Trigger Warnings, Content Warnings, and Content Notes
May 23, 2024 4:34 AM   Subscribe

"We present the results of a meta-analysis of all empirical studies on the effects of these warnings. Overall, we found that warnings had no effect on affective responses to negative material or on educational outcomes. However, warnings reliably increased anticipatory affect. Findings on avoidance were mixed, suggesting either that warnings have no effect on engagement with material or that they increased engagement with negative material under specific circumstances."

Full Abstract:

"Trigger warnings, content warnings, or content notes are alerts about upcoming content that may contain themes related to past negative experiences. Advocates claim that warnings help people to emotionally prepare for or completely avoid distressing material. Critics argue that warnings both contribute to a culture of avoidance at odds with evidence-based treatment practices and instill fear about upcoming content. A body of psychological research has recently begun to empirically investigate these claims. We present the results of a meta-analysis of all empirical studies on the effects of these warnings. Overall, we found that warnings had no effect on affective responses to negative material or on educational outcomes. However, warnings reliably increased anticipatory affect. Findings on avoidance were mixed, suggesting either that warnings have no effect on engagement with material or that they increased engagement with negative material under specific circumstances. Limitations and implications for policy and therapeutic practice are discussed."

Citation:

Bridgland, V. M. E., Jones, P. J., & Bellet, B. W. (2023). A Meta-Analysis of the Efficacy of Trigger Warnings, Content Warnings, and Content Notes. Clinical Psychological Science, 0(0). https://doi.org/10.1177/21677026231186625

Note that the Association for Psychological Science has a more public-facing writeup about the research, if you'd prefer a non-researcher-oriented read. If, instead, you'd like to read more research, you might take a look at the articles citing/responding to the meta-analysis.
posted by cupcakeninja (75 comments total) 17 users marked this as a favorite
 
Of the responding articles, I am particularly keen to read Jones, P. J., Bridgland, V. M., & Bellet, B. W. (2023). Content warnings reduce aesthetic appreciation of visual art. Psychology of Aesthetics, Creativity, and the Arts. Abstract looks interesting; the final sentence did make me chuckle.
posted by cupcakeninja at 4:37 AM on May 23 [1 favorite]


Did the study also investigate the consequences of the content warnings the MPAA has been applying to R-rated movies since September of 1990, e.g., "Rated R for strong/graphic violent content, disturbing material, sexual content, graphic nudity, drug use, and pervasive language"?
posted by Faint of Butt at 4:43 AM on May 23 [15 favorites]


I haven’t finished reading but this is the actual first two sentences:

“ A trigger warning, content warning, or content note refers to a statement intended to help individuals prepare for or avoid content. These warnings differ from other, older types of content labeling (e.g., Motion Picture Association ratings intended to provide guidance for parents regarding content appropriacy for children; Motion Picture Association, Inc., 2023) in that they aim to protect individuals whose unique experiences have left them emotionally vulnerable to specific material.”
posted by brook horse at 4:46 AM on May 23 [20 favorites]


It’s context dependent but I like the idea I read about some streaming service having content warnings in the show description or somewhere like that, so that you could seek them out specifically if you chose, but weren’t confronted with them at the start of the episode. If content warnings feel like a jump scare when they’re shown the latter way, maybe it makes sense to have them be something you can look up if you’re already concerned.
posted by throwitawayurthegarbageman at 4:51 AM on May 23


Indeed, despite most articles concluding trigger warnings are not helpful, they continue to be widely used by the public.

Give the people what they want!
posted by chavenet at 5:00 AM on May 23 [2 favorites]


what do you mean not helpful, you put an "animal cruelty" warning on something, i am not going to read it or watch it, that's helping me because i don't have to spend the rest of my day trying to distract myself from my own damn brain

fuck this study
posted by seanmpuckett at 5:02 AM on May 23 [56 favorites]


I haven’t finished reading but this is the actual first two sentences:

Well, that's on me for not RTFA. But considering the fact that MPAA ratings can be used by individuals who seek warnings about specific content, and that content warnings can be used by parents who wish to protect (or insulate) their children, I'm not seeing the objective difference.
posted by Faint of Butt at 5:06 AM on May 23 [6 favorites]


Following on from what seanmpuckett said, seems to me like warnings are much more useful if you feel you have a meaningful degree of control over your engagement with the content. Using two of the examples in the study, having a trigger warning on course material may well lead to a greater overall negative emotional response. A student might start spiraling imagining that their grade depends on being exposed to the most upsetting trigger they can imagine, and potentially having to confront it in a discussion in a crowded classroom full of their peers. On the other hand, a placard in a museum clearly indicating how to bypass an exhibit might just elicit a "yeah, no, not going in there" response.
posted by nanny's striped stocking at 5:17 AM on May 23 [19 favorites]


I used to be an English teacher. I retired just in time (5-6 years ago). Students were just starting to request trigger warnings. Given that the stories I liked to use in the classroom often dealt with emotionally freighted material (which teenagers and adults gravitate towards anyway), much of the material merited content alerts.

No big deal. Trigger warnings are just a social convention these days, like linguistic conventions and fashion trends. Older people find the necessity of introducing new material with these prefaces annoying because learning new habits is never easy. However, as this meta-analysis shows, they don't really do much. Issuing trigger warnings doesn't cost much for the instructor/exhibitor/podcaster, but following this convention shows a little respect for those (usually youngish) people who expect them.
posted by kozad at 5:27 AM on May 23 [21 favorites]


In a thread on a prior study, it was noted that the study had completely eliminated any test subjects with mental health concerns (such as PTSD) that content and trigger warnings are aimed at. Also, the author cites open white supremacist Jonathan Haidt and includes several of the usual Atlantic and Guardian articles whining about how the wokesters are ruining higher education. And finally, two of the authors are scholars at the Heterodox Academy, a notorious anti-"woke" organization whose entire existence seemed to revolve trying to prove that kids these days are being mind-controlled into transgender furries who are also Hamas terrorists.
posted by Glegrinof the Pig-Man at 5:29 AM on May 23 [97 favorites]


I would not expect that CWs would have any large societal effect. What I would expect (and it's not mentioned in the abstract) is that it's helpful for certain sets of individuals to either prepare for or make choices.

