I think social media undoubtedly plays a monstrous role in all this. But I wonder whether other elements, such as what could have happened in these kids’ earlier years—perhaps something they were too young to handle (9/11?)—might’ve laid the foundation and made these children more vulnerable.
Maybe I'm way off here.
No matter the foundation, if many adults can't use social media responsibly (a vague term, I know), why allow our children to use it?
It would be interesting to see if the trend is also reflected in homeschoolers. This might be one sub-group of the population that is less likely to use social media/ or use it to a lower extent. There is less peer pressure, closer ties to family, and a smaller group of closely connected friends. I coordinate co-ops for homeschooled high school students and have observed that phones rarely play any role in their interactions. A group of them actually decided to delete all their social media apps on their phones; they have kept it up and encourage each other. They have decided to use their phones only for making arrangements or taking photos. Many of them have clearly decided that they prefer to be grounded in reality rather than lost in space.
Feb 16, 2023·edited Feb 16, 2023Liked by Jon Haidt
Thank you for all your work, professor! I’ve been going around preaching your and twenge’s books to my community for years (proof here, my very long talk I used to give pre-Covid: , https://gaty.substack.com/p/unplugging-your-kids ), absolutely invaluable insights.
Anyway, I’ll write about this in more detail later, but did you know that even with the end of lockdown, kids are still doing virtual school, just in person? That is, they go to school, but spend all their classes on tablets watching videos iand virtual slide decks. One of my patients told me this a couple months ago and I thought he was exaggerating but since then I’ve encountered the same story without prompting several other times! And at least one kid could easily still access YouTube during class so was fooling around instead of following along. And then the teacher gets mad and sends them to me for “adhd” - and I’m like, y’all are the ones giving them screens all day!! I don’t think this will end well…
For anyone interested in a non-screen cause of teen anxiety, as demonstrated by a recent survey on parent priorities, please see here:
Thanks for this, and especially the timely response to critiques. The timeline before 2000 was one of my own curiosities. Glad someone else in the class was willing to raise their hand and ask the question. The Looping Effect is one of the things I seem to be observing and attempting to counteract within our teen case study of 2.
There are 3 major achievements to be made during adolescence and early adulthood. One is forming a sense of identity, separate from one's family, often beginning with group membership. A second is the capacity for emotional intimacy with peers. But a third, often overlooked, is learning to manage affective information.
I distinguish this from managing affect itself. One of the challenges young people face is being able to set aside how something makes them *feel*, whether positive or negative, and not treat it as pivotal or deciding information when reasoning through decisions. For me, this difficulty/challenge is fundamental to why teens and young adults are not only more prone to *take* risks, but also more prone to be *risk-averse* in many situations: they are convinced by their feelings. So, they might avoid some action, out of great fear of the imagined consequences, and adopt a different riskier course of action, because it creates the impression of greater safety.
Flipping this around, face with an adolescent who could say "You know, I really don't like the guy, BUT he makes a good point", or "It sounds like it would be a lot of fun, but I have an important exam tomorrow, so I'll have to bail out", we would perceive them as "mature", BECAUSE they show evidence of being able to separate feelings from decision-making. I would contend that where you can talk a 40 year-old down off a ledge, it is substantially more difficult to persuade a 16 year-old that their perception of things being hopeless is wrong. For those who have not yet learned to manage how they incorporate affective information, what they feel in their gut makes for an open and shut case, impervious to other information.
I mention this because one of the hallmarks of what I refer to as "the adolocentric society" is the increasing emphasis on affective information as important and even crucial. Heck, we even see it in the field of management where people are encouraged to "go with their gut". But, as Daniel Kahneman has amply illustrated, decisions made on the basis of those first 300msec of "feelings" are generally not the best decisions.
The emergence of mobile devices that permit us to respond immediately and impulsively, without reflection (I submit sampling Youtube comments as evidence!), and be impacted on by an onslaught of messages and images explicitly intended to drive our affect, is not solely responsible for the declines in indices of mental health noted in the CDC report, and generally discussed here. But it sure as shooting hasn't helped, and substituting actual social interaction with curated indirect contact, via smartphones, during the pandemic, has only amplified this trend further. I think, however, the roots of this go back to the early '80s.
