I'm Being Trained and So Are You - Barry's Economics

I am being trained, and so are you. So, about a month ago, I posted a video that got a quarter of a million views in 2 days. And I know that in YouTube terms, those numbers put me somewhere between like a regional story about, I don't know, flooding and a poodle that fell off a surfboard. But, to me, it was real. But, ever since, there's been a deeply uncool part of my brain that just keeps going to me obsessively, "Do it again, Barry. Do it again. Make Make the machine go bang bang. Maybe a hotter take. Have you Have you got a hotter take? Maybe something hot?" It's something that I recognize that's built into this system by people who are very good at their jobs. Because right now, in Silicon Valley, there's a man in flip-flops and a clicker that's literally just saying out loud to me, "Good boy. And you're a good boy. Yes, you are. Get more views. You carry on doing it. Yes, you are. You're a good boy." while sipping at Huel. And it's the same manipulations if you're watching videos. Your timeline, your news feed, your Reels feed. I am being trained, and so are you. So, over the next, however long this video is, I want to show you the science of exactly how that training works and what it does to people that are caught inside of it, like us. In 1938, a psychologist called B. F. Skinner put a rat in a box. And that doesn't sound extraordinary, but it changed the way we see the world. The box had a lever. When the rat pressed the lever, food appeared. The rat learned to press the lever because even rats understand the principles of Deliveroo. But, Skinner didn't just test one way of rewarding the rats, right? He tested several ways. Sometimes, the food came every single time the rat pressed the lever. Sometimes, it came every fifth press. Sometimes, it came randomly with no pattern whatsoever. And he measured which schedule produced the most obsessive-compulsive, impossible-to-stop lever pressing. And the answer was, random rewards. What Skinner called a variable ratio reinforcement schedule. Because when the rat never knew which press would deliver the food, when the reward was unpredictable, it just kept pressing endlessly, frantically. It couldn't stop. Even when the food stopped coming entirely, press press, press press press. Skinner himself was reportedly disturbed by how effective this type of reward was. And the research was so reliable, so repeatable, so devastating in its implications, that he spent years worrying about what it meant for human society. And to be fair, he should have worried more. Cuz the rat's lever only dispenses food, right? Cut to 2026, and our levers dispense food, status, validation, a sense of meaning, and occasional abuse from strangers. Oh, and if you're lucky, very lucky, a brand deal from a mattress company. Because if you post a video, sometimes it gets 2,000 views, sometimes 8,000, and sometimes, if the algorithmic gods are feeling generous, 400,000 views, a million views. And you'll never know which press is going to deliver the food. You'll never know why one video flies and another video dies. You just keep pressing. Like, I've spent hours trying to reverse engineer what makes videos fly. And the best idea I have at the moment is phases of the moon. And it's the [clears throat] same manipulations if you're watching videos. Your timeline, your news feed, your Reels feed. Sometimes, it's a great video, sometimes it's one you don't like, but you keep scrolling for that hit, the one that really gets you, the one that makes you laugh, the one that entertains you, that makes you forget about things, that gives you something to think about, that just hit of dopamine. Because we are all the rat. Instagram, TikTok, YouTube. They're not platforms. They are Skinner boxes with better graphic design, but a much worse privacy policy. >> The main thing is what we what we call schedules of reinforcement. Reinforcement is what the layman calls reward, and you can schedule it uh so that a reward occurs every now and then when a pigeon does something. We usually use a response with a pigeon pecking a little disc at a little spot in the wall. You don't reinforce every time, you're every perhaps every tenth time, or perhaps only once every minute or something like that. And there is a good example of how you can move from uh the uh the pigeon to the human case because one of the one of the schedules is very effective with with rats or pigeons is what we call a variable ratio schedule, and that is at the heart of all gambling devices and has the same effect. The pigeon can become a pathological gambler just as a person can. People gamble because of the schedule of the reinforcement that follows. And this is true of all gamblings as they all have variable ratios built into them. So, what we've learned from the pigeon, it made it possible to interpret this vast field very effectively. Because the variable ratio reinforcement schedule that Skinner identified in 1938 as the single most addictive pattern ever discovered in behavioral science is now the business model of every major social media company on Earth. But, they have way more data to design that lever and make it even more addictive. And that's what they do to viewers. But, what they do to creators, huh, [sighs] is even more sadistic. >> [music] >> So, Nicholas Perry was a classically [music] trained violinist, soft-spoken, thoughtful, vegan. He dreamed of Broadway. Look at him there. Isn't he a sweet-looking lad? The hope in those eyes. >> [snorts] >> So, in 2014, >> [music] >> he started a YouTube channel sharing recipes and playing music, like gentle, wholesome stuff. You know, green smoothies [music] and violin pieces. And of course, nobody watched it. Why would you watch that? All right, Barry. Then one day, just for a joke, he tried something called mukbang. >> Well, hello, my little sloth. Sit down. We're going to eat together. I'm going to think happy thoughts right now and show you >> [sighs and gasps] >> the food. So, which is a genre of video originating from South Korea where a creator eats a huge quantity of food on camera while talking. And his first mukbang got 50,000 views in 2 weeks, more than he'd ever gotten before combined. The Skinner lever had delivered food. Literally, in this case. >> Sushi. The algorithm had