The fact that they don't have overall effects isn't surprising.
posted by CheeseDigestsAll at 5:30 AM on May 23 [10 favorites]


Bellet et al 2018 was a sample of only "non-traumatized" participants. Boysen et al 2021 also seems to be a survey of participants who don't necessarily have a history of trauma or PTSD.

I don't have time to look through all the articles in the meta-analysis because I should be getting ready for work, but ... it seems obvious to me that "are trigger warnings helpful to people who don't have a history of PTSD?" is not the interesting or relevant question. The interesting and relevant question is "Are trigger warnings helpful to people who do have PTSD" - and specifically, questions like "Are trigger warnings about suicide helpful to people who have suicidal ideation? Are trigger warnings about suicide helpful to people who have lost loved ones to suicide? Are trigger warnings about sexual assault helpful to survivors of sexual assault?"

Just like a warning about peanuts is not helpful to me as a person who doesn't have a peanut allergy, you wouldn't necessarily expect a trigger warning to be helpful to a person who doesn't have triggers, and you shouldn't be surprised that it isn't! If you're not specifically looking at the group of people who want to have trigger warnings on things because they have been triggered in the past, then how are you going to know if trigger warnings are helpful for that group of people specifically?

(with the caveat, of course, that "people who have trauma" and "people who get triggered" and "people who want trigger warnings" are not completely overlapping circles in a Venn diagram.)
posted by Jeanne at 5:31 AM on May 23 [31 favorites]


As another advantage, content warnings can reassure the audience that the presenter of the material has actually familiarized themself with the material, something that is not always guaranteed.
posted by Faint of Butt at 5:34 AM on May 23 [7 favorites]


I haven't RTA yet - but it always strikes me as a strange coincidence that the Kind of People who staunchly maintain that Content / Trigger warnings are unnecessary, are *also* the same Kinds Of People who insist that respecting someone Pronouns is unnecessary ....

It's a very specific form of solipsism: "This Thing doesn't matter to me, so it shouldn't matter to anyone else !"

Then they go out of their way to illustrate just how much this Thing *does not matter to them* - and other people should stop making them have to engage / think about it.

Strange.
posted by Faintdreams at 5:37 AM on May 23 [26 favorites]


I just wanted to follow up on my previous comment to make it clear that I am not arguing against trigger warnings for course material; I definitely know first-hand the sinking, suffocating feeling of sitting down for what I thought was going to be a normal class only to find out the next hour was going to be my own personal Ludovico Technique session. It seems like a really tough balancing act for educators, especially US high school teachers level given the highly politicized climate and the fact that US students are basically inmates, but I'm glad society is grappling with this and other big questions related to mental health and accessibility.
posted by nanny's striped stocking at 5:58 AM on May 23 [3 favorites]


Content warnings on material in the syllabus you will unavoidably have to discuss: yeah, makes sense it wouldn't reduce anxiety by much. At least they showed that while the content warnings might not be helping, they're probably not hurting (educational outcomes) either.

Even content warnings / trigger warnings on avoidable things - for example fanfiction on the internet - are not all that straightforward, though. For one thing, there are thousands of things that might be triggering to someone, but we only warn for a handful of the most common triggers. For another, it can create these ideas around responsibility - it's your responsibility to warn me should be coupled with it's my responsibility to heed your warnings, but it isn't always.

There's also this idea that trigger warnings "don't cost anything" but I'm not sure that's actually true - as an author you have to put yourself into a certain kind of headspace to think of all the ways something you're writing (drawing, etc) might upset someone else, and that's not nothing, especially if you're in a milieu that expects a lot of trigger warnings. It's not necessarily a huge burden but it will discourage some people. The presence or absence of trigger warnings on a piece of art can be used as a way to determine who is or isn't part of the in-group and as a vector for bullying, same as any other social expectation.

Beyond that, it's also a recognized phenomenon that if trigger warnings are something you can sort on - like a tagging system - at least as many people will use them to search for certain kinds of content, as will use them to avoid that content.

I'm not like anti trigger warning by any means, but they're not as straightforward as we pretend they are. They are a compromise, a very imperfect one.
posted by subdee at 6:01 AM on May 23 [12 favorites]


Yes (to many comments above), this meta-analysis seems to me to have gaps. I was particularly struck by the absence of discussion of, uh, many of the factors that go into my decisions about using any form of warning in online discussion or in the classroom. It did, however, give me some interesting points to consider about research and these kinds of warnings.

seanmpuckett, I feel you on the animal cruelty warnings.
posted by cupcakeninja at 6:04 AM on May 23 [4 favorites]


But considering the fact that MPAA ratings can be used by individuals who seek warnings about specific content, and that content warnings can be used by parents who wish to protect (or insulate) their children, I'm not seeing the objective difference.

Isn't the MPAA infamously opaque about what their criteria is? And aren't the values reflected by their ratings prescriptive because of how ratings are used to police the film industry? And isn't it an open secret that there are a ton of loopholes to exploit so long as you patronize the values of the MPAA?

I always view content warnings as a show of empathy. They're usually issued by a creator or a presenter (as opposed to an outside organization) and when they're done well, the warnings usually reflect an understanding that some of the content might be disturbing for people who have had a certain set of experiences.
posted by RonButNotStupid at 6:08 AM on May 23 [11 favorites]


I must be weird because sometimes I look at CWs and say "OK not looking at that, don't need $ATROCITY_IMAGE today."

Also about the assertion upthread that CWs are intended only for people with PTSD. That's news to me.
posted by Aardvark Cheeselog at 6:21 AM on May 23 [11 favorites]


It's all about the tolerable level of permanent damage for other people or, as the brilliant Innuendo Studios would have it, Didoing

As mentioned above, the study excludes people who'd potentially be more meaningfully triggered. But really, even if this just proves there's no effect for those that aren't, the objection to a couple of lines of text about things that might be unpleasant for some people doesn't matter to content warning opposers, because... other people's potential discomfort doesn't matter to them. Just be triggered, even if I'm not that impacted by something that might help someone else. Dislike trigger warnings all you like, but what it means is you think someone else's discomfort isn't worth your giving of a shit.