Feb 16, 2023·edited Feb 16, 2023Liked by Jon Haidt, Zach Rausch
Great post. Since lead is posited as a cause of the peak in boys suicidality in the 90s, any plausible similar explanation for the recent peak that has not yet been ruled out? I was a bit surprised when you suggested that girls fared worse during the pandemic because they dove into social media. To me the more plausible explanation is that girls are more social and experienced more sadness at not being as connected to their friends in real life. I know extroverted adults who experienced the same (as an introvert, I suffered very little ;)). Also, I keep wondering about to what extent these patterns are specific to specific groups/demographics. Maybe you have covered that somewhere already, but I think generalizing to boys and girls broadly makes it seem more plausible that it's a mass phenomenon (social media) vs. something else affecting only a subset of boys/girls but nevertheless increasing the rates. Finally, I suspect that at least for some of this it is a measurement artifact. Best to calibrate evaluations by the quality of the evidence (objective indicators of poor mental health > subjective).
A partial critique of the answer to question one: Even if teens' time spent hanging out with friends in the strictest sense (outside of school settings) had already been decimated over the pre-Covid decade, wouldn't the transition to virtual learning, including the falling time spent in-person with clubs and sports, have still reduced the amount of face-to-face time with peers even further? That was my experience being in high school during the pandemic: even though I hadn't previously spent loads of time just "hanging out" at friends' houses, the year+ on virtual learning cut out that time I had eating lunch with friends, bantering in the hallway or on the bus, meeting people at after-school clubs, telling funny stories while working on a Photoshop etc. etc. and that was just awful for me. Indeed, the authors of that study note that "social baseline theory suggests, at a minimum, being in relatively close proximity to others imparts physiological benefits." Seems important to note that, even though friend time fits into the pre-Covid trend, if you look at the 15-24 line on the "social isolation" graph, it had previously been going up especially in the preceding few years, but went up from 2019 to 2020 by more than it had in the total 2012 to 2019 span.
I'll admit that the sample of Gen-Zers who read stuff like this probably enjoyed school much more than the average member of our generation, though, and there is evidence that youth suicides go up during the school year and on school days. There was a lot less work when school was online, so maybe the reduction of schoolwork time had a (short-term) positive effect that partially masked the pernicious effect of reduced face-to-face time at school, extracurriculars and sports. But it all depends on what it got replaced with—is schoolwork time worse for the average high schooler (in the short-run) than YouTube time? Instagram/TikTok time? I don't know.
This is fascinating. So much was made of covid's impact, and I think the context of the broader downward spiral of mental health prior to covid has always been left out. Thank you for bringing clarity. It's important for two reasons: 1) If teen mental health decline was related to covid, it can be easily dismissed as something that was extremely unlikely to happen and will likely not happen again. 2) Because of #1, no one would do anything about it. But if it's viewed as being a continuation of a downward trend with no end in sight, then maybe the issue will get some attention. Maybe.
Feb 16, 2023·edited Feb 17, 2023Liked by Jon Haidt
I really appreciate you sharing the passage in Haidt's book about The Looping Effect. It's something I got caught in myself when I was diagnosed with Depression, Anxiety, and Asperger's in middle school (I'm 31 now).
While the answers to my troubles could now be explained under these new labels, at the same time it pigeonholed me into an identity I had little desire to change. Every mental problem in my life could now be easily explained away by the "chemicals in my brain". My concern is while diagnoses can give answers to a person's suffering, I see less effort being placed toward inspiring people to move past their diagnoses to become stronger and wiser. While the stigma of mental illness is decreasing, it also seems to be having a paradoxical effect in some circles where it's more popular now to have some diagnosis than none at all, which in turn further entrenches people in mental illness identities.
These days I barely mention my diagnoses, but when I do I like to say: Labels are an Explanation, not the Destination. I Accept the Diagnosis, but I Refuse the Verdict. For anyone reading this who can relate: please don't let your psychiatric diagnoses define you as an individual!
If I connect your very pertinent observation to the Italian situation it makes sense this side of the Ocean because there is a DELAY IN ADOPTION so unlike their American peers Italian teenagers still spent time with friends face to face and many millions of them either had restriction in the use of phone or no smartphone before COVID19. I interviewed many mothers across Italy this summer and it is clear that COVID19 not only increased exponentially the distribution of smartphone to teens but it also lead giving smartphone to younger children (7-10) with awful consequences.
