot of them.
Good luck.
Well, you know what?
Yeah, you're right.
There's too many to talk about all of them.
Why don't we start with the strongest correlations that they've found and let's look at the measures that went into that.
Oh, so you're doing what I said.
You're going to the results and you're talking about measures.
Okay, fine.
I flip-flopped.
Yeah, that's all right.
Yeah, so what would you like to talk about, Pat?
What correlation?
We can jump around.
We said we're going to jump around.
We're at the results and method section now.
Yeah, yeah, we're jumping around.
Okay, so one of the first things they report here is some correlations between the dark tetrad traits and dark humor measures.
This is in table two.
And they found some pretty big correlations there, Chris, you know, by the standards of my discipline, between Machiavellianism and the dark humor trait of correlation of 0.53.
Very significant at.001, despite there being a relatively small sample size.
And we have a correlation of.56 between sadism and this dark humor trait.
So there you go.
Machiavellianism, this tendency to kind of, you know, a willingness to manipulate people, to get what you want.
This is the concepts we're talking about here.
And this dark humor trait.
Remember...
We're not quite clear about what that document trait is.
I mean, you would be if we'd gone through this properly and read the introduction section for you, but we didn't.
We're jumping around.
So just put a pin in that for the moment.
We're going to look at the measure.
And sadism.
So what does this mean?
Does it mean that they torture small animals, that they're a serial killer or something?
Well, we'll find out because it's all about what questions they ask these participants in this self-report measure.
And just remember, when you think about self-reports, This is what people are willing to tell you.
And remember that these are psychology students participating for course credit.
And think about how...
No, just think about yourself in that situation.
Perhaps, you know, how comfortable you might feel admitting to things.
Think about social desirability and so on.
There's going to be some stuff going on there, perhaps.
But first, let's look at those measures, eh?
Yeah, I like this because...
So many of them describe me, Matt.
So, like, if you're looking at sadism, the self-report measures are stuff like watching a fistfight excites me.
I really enjoy violent films and video games.
It's funny when idiots fall flat on their face.
That is funny.
You would strongly agree to all of those, wouldn't you?
I enjoy watching violent sports.
Some people deserve to suffer.
Just for...
The problem is...
Just for kicks, I've said mean things on social media.
Do you deny that, Chris?
I know how to hurt people with words alone.
You do.
I swear to God.
So you would answer five, just to be clear, on all of those items.
Let's be honest.
Yeah, that's possibly true.
And they have a one to five.
So, yes, but is that really measuring?
We'll get to that.
We'll get to that later.
Just for the moment, let's just admit that you've maxed out their sadism scale.
Whether or not you are a sadist, we'll get to that later, Chris.
Because you do like mixed martial arts.
You like participating in it.
I do like idiots falling flat.
Like, metaphorically.
Metaphorically.
Yeah, it's very clear.
It's idiots, right?
Like, it's not...
So, I'm assuming, like, bad idiots.
Not just, like, people that are not intelligent.
People who deserve it.
So, yeah.
Yeah, so that's, like, the kind of self-report.
And they have the same thing.
Other items for Machiavellianism and psychopathity.
Would you like to answer some of these?
Shall I see?
Well, I would...
No, I like this one.
I want to ask you this, Matt.
One to five.
Okay?
Strongly disagree.
Let's psychoanalyze you.
It's not wise to let people know your secrets.
Agree.
I agree with that.
Five.
Whatever it takes, you must get the important people on your side.
Yeah, strongly agree.
Avoid direct conflict with others because they may be useful in the future.
You've chosen this one to make me look bad.
Look, I didn't get to...
Answer the question!
Do you agree with that?
I didn't get to full professor by...
Walking into brick walls.
Keep a low profile if you want to get your way.
Actually, look, all of this is great advice for getting ahead in academia.
Stop skipping the answers.
So what about that?
Keep a low profile if you want to get your way.
Are you agreeing with that?
Somewhat, somewhat agree.
I generally agree.
Does manipulating the situation take planning, Matt?
Does it?
It does for me.
Yeah.
Is flattery a good way to get people on your side?
Well, I don't flatter people.
Skillful flattery.
Skillful flattery.
You know it's a good way.
Yeah, okay.
All right.
It is.
I like that one.
Do you love it when a tricky plan succeeds?
Who doesn't, Chris?
Who doesn't?
Who doesn't enjoy a good tricky plan?
So, you would, on this, be scoring quite high on Machiavellianism, I think.
I accept that.
Look, you were big enough to accept that you are a sadist.
Apparently.
Yeah, I couldn't have the scale.
And I'll admit that I am somewhat Machiavellian.
But, you know, that's how I got where I am.
Yeah.
So there are other ones.
What about this one?
Can I ask you these, Matt?
There's just like seven of them.
Okay.
I'll put them in the form of questions for you.
So do people see you as a natural leader, Matt?
They do not.
I know they don't.
No, they don't.
People think you have a unique talent for persuading people?
No.
No, I don't think so.
I wouldn't say that.
Hey, you have to agree so quickly.
Are group activities dull without you?
Yeah, well...
More dull?
I'm never there.
I'm not there.
How do I know?
This podcast would be a lot.
That's true.
I've seen you.
Do you know that you're a special person because people keep telling you that?
People don't tell me that so much.
Oh, this is good.
Okay.
Do you have some exceptional qualities?
Oh, look.
Who doesn't?
I think I have a few.
Come on, I've got a few.
Yeah, I've got a few.
Are you likely to become a future star in some area?
A ship has sailed, man.
Whatever degree of stardom I've obtained, I've peaked.
I've got as far as I'm going to get.
And do you like to show off every now and then?
Yes, if possible.
If you'd like to continue listening to this conversation, you'll need to subscribe at patreon.com slash decoding the gurus.
Once you do, you'll get access to full length episodes of the decoding the gurus podcast, including bonus shows, gurometer episodes, and decoding academia.
The decoding the gurus podcast is ad free and relies entirely on listener support, subscribe,
We'll save the rainforest, bring about global peace, and save Western civilization.
And if you cannot afford $2...
You can request a free membership, and we will honor zero of those requests.