But sure. Let's have a study on whether antibiotics might cause allergies but exclude everyone with allergies from the group we study. Good job.
posted by onebuttonmonkey at 6:22 AM on May 23 [13 favorites]


The MPAA’s main purpose is not to provide helpful warnings to the public. Rather, it’s to do the absolute minimum work possible to self-regulate the industry…in order to keep the government from stepping in. It is all about protecting the film industry and not really about helping consumers make better media choices.
posted by Doleful Creature at 6:26 AM on May 23 [4 favorites]


There was a fascinating episode of the Search Engine podcast on this topic very recently, interviewing a researcher who had done work with subjects with PTSD (amongst many studies on the topic).

One point of note was that actual triggers for PTSD are often small things associated with the very personal memory of trauma, rather than the thing itself... eg, if you get shot at while eating Neapolitan ice cream, maybe Neapolitan ice cream becomes a trigger. And good luck getting trigger warnings for that.
posted by kaibutsu at 6:29 AM on May 23 [17 favorites]


way i've seen this? note everything.

hypothetical semester

session 1: Y (relatively non-controversial topic)
okay

session 2: X ...
fX! i mean seriously. f it. why is everyone on about X. don't they know what i've been through?! y'all can just sit around and chat about X? this is something i have had to deal with my entire life. My. Entire. Life. i don't care if i "Fail." if i have to sit through another group of uninformed people spouting off about X for another minute-
*note: no demerits for missing class (syllabus)
oh, okay. well, not okay. but okay

session 3: Z &c
posted by HearHere at 6:30 AM on May 23


>As mentioned above, the study excludes people who'd potentially be more meaningfully triggered

This is not quite true. In this thread, Glegrinof the Pig-Man did mention one study posted to MF previously that excluded folks with a trauma backround. However, the metastudy that is the subject of this post explicitly discusses whether each of the individual studies included folks with a trauma background or without, and most of them did not exclude:

Overall, 142 of the 144 effect sizes meta-analyzed (98.6%) included trauma survivors in their sample. Indeed, only two effect sizes in the meta-analysis excluded trauma survivors (referred to as “trauma-naive” in Table 1; i.e., Bellet et al., 2018, 2020), but several others exclusively studied trauma survivors (referred to as “trauma only” in Table 1; e.g., Jones et al., 2020) or examined subsamples of trauma survivors (e.g., Bruce & Roberts, 2020).
posted by mrgoldenbrown at 6:32 AM on May 23 [8 favorites]


Dislike trigger warnings all you like, but what it means is you think someone else's discomfort isn't worth your giving of a shit.

I always get a chuckle from the disproportionate outrage against content/trigger warnings generated online, it's peak "im not mad. please dont put in the newspaper that i got mad" energy.
posted by Glegrinof the Pig-Man at 6:32 AM on May 23 [10 favorites]


i seem to see this discussion brought up every few years, 100% always by people who are highly invested in downplaying the importance of content tagging. always a lot of goalpost-moving and narrow parameters. and i'm just reminded of every other instance where there's all this energy expended to try and convince people they don't have to show people a common courtesy when they could just show them that courtesy. like just be decent goddamn, save us all the trouble.
posted by Aya Hirano on the Astral Plane at 6:34 AM on May 23 [9 favorites]



Content warnings on material in the syllabus you will unavoidably have to discuss: yeah, makes sense it wouldn't reduce anxiety by much.


If that’s how people are using them, they’re using them wrong. When I received content warnings as a person with identified PTSD, it was so that I could not attend class for that session and/or get an alternate assignment. What is even the point of a trigger warning when you then can’t avoid the trigger?
posted by corb at 6:36 AM on May 23 [11 favorites]


For what it's worth, it wasn't my intention to assert that content warnings are intended only for people with PTSD. But if the question is "do trigger warnings reduce anxiety," I would expect there to be a pretty big difference for people who would be triggered by the content in question and people who wouldn't be.
posted by Jeanne at 6:41 AM on May 23 [6 favorites]


Beyond that, it's also a recognized phenomenon that if trigger warnings are something you can sort on - like a tagging system - at least as many people will use them to search for certain kinds of content, as will use them to avoid that content.

Is this a criticism of trigger warnings? That they have even more uses than intended?
posted by Dysk at 6:43 AM on May 23 [8 favorites]


I haven’t had time to delve into the study but it seems to be both asking the wrong question, not aimed at the right level of aggregation, and underpowered. Every movie we watch is looked up in the doesthedogdie database for a specific trigger, and we don’t watch it if it’s in there. This study seems to be kinda like “if we give everyone Tylenol, will the headache rate go down”? Like who gives a shit.
posted by MisantropicPainforest at 6:59 AM on May 23 [5 favorites]


I've finally put a slide in my infosec course intro to the effect that folks with increased anxiety may wish to reconsider taking the course, and I won't be upset if they drop.

That's after a couple-three student mental-health incidents serious enough for me to file student-of-concern reports. One was apparent to other students to a class-disrupting extent.

I always have warned for especially upsetting content (e.g. I discuss stalkerware and spyware, which inevitably brings up harassment and abuse), but I didn't expect garden-variety adversarial thinking or discussion of bog-standard attacks and attacker motives would do that much harm.

Live and learn... and give people choices. I can't just not teach the upsetting stuff, but my class is an elective, so I can offer appropriate warnings as early as possible.
posted by humbug at 6:59 AM on May 23 [10 favorites]


One way to investigate the effects of trigger warnings is to ask an intermediate question: “do victims of childhood sexual assault experience distress after viewing material graphically depicting childhood sexual assault?” The answer is obviously ‘yes’. Now ask, “do trigger warnings reduce the likelihood that a victim of childhood sexual assault will view material graphically depicting childhood sexual assault?” Again an obvious yes. QED.
posted by MisantropicPainforest at 7:04 AM on May 23 [5 favorites]


I'm not like anti trigger warning by any means, but they're not as straightforward as we pretend they are. They are a compromise, a very imperfect one.