In an effort to contribute to your "crosscultural study" I take this opportunity to share the first study that establish a causal link (an Italian Professor from Bocconi is co-author) https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3919760
I strongly support your exploration of the 27 different European Countries (because Europeans only exists in the mind of Americans -sorry but reality matters) It will be easier to prove the relation between screen introduction in teens life and mental health deterioration because if it happens simultaneously in 27 different cultures that have different 27 ways of life some of them highly centred on relationships and some less, some with more family participation some less... well there is no other way to explain the phenomenon but with smartphones.
Again thank you for your inspiring intellectual courage.
Honest question about that chart on "Time Spent with Friends": so, since it's tracking cohorts based on *current* age (which is why Gen Z had tons of time with friends a decade ago -- they were children), how do we know that sharp decline isn't just a fairly normal transition for mid-teens -> early thirties?
Edit: in other words, is there a way to see that chart while stabilizing the ages? like, "here's how much time 15-year-olds in '22 spent with friends, vs. how much time 15-year-olds in '12 spent with friends" (rather than vs. the same cohort as 5-year-olds in '12).
So glad that you guys got around to answering some of the questions raised! Another issue I saw come up in a few questions that wasn't addressed is what if any correlation there is with all the gender identity and trans issues that have been increasingly raised in recent years. I'm not sure exactly when it started, but a cursory search brings up links to stories about pronouns and bathrooms from 2016. I notice in the graphs shown that the uptick in boy suicides is particularly notable from 2016.
I realize this is a huge can of worms and people here will likely fall on both sides of the actual issue itself, but it seems to me pretty uncontroversial to say that teens identifying as LGBTQ+ are at higher risk of suicide, so more young people identifying as such would logically mean more suicides.
One source puts the LGBTQ+ suicide rates at 4 times that of the general population.
It's really interesting that the big drop-off in time spent with friends happens at the same time as the rise of smartphones and social media, because I assumed teens adopted these technologies because of in-person time they were already missing, rather than the reverse where adoption displaced in-person time. Even before smartphones, it seemed to me that in-person friend time had declined hugely since my childhood for a variety of reasons.
42% of high schoolers experiencing persistent feelings of sadness or hopelessness is so unbelievably striking and bleak.
I think social media undoubtedly plays a monstrous role in all this. But I wonder whether other elements, such as what could have happened in these kids’ earlier years—perhaps something they were too young to handle (9/11?)—might’ve laid the foundation and made these children more vulnerable.
Maybe I'm way off here.
No matter the foundation, if many adults can't use social media responsibly (a vague term, I know), why allow our children to use it?
It would be interesting to see if the trend is also reflected in homeschoolers. This might be one sub-group of the population that is less likely to use social media/ or use it to a lower extent. There is less peer pressure, closer ties to family, and a smaller group of closely connected friends. I coordinate co-ops for homeschooled high school students and have observed that phones rarely play any role in their interactions. A group of them actually decided to delete all their social media apps on their phones; they have kept it up and encourage each other. They have decided to use their phones only for making arrangements or taking photos. Many of them have clearly decided that they prefer to be grounded in reality rather than lost in space.
Thank you for all your work, professor! I’ve been going around preaching your and twenge’s books to my community for years (proof here, my very long talk I used to give pre-Covid: , https://gaty.substack.com/p/unplugging-your-kids ), absolutely invaluable insights.
Anyway, I’ll write about this in more detail later, but did you know that even with the end of lockdown, kids are still doing virtual school, just in person? That is, they go to school, but spend all their classes on tablets watching videos iand virtual slide decks. One of my patients told me this a couple months ago and I thought he was exaggerating but since then I’ve encountered the same story without prompting several other times! And at least one kid could easily still access YouTube during class so was fooling around instead of following along. And then the teacher gets mad and sends them to me for “adhd” - and I’m like, y’all are the ones giving them screens all day!! I don’t think this will end well…
For anyone interested in a non-screen cause of teen anxiety, as demonstrated by a recent survey on parent priorities, please see here:
https://gaty.substack.com/p/the-good-portion-a-warning-to-parents
Thanks for this, and especially the timely response to critiques. The timeline before 2000 was one of my own curiosities. Glad someone else in the class was willing to raise their hand and ask the question. The Looping Effect is one of the things I seem to be observing and attempting to counteract within our teen case study of 2.
There are 3 major achievements to be made during adolescence and early adulthood. One is forming a sense of identity, separate from one's family, often beginning with group membership. A second is the capacity for emotional intimacy with peers. But a third, often overlooked, is learning to manage affective information.