This jogged loose a comment I heard on a podcast a while back (can't remember which one) where the host mentioned a fan asking for content warnings on depictions of food (in a fanart context maybe?) Not disordered eating, not like, dishes that are identifiably a dead animal, just food in general. That did strike me as a bit of an out-there request; if the unexpected sight of food itself is deeply upsetting to you, I think unfortunately you are just going to have to develop coping strategies for that or adopt a very constrained lifestyle.

One point of note was that actual triggers for PTSD are often small things associated with the very personal memory of trauma, rather than the thing itself... eg, if you get shot at while eating Neapolitan ice cream, maybe Neapolitan ice cream becomes a trigger. And good luck getting trigger warnings for that.

Yeah as a child I was triggered by, of all things, the then-popular song "A Whole New World" from Aladdin for a while because I associated it with an older relative's traumatic health crisis. Not really any way to protect against that.

I'm not sure what I'm trying to get at here; on one hand I don't think you can just mandate a list of "The 14 Triggers" in some kind of analogy to the 14 EU allergens, but on the other hand it's clear that most traumas and triggers can be grouped in a small number of broad categories like "physical or sexual violence", "emotional abuse, bullying, and systemic discrimination", "death and illness", and "addictive behaviors". It's a tricky balancing act, and like I said in my previous comment, I don't envy those who have to navigate these issues in a professional context.
posted by nanny's striped stocking at 7:14 AM on May 23 [3 favorites]


I care less about the specifics of studies and more about the fact that a lot of people describe their internal experiences of trauma and distress and there's a bunch of people who respond to that with "NO YOU DON'T SHUT UP".
posted by Pope Guilty at 7:29 AM on May 23 [16 favorites]


A lot of folks seem to be assuming bad faith on the part of the metastudy authors, without reading either the paper or the layperson summary. If she is a right wing anti woke activist who doesn't care about anyone, she is good at disguising her intentions.

She herself still uses content warnings when teaching classes for example.

Her stated goal is to help people by giving them an effective tool. “We need more strategies to give people versus just putting a warning on something and assuming that is going to give them a toolkit for mental health.”
posted by mrgoldenbrown at 7:31 AM on May 23 [21 favorites]


One of the few non-evil uses for generative AI could be scanning content for anything the user wants. I agree it's unreasonable to expect teachers and artists to be able to anticipate every possible trigger, but it would be nice if people with disagreeable limbic systems had tools at their disposal.
posted by Reyturner at 7:35 AM on May 23 [2 favorites]


Is this a criticism of trigger warnings? That they have even more uses than intended?

It's not a criticism per se, BUT in my experience, people being able to search for things (via tagged trigger warnings) can create an identifiable audience which can result in MORE art being about those things. Which, if that art is in the same place as the original art, with the culture of tagging still in place, it will also be tagged and can presumably also be avoided. But the point is that trigger warnings aren't as simple as we pretend they are. There's this whole social scene that's sprung up around who uses what warnings in what ways.

In general I'm a fan of the warnings because if we didn't have "don't like, don't read" culture online there would be a lot more calls to outright ban things that people don't like... though sometimes putting the searchable triggers on online stuff means you can find it more easily, and then quantify how much "bad stuff" is out there, and that can lead to MORE calls to ban all the bad stuff...
posted by subdee at 7:53 AM on May 23 [1 favorite]


There was a fascinating episode of the Search Engine podcast on this topic very recently, interviewing a researcher who had done work with subjects with PTSD (amongst many studies on the topic).

I didn't read this article, but I did recently listen to this episode, and the researcher he interviewed is one of the researchers on this paper - Victoria Bridgeland.

My takeaway wasn't that Bridgeland is some odious figure - the history of trigger warnings presented in the podcast is interesting, namely that it was organic and largely via the online FanFic world, not rooted in any psychological study. And so she was curious to actually study them as a psychologist.

But it did appear to me, as someone who has somewhat used them in a classroom, that she was asking the wrong questions. Or at least, wasn't asking all of the potential questions. I believe her studies that found that warning people "you are about to see something bad" doesn't make the bad thing less disturbing - according to her, the idea that people can prepare themselves turns out not to be true -OK. And she found that most people don't avoid potentially disturbing content, and what actually triggers people's trauma response can be random (as already mentioned). All of that is seems reasonable.

But then (in the podcast) she concludes that they serve no purpose, and I found that to be not convincing. She never considered the reasons why I, an educator, would use them. Mainly, I'd say something along the lines of "Hey, just a heads up, this video we're going to watch gets into [dark topic], if you need to take a breather at any point that's fine, take all the time you need." For me it was always about demonstrating empathy, taking the time to preview anything I was going to show to students (which sometimes resulted in me tossing the material all together - seriously, some educational material from the 90s is....not great), etc. Students don't even expect that much in my experience - they just don't want to have to consume upsetting material unless there is some sort of clear educational reason for doing so, and they'd don't want to be penalized if they need to engage with the material on their own terms.
posted by coffeecat at 8:25 AM on May 23 [17 favorites]


TW: few non-evil uses for generative AI could be scanning content for anything the user wants. I agree it's unreasonable to expect teachers and artists
posted by HearHere at 8:36 AM on May 23


I have a lot of thoughts about this, but to jump on one that hasn't been brought up here yet...honestly. If you teach Ovid's elegant poems of rape and other forms of sexual coercion in a classroom as if there was nothing objectionable about the content at all, nothing worthy of note, nothing to see here, just a bunch of rapes and trees and brooks, you are doing a pedagogical harm that goes beyond provoking an immediate trauma response. I'm not talking about somehow promoting or excusing rape, I'm talking about what happens when authority figures treat sexual violence as a matter of aesthetic concern only, not as something that has real-life impact.

When you teach, say, the Iliad or the Odyssey, each has, in addition to the glorification of physical violence, built-in laments for the resulting loss of life and the impact of death on the people left behind. It seems like everything stops in the Odyssey every three books so people, mostly men, can cry together about how awful the experience of the Trojan War was. Not really so much for the sexual violence. It is fucking bizarre to teach texts supposedly about life as if sexual violence were merely an aesthetic phenomenon, and you could easily, easily have a course including Ovid taught to freshmen in this century where no one even acknowledges in-course that it is rape.

I still remember the day in high school when a particularly high-strung student came to another class in tears because she'd just learned in Latin class what a catamite was, she'd figured out the implications but they obviously weren't addressed in the class, and the response of some of the other students was...to make fun. Of someone being upset to learn about a Roman sexual practice that was often though not exclusively pedophilic. Because they were kids who got no guidance on how to deal with the reality of what they were talking about, or even an acknowledgement that it was awful for those actually involved. Now, this girl would have been the poster child for both the snowflake and the limousine liberal stereotypes, but, still, that was wrong.

You don't have to address this issue with trigger warnings/content notes, but it is a way, albeit imperfect, to do so.
posted by praemunire at 8:50 AM on May 23 [21 favorites]


(skip this comment (you can press 'j' if you're on a desktop) if you don't need to read today about sexual assault in movies.)



A while back I watched a pretty famous film: Pedro Almodovar's Talk to Her. I'd seen it described in various places as a feminist movie, and Almodovar as a feminist director. I'd seen another movie of his that I liked. I thought, based on reviews I saw, that it was about men who "seem happy to devote their lives to [comatose] women who do not, and may never, know of their devotion." A "masterful, compassionate work." Something gentle and bittersweet - "like all great doomed affairs, 'Talk to Her' is full of lovely, sweet suffering."

I had read multiple reviews and still somehow had no idea that regardless of anything else, this was a movie about rape, which, it turned out, both happens offscreen and is shown as an extremely graphic animated fantasy in which a woman's body is explicitly and solely an object for a poor besotted man to give his "devotion" to. The movie is the story of the rapist, and the gentle compassion is absolutely for him, not for the victims who are incapable of consent. It more than ruined my day: exploring why people do terrible things, or creating some sympathy for those people, is one thing, but it felt to me like this went way beyond that.

After I watched it, pretty nauseated and furious, I went back and looked for reviews that actually mentioned any of this. Because even if Almodovar's goal was to highlight how fucked up even "innocent" rape can be (which is really not clear), how could it be that discussions of the film didn't center said fucked-up-ness? But most of the reviews I saw either didn't mention it at all, or completely danced around it with the vaguest of references. The one review I saw before I decided to stop that actually used the word "rape" and spelled things out nonetheless said "The action of Talk to Her could be horrific, or at the very least nauseating. Yet such is the imaginative warmth that Almodovar conveys for his two male leads, combined with his stylisation and modification of the real world in which the rape happens, [that] revulsion is neutralised through a combination of sympathy and alienation. Almodovar so expertly manages his movie's perspectives that the rape looks like a subsidiary event in an essentially heartwarming, tragicomic fable of 'relationships'. It leaves unanswered the question of how exactly we are supposed to think and feel about this rape, or if it is a rape at all." (You might take a guess about the gender of the reviewer.)


Anyway, that movie and those reviews came out in 2002, and I'd like to believe that today, even if they didn't have somewhat different interpretations of the film, more reviewers would at least not dance around the fact that rape occurs in it, is central to it, and you're absolutely going to see a depiction of it, because they would understand that there are some categories of things that at least some people are strongly going to want to know about when deciding whether to watch. And more of them might understand, today, that not everyone is able to 'neutralize their revulsion'.

tl;dr Maybe trigger warnings in their current format are overused, maybe they aren't, maybe the format could be better, maybe not, but I'm grateful for whatever increased cultural awareness makes it more likely to be given this kind of information up front. And less likely to find yourself stunned and asking "wait, did no one else even register this stuff?"
posted by trig at 9:30 AM on May 23 [17 favorites]


I'm going to keep using content warnings in my writing, the negative effect appears to be small for the general population, and some people do desist from engaging in material based on a warning. It would be interesting to see explorations of alternative strategies for reducing distress though: for example, does offering a brief guided meditation as an option before engaging with material lead to people doing the meditation, and if so is that helpful?
posted by Chrysopoeia at 9:32 AM on May 23 [1 favorite]


One of the few non-evil uses for generative AI could be scanning content for anything the user wants.

Doesn't even need to be generative, though the concept of "discovering" triggers (vs simply doing multilabel classification) is an interesting one. In a previous job, discovering new terms via a model was always the holy grail, but we sucked at it to the degree that waiting for people complain to X was missing was more efficient than reviewing all the proposed new terms.

Tangentially, the one "OMG we must put 'AI' in our product!!!!" feature I've seen that's actually pretty good is Amazon's review summaries. It is hilarious that it always has to produce a negative ("that said, ..."), and I suspect they're doing slot filling rather than "true" seq2seq.
posted by hoyland at 9:35 AM on May 23 [2 favorites]


As someone working in the GLAM space, I think this analysis really glosses over what purpose content warnings serve in a pedagogical/social sense. In our library, we've fairly recently started added explanatory text in the front matter of some of our archival digital collections that are, for example, hella racist. Does that text (is it a content warning? I don't know) mitigate the affective response of someone who looks at that material? I don't know -- and that's where the authors seem to have started and ended their analysis. But I am fairly certain that including text like that demonstrates that we are familiar with the contents of our collection, we know our research community, and we want to provide access to materials in a thoughtful and pedagogically-appropriate way. We may do it wrong - but our researchers know we've made an effort and know how to have a conversation with us about doing better. That's....not nothing.
posted by pantarei70 at 9:59 AM on May 23 [12 favorites]


TW, CW, etc are just an evolution of age-rating systems, except it’s more detailed. They are completely optional to even access, except in academic settings. So, if you don’t like them, don’t read them. I’ve put content warnings on all of the stories I’ve published (mostly horror) and none of it has been criticized by the actual readership. Usually, it’s just criticized by people outside using a bunch of dumb hypotheticals.

When Judy Dench criticized them as to plays, she knows that people can just access a complete plot outline of the play beforehand, right? Should these not exist either? I mean, they totally spoil the play, don’t they?
posted by 90s_username04 at 10:06 AM on May 23 [1 favorite]


she knows that people can just access a complete plot outline of the play beforehand, right?

Maybe we could bring back the tradition of describing what happens in every book chapter section of content before it starts: "IN WHICH our Hero Encounters a Site -- Professional Colors are Pondered -- $$$$$ Must be Obtained -- A Username is Created -- A Comment is Posted -- Another -- A Lull -- Violent Objections -- A Banning Occurs!"

No one could object to this, the faithful return of a beloved tradition.
posted by trig at 10:20 AM on May 23 [16 favorites]


The last time we had a post about these sorts of studies, it was clear the investigators had made no attempt to distinguish between "Here's an opportunity to avoid something that might upset you..." and "Buckle up! You're about to see something that might upset you!" It wasn't even clear they even understood the distinction. Which made the whole investigation utterly worthless to my mind.

So I'm not bothering with the article unless someone says they definitely made that distinction.
posted by straight at 10:24 AM on May 23 [4 favorites]


I get a lot of value out of online CWs. I nope out of clicks all the time. I think if they were offered in a lit class I were in, I'd ask for an opt-out mechanism (e.g. email only to CW subscribers). Surprise is an essential element of storytelling, imvho.

Gotta rtfa.
posted by j_curiouser at 10:27 AM on May 23 [1 favorite]


I sometimes wonder how much of all of this controversy could have been avoided if instead of calling them content warnings, they were just called 'content notes' or 'tropes'.

As we've learned from the existence of TVTropes, people will happily categorize every beat of a 12 season television show as each of three different tropes and then argue ad nauseum about whether the character's reaction shot was any old Fourth Wall Break or specifically an Aside Glance. It isn't describing content or having content described that bothers people, apparently, but the idea that it's a thing they should do to help others.
posted by jacquilynne at 10:45 AM on May 23 [6 favorites]


"content tagging" seems like a perfectly useful thing to me - like a director's commentary one can opt into - if you market it as something anyone can make use of for whatever filtering purpose they desire.
posted by I-Write-Essays at 11:00 AM on May 23 [2 favorites]


The last time we had a post about these sorts of studies, it was clear the investigators had made no attempt to distinguish between "Here's an opportunity to avoid something that might upset you..." and "Buckle up! You're about to see something that might upset you!" It wasn't even clear they even understood the distinction. Which made the whole investigation utterly worthless to my mind.

So I'm not bothering with the article unless someone says they definitely made that distinction.



For one, this is a meta analysis.

Second, in the time it took to write your comment you can look exactly what the warnings are!

You can also look at the effect sizes for each study (meaning for each warning) and decide for yourself.

Third, researchers are very often operating under significant contraints. The reason they don't do things that people want them to do is very often because of one of these constraints. That doesn't make the endeavor worthless.
posted by MisantropicPainforest at 11:12 AM on May 23 [6 favorites]


I wish I knew how to tell whether it was correlation or causation, but the rise of the term "trigger warning" came right about the same time when there were a lot of young people returning from the wars in Iraq and Afghanistan, many presumably with PTSD from the experience, and going to back to civilian life and getting their college educations. Part of me wants to see some connection there.
posted by Zalzidrax at 11:16 AM on May 23 [3 favorites]


I don't know. Maybe there's some significant scholarly value here. I'm not in the headspace to do a deep reading and analysis today even if I were qualified to make some any kind of judgment based on the research, which I am definitely not. My main takeaway is, here's yet another example of "oh hey, this little thing that people do to show respect and empathy toward each other, you know to sort of smooth the way a little given how shitty the world can be, just a little recognition of our shared humanity we sometimes do -- yeah that thing is useless and you don't have to do it anymore, and if you got used to it, better get over it".

There must be dozens of other trivial things we do for each other each day that could be eliminated entirely. Just imagine the tremendous increase in efficiency if we as a society could someday manage to identify and rid ourselves of all of them. Hats off to the scholars out there doing this important work.
posted by Two unicycles and some duct tape at 11:45 AM on May 23 [3 favorites]


My main takeaway is, here's yet another example of "oh hey, this little thing that people do to show respect and empathy toward each other, you know to sort of smooth the way a little given how shitty the world can be, just a little recognition of our shared humanity we sometimes do -- yeah that thing is useless and you don't have to do it anymore, and if you got used to it, better get over it".

There must be dozens of other trivial things we do for each other each day that could be eliminated entirely. Just imagine the tremendous increase in efficiency if we as a society could someday manage to identify and rid ourselves of all of them. Hats off to the scholars out there doing this important work.
'

This is, fundamentally, anti-science. If the results they found were different, you would laud them for confirming something everyone knows. Do you think they should 1) not do meta analyses of trigger warning experiments or 2) do meta analyses of trigger warning experiments, but if the results come out one way or the other, not publish the results or 3) fake the results to be more palatable?
posted by MisantropicPainforest at 12:41 PM on May 23 [8 favorites]


Okay, so I went and read this thing, and it's much dumber than I assumed.

They acknowledge that some studies measure how someone feels if they're given a trigger warning before they are exposed to something, and other studies measure whether trigger warnings change people's behavior regarding whether or not they choose to expose themselves to something. But these are two completely different theories of what trigger warnings are and why they might be helpful. (And both theories are wrong, which I'll get to in a minute.) But the authors go right ahead and lump both of these together so they can make a blanket statement about whether trigger warnings "work."

(It's like I asked people "Do you care if I ruin one of the surprises in Empire Strikes Back?" and counted how many people say yes. And you just ran up to people and said either "SPOILERS! VADER IS LUKE'S DAD!" or "VADER IS LUKE'S DAD!" and count how many people in each group were annoyed by this behavior. And then we add up both sets of numbers to find out whether spoiler warnings "work".)

When in fact I would argue that the real question of interest is "If you give people more agency over how, when, and whether they are exposed to certain kinds of material, does that increase their happiness and mental well being?" Neither of those kinds of studies are measuring that.

And just in case I was worried I was being uncharitable, the authors prove they have no clue what any of this is about by discussing at length, "In fact, many times when you tell people ahead of time that something contains possibly upsetting material, they watch it more! That's the opposite of what trigger warnings are for!"