I distinguish this from managing affect itself. One of the challenges young people face is being able to set aside how something makes them *feel*, whether positive or negative, and not treat it as pivotal or deciding information when reasoning through decisions. For me, this difficulty/challenge is fundamental to why teens and young adults are not only more prone to *take* risks, but also more prone to be *risk-averse* in many situations: they are convinced by their feelings. So, they might avoid some action, out of great fear of the imagined consequences, and adopt a different riskier course of action, because it creates the impression of greater safety.
Flipping this around, face with an adolescent who could say "You know, I really don't like the guy, BUT he makes a good point", or "It sounds like it would be a lot of fun, but I have an important exam tomorrow, so I'll have to bail out", we would perceive them as "mature", BECAUSE they show evidence of being able to separate feelings from decision-making. I would contend that where you can talk a 40 year-old down off a ledge, it is substantially more difficult to persuade a 16 year-old that their perception of things being hopeless is wrong. For those who have not yet learned to manage how they incorporate affective information, what they feel in their gut makes for an open and shut case, impervious to other information.
I mention this because one of the hallmarks of what I refer to as "the adolocentric society" is the increasing emphasis on affective information as important and even crucial. Heck, we even see it in the field of management where people are encouraged to "go with their gut". But, as Daniel Kahneman has amply illustrated, decisions made on the basis of those first 300msec of "feelings" are generally not the best decisions.
The emergence of mobile devices that permit us to respond immediately and impulsively, without reflection (I submit sampling Youtube comments as evidence!), and be impacted on by an onslaught of messages and images explicitly intended to drive our affect, is not solely responsible for the declines in indices of mental health noted in the CDC report, and generally discussed here. But it sure as shooting hasn't helped, and substituting actual social interaction with curated indirect contact, via smartphones, during the pandemic, has only amplified this trend further. I think, however, the roots of this go back to the early '80s.
Great post. Since lead is posited as a cause of the peak in boys suicidality in the 90s, any plausible similar explanation for the recent peak that has not yet been ruled out? I was a bit surprised when you suggested that girls fared worse during the pandemic because they dove into social media. To me the more plausible explanation is that girls are more social and experienced more sadness at not being as connected to their friends in real life. I know extroverted adults who experienced the same (as an introvert, I suffered very little ;)). Also, I keep wondering about to what extent these patterns are specific to specific groups/demographics. Maybe you have covered that somewhere already, but I think generalizing to boys and girls broadly makes it seem more plausible that it's a mass phenomenon (social media) vs. something else affecting only a subset of boys/girls but nevertheless increasing the rates. Finally, I suspect that at least for some of this it is a measurement artifact. Best to calibrate evaluations by the quality of the evidence (objective indicators of poor mental health > subjective).
A partial critique of the answer to question one: Even if teens' time spent hanging out with friends in the strictest sense (outside of school settings) had already been decimated over the pre-Covid decade, wouldn't the transition to virtual learning, including the falling time spent in-person with clubs and sports, have still reduced the amount of face-to-face time with peers even further? That was my experience being in high school during the pandemic: even though I hadn't previously spent loads of time just "hanging out" at friends' houses, the year+ on virtual learning cut out that time I had eating lunch with friends, bantering in the hallway or on the bus, meeting people at after-school clubs, telling funny stories while working on a Photoshop etc. etc. and that was just awful for me. Indeed, the authors of that study note that "social baseline theory suggests, at a minimum, being in relatively close proximity to others imparts physiological benefits." Seems important to note that, even though friend time fits into the pre-Covid trend, if you look at the 15-24 line on the "social isolation" graph, it had previously been going up especially in the preceding few years, but went up from 2019 to 2020 by more than it had in the total 2012 to 2019 span.
I'll admit that the sample of Gen-Zers who read stuff like this probably enjoyed school much more than the average member of our generation, though, and there is evidence that youth suicides go up during the school year and on school days. There was a lot less work when school was online, so maybe the reduction of schoolwork time had a (short-term) positive effect that partially masked the pernicious effect of reduced face-to-face time at school, extracurriculars and sports. But it all depends on what it got replaced with—is schoolwork time worse for the average high schooler (in the short-run) than YouTube time? Instagram/TikTok time? I don't know.
This is fascinating. So much was made of covid's impact, and I think the context of the broader downward spiral of mental health prior to covid has always been left out. Thank you for bringing clarity. It's important for two reasons: 1) If teen mental health decline was related to covid, it can be easily dismissed as something that was extremely unlikely to happen and will likely not happen again. 2) Because of #1, no one would do anything about it. But if it's viewed as being a continuation of a downward trend with no end in sight, then maybe the issue will get some attention. Maybe.