No. The purpose of trigger warnings are not to discourage people from watching something. They are to give people more agency in deciding whether or not to watch something.

My initial hypothesis that this was probably dumb and not worth my time was correct.
posted by straight at 1:09 PM on May 23 [6 favorites]


As we've learned from the existence of TVTropes, people will happily categorize every beat of a 12 season television show as each of three different tropes and then argue ad nauseum about whether the character's reaction shot was any old Fourth Wall Break or specifically an Aside Glance. It isn't describing content or having content described that bothers people, apparently, but the idea that it's a thing they should do to help others.

There's been a backlash to that as well, and personally I find that kind of hyper categorization more offensive than generalized content warnings for depictions of sex and violence (which are actually pretty similar to the hated MPAA warnings, but come from a different place of origin, so they're seen as helpful instead of oppressive).

Once you reach a certain point, you're treating artistic experience the same way you treat a fast casual meal--if there's anything that doesn't provoke the exact right feelings, complain and send it back, just like you'd complain about getting ketchup instead of spicy ranch for your chicken tenders. And as subdee noted, detailed descriptive warnings originated in the fanfiction community, where they were basically intended as getting off guides. Which, fine, but the entire world of art does not need to be tailored to specific orgasmic needs for it to be valuable.

If you want to have a basic ability to engage with art, you have to have the ability to go into some things cold.
posted by kingdead at 1:18 PM on May 23 [4 favorites]


"But the authors go right ahead and lump both of these together so they can make a blanket statement about whether trigger warnings "work.""

Did they say this somewhere? I didn't see it in the article. I couldn't find anywhere where they make a blanket statement that trigger warnings don't work.

FTFA: "Because of their ambiguous usage, we defined trigger warnings by their intent: to help individuals prepare for or avoid content likely to trigger memories or emotions relevant to past experiences."

Seems like this is exactly the definition you're chastising the authors for not adopting?
posted by MisantropicPainforest at 1:30 PM on May 23 [1 favorite]


And finally, two of the authors are scholars at the Heterodox Academy, a notorious anti-"woke" organization whose entire existence seemed to revolve trying to prove that kids these days are being mind-controlled into transgender furries who are also Hamas terrorists.

Conservatives are triggered by content warnings.
posted by AlSweigart at 1:47 PM on May 23 [4 favorites]


Seems like this is exactly the definition you're chastising the authors for not adopting?
Not who you are responding to, but I am making an argument that the operational definition of trigger warnings in this study is inadequate when examining how they are actually used in the wild. From my pov, the definition they've adopted is overly reductive, though I'm sure exactly what they were attempting to measure in the study.
There's been a backlash to that as well, and personally I find that kind of hyper categorization more offensive than generalized content warnings for depictions of sex and violence (which are actually pretty similar to the hated MPAA warnings, but come from a different place of origin, so they're seen as helpful instead of oppressive).
This hyper categorization is perfectly normal and totally fine in communities that require fine detail. I'm in a Discord that recommends romance novels, and the warnings can't just be for "sex" because that would comprise 95% of the recommendations to the point of uselessness.
posted by MagnificentVacuum at 2:14 PM on May 23 [4 favorites]


I highly recommend the Search Engine podcast mentioned above, which features an interview with Bridgeland, a lead author on the meta-analysis. She seemed pretty clear on the limits of science in this area, since, for example we paradoxically cannot directly study the effect easily due to informed consent rules. And where the evidence was thin she suggested more research, (but I suppose that's the code to every question a scientist asks: more grants plz!)

It's not a criticism per se, BUT in my experience, people being able to search for things (via tagged trigger warnings) can create an identifiable audience which can result in MORE art being about those things.

A long long time ago, there was an IRC quote database (now thankfully offline) which featured an up/down voting system, which it used to hide "bad" jokes / quotes. Well, everywhere but the "worst" page, which one of the most downvoted quotes described as "page with the best racist jokes." Some clique of people began downvoting offensive quotes as a result of this system, which clearly kickstarted a feedback loop of more racist behavior and more quote submissions.

Similarly, there are anecdotes about using "proana" content warning tags as an organizing principle. If you add a content warning that only six in a hundred trauma survivors use to avoid, and more use the metadata to self-trigger, it's not as immediately obvious that they are a net positive online.

Paradoxically, for a community concerned with content warnings and literally has the name "filter" in the domain name, there's no tagging convention one could use to customize the "my mefi" page excluded tags list with.
posted by pwnguin at 2:16 PM on May 23 [3 favorites]


I'm not the intended audience for trigger warnings. I don't have PTSD. But sometimes I just don't have the energy to manage my emotional reactions to graphic violence or sexual assault or something anyway. Like -- I'm at work. I've got a meeting. I can't be crying right now. (And the older I get, the more stuff makes me cry. Even fiction -- because I'm more aware of real-world parallels). Plus, I'm a parent, and I've got a kid who wanted to see some horror movies and then had a really hard time with graphic intrusive thoughts afterward. She doesn't want to see anything like that again for a while either.

So anyway, regardless of their utility for trauma survivors, I as person who has not experienced trauma value the warnings that help me control how much of the world's infinite sadness I let into my brain each day. Because I can only manage so much. And my limbic system just did not evolve to deal with the internet.
posted by OnceUponATime at 2:24 PM on May 23 [9 favorites]


I'm always of a lot of different minds about content warnings. I appreciate them for my own consumption, i think they're a marvelous courtesy, and i approve wholeheartedly of people making them available voluntarily. (I also think that in pedagogical contexts, or other contexts where people have limited or no ability to opt-out, they're absolutely essential for at least the worst/most common categories of upsetting content, such as rape and graphic violence.)

But there are also communities where people turn into cops about them, and turn them from a courtesy and a way to be gentle with each other into an excuse for social (and sometimes mod-level) enforcement, and i am extremely not okay with that, especially as the level of "personal political preference" starts to rise (and i say that as someone with extremely Left politics!) Content warnings are always already political, and when people lose sight of that, things can get pretty gross and you can get a lot of lateral aggression. As i've said, in and about these contexts:

You know what one of the top five most common phobias in North America is? Cynophobia. You know what nobody ever, ever jumps your shit for not putting a CW on? Pictures of cute dogs.
posted by adrienneleigh at 2:38 PM on May 23 [19 favorites]


This is, fundamentally, anti-science.