I really appreciate you sharing the passage in Haidt's book about The Looping Effect. It's something I got caught in myself when I was diagnosed with Depression, Anxiety, and Asperger's in middle school (I'm 31 now).
While the answers to my troubles could now be explained under these new labels, at the same time it pigeonholed me into an identity I had little desire to change. Every mental problem in my life could now be easily explained away by the "chemicals in my brain". My concern is while diagnoses can give answers to a person's suffering, I see less effort being placed toward inspiring people to move past their diagnoses to become stronger and wiser. While the stigma of mental illness is decreasing, it also seems to be having a paradoxical effect in some circles where it's more popular now to have some diagnosis than none at all, which in turn further entrenches people in mental illness identities.
These days I barely mention my diagnoses, but when I do I like to say: Labels are an Explanation, not the Destination. I Accept the Diagnosis, but I Refuse the Verdict. For anyone reading this who can relate: please don't let your psychiatric diagnoses define you as an individual!
Grazie mille Professore! Thank you so much.
If I connect your very pertinent observation to the Italian situation it makes sense this side of the Ocean because there is a DELAY IN ADOPTION so unlike their American peers Italian teenagers still spent time with friends face to face and many millions of them either had restriction in the use of phone or no smartphone before COVID19. I interviewed many mothers across Italy this summer and it is clear that COVID19 not only increased exponentially the distribution of smartphone to teens but it also lead giving smartphone to younger children (7-10) with awful consequences.
In an effort to contribute to your "crosscultural study" I take this opportunity to share the first study that establish a causal link (an Italian Professor from Bocconi is co-author) https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3919760
And the data from a study on Italian children during Covid https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3919760 -
This study from JAMA https://pubmed.ncbi.nlm.nih.gov/31305878/ (you probably have it already) but just in case.
I strongly support your exploration of the 27 different European Countries (because Europeans only exists in the mind of Americans -sorry but reality matters) It will be easier to prove the relation between screen introduction in teens life and mental health deterioration because if it happens simultaneously in 27 different cultures that have different 27 ways of life some of them highly centred on relationships and some less, some with more family participation some less... well there is no other way to explain the phenomenon but with smartphones.
Again thank you for your inspiring intellectual courage.
Honest question about that chart on "Time Spent with Friends": so, since it's tracking cohorts based on *current* age (which is why Gen Z had tons of time with friends a decade ago -- they were children), how do we know that sharp decline isn't just a fairly normal transition for mid-teens -> early thirties?
Edit: in other words, is there a way to see that chart while stabilizing the ages? like, "here's how much time 15-year-olds in '22 spent with friends, vs. how much time 15-year-olds in '12 spent with friends" (rather than vs. the same cohort as 5-year-olds in '12).
I would love to see further back in time for figure 4, is that data available?
So glad that you guys got around to answering some of the questions raised! Another issue I saw come up in a few questions that wasn't addressed is what if any correlation there is with all the gender identity and trans issues that have been increasingly raised in recent years. I'm not sure exactly when it started, but a cursory search brings up links to stories about pronouns and bathrooms from 2016. I notice in the graphs shown that the uptick in boy suicides is particularly notable from 2016.
I realize this is a huge can of worms and people here will likely fall on both sides of the actual issue itself, but it seems to me pretty uncontroversial to say that teens identifying as LGBTQ+ are at higher risk of suicide, so more young people identifying as such would logically mean more suicides.
One source puts the LGBTQ+ suicide rates at 4 times that of the general population.
https://www.thetrevorproject.org/resources/article/facts-about-lgbtq-youth-suicide/
I've also found data showing that the LGBTQ+ community is growing, and that nearly 1 in 5 identifying as transgender are aged 13-17.
https://thehill.com/changing-america/respect/diversity-inclusion/585711-us-lgbtq-population-hits-20-million/
https://williamsinstitute.law.ucla.edu/publications/trans-adults-united-states/
It's really interesting that the big drop-off in time spent with friends happens at the same time as the rise of smartphones and social media, because I assumed teens adopted these technologies because of in-person time they were already missing, rather than the reverse where adoption displaced in-person time. Even before smartphones, it seemed to me that in-person friend time had declined hugely since my childhood for a variety of reasons.