I think there's a comfortably wide margin between "fundamentally anti-science" and "generally skeptical of what appears to be science driven by conservative ideology".

If the results they found were different, you would laud them for confirming something everyone knows.

Making statements unsupported by evidence is, fundamentally, anti-science.

Do you think they should 1) not do meta analyses of trigger warning experiments or 2) do meta analyses of trigger warning experiments, but if the results come out one way or the other, not publish the results or 3) fake the results to be more palatable?

You're clearly trolling with 2) and 3), in line with the unsupported bad-faith assumptions you've already made, so I'll ignore those.

I don't think I'm alone in recognizing that meta analyses are sometimes problematic. It's easy for researchers to come to misleading conclusions by cherry-picking the studies they want to include or by grouping together studies with substantially different methodologies. Throw in ideological bias on top of that and, yeah, I'm skeptical. I'm skeptical about the motivations for these researchers in particular to pursue this particular field of research, I'm skeptical about their objectivity, and I'm skeptical about their conclusions. So yeah, I'm uninterested in trigger warning meta analyses from a group of researchers with obvious ideological bias who came to a conclusion that -- surprise! -- supports their ideological bias. Call that anti-science if you like.
posted by Two unicycles and some duct tape at 2:39 PM on May 23 [8 favorites]


Funnily, from an evidence based practice perspective, their conclusion actively supports using trigger warnings because it INCREASES engagement with triggering content while providing an identical decrease in anxiety (as compared to no trigger warning) after viewing, which is the fundamental basis of trauma therapy.
posted by brook horse at 3:40 PM on May 23 [8 favorites]


I'm a chicken and avoid anything with a hint of triggering. My life is full and interesting enough without upsetting the apple cart.
posted by Czjewel at 3:49 PM on May 23 [2 favorites]


Glegrinof the Pig-Man, I appreciate your comments here. I’m not an expert in psychology, nor warnings, so I neither knew everyone being cited here, nor checked author affiliations. Had I realized those things, I would likely have added something in extended description.
posted by cupcakeninja at 4:18 PM on May 23 [2 favorites]


So, for instance they summarize some of their findings in the abstract like this:

Findings on avoidance were mixed, suggesting either that warnings have no effect on engagement with material or that they increased engagement with negative material under specific circumstances.

Now, the whole point of a trigger warning (as opposed to just not showing people something that might be upsetting) is that you don't know whether it would be upsetting or not so you tell them about it and let them decide. If you warn people and 95% think it's fine and 5% think they'd rather not, then that's working as intended. If 75% think it's fine and 25% think they'd rather not, then that's working as intended. But their meta-analysis would say the trigger warnings were more effective in the second case.
posted by straight at 4:51 PM on May 23 [4 favorites]


Which is something they explicitly acknowledge in the limitations section: these studies are underpowered and thus can’t detect heterogeneous treatment effects.
posted by MisantropicPainforest at 5:13 PM on May 23 [1 favorite]


I was violently abused as a child. Like other people with trauma, a lot of my triggers are weird and random -- there's no way to warn me, and I don't expect it. But at least half a dozen times in the last year or so, I've sat down to watch what I thought would be mindless TV and ended up in tears because of unexpected child abuse. "Dead Boy Detectives" is the most recent example.

I deserve to be able to make informed decisions about stuff like movies, TV, and books. Of course there's no easy, one size fits all solution for everything from fanfic to university curricula. But I'm always surprised at how many people don't think we should even try.
posted by mostlymartha at 10:43 PM on May 23 [6 favorites]


What I'm saying is I still would have chosen to watch "Dead Boy Detectives" had I known about the child abuse. It was a really good show that for the most part handled the subject well. I can't avoid everything that triggers me. But I deserved to make an informed decisions and go into things without the right frame of mind.
posted by mostlymartha at 10:46 PM on May 23 [5 favorites]


I always get a chuckle from the disproportionate outrage against content/trigger warnings generated online, it's peak "im not mad. please dont put in the newspaper that i got mad" energy.

Content warning: woke mind control
posted by flabdablet at 4:04 AM on May 24 [1 favorite]


Call that anti-science if you like.

you literally just discounted something because of your own ideological biases, declining to even look at the evidence because you'd already decided it was wrong, so,

I find excessive content warnings a little eye-rolly, presumably people will still use them if they want to though and the world will keep turning.
posted by Sebmojo at 5:14 AM on May 24


Zalzidrax, I have a scrap of anecdata about your hypothesis— the first time I met classroom content warnings was about 2003? 2004? and included ones about experiences of war.

It was in a statistics class. Every so often the prof would comment that we were about to work through data taken from innately painful events - war, cancer, discrimination, poverty, etc. I forget his phrasing, but i remember that it landed for me both as someone with the day’s painful experiences and later as someone so *far* from it that i was grateful for the reminder not to ignore the real side.

Why? W-hat? Y-hat.
posted by clew at 8:28 AM on May 24 [5 favorites]


"Dead Boy Detectives" had unexpectedly dark themes for a teen-drama type show. Like the writers had a choice of emotionally gripping topics / themes / backstories to include, and they picked the most hardcore thing they could think of every single time.
posted by subdee at 9:28 AM on May 24 [1 favorite]


The obvious solution seems like site-wide settings to show or not show content warnings. In a classroom you could have it listed on the syllabus or somewhere people don't have to look.

I have wondered how common it is for trigger warnings to be counterproductive. For me it is obvious that the entire category of child abuse is more distressing than a specific instance of it that shares little in common with your experience, so bringing up the former before the latter seems weird. But I could see it being the other way around.
posted by hermanubis at 9:29 AM on May 24


« Older "a peculiarly British disease which we aim to...   |   Out Of The Shadows, Into The Spotlight [or so they... Newer »


You are not currently logged in. Log in or create a new account to post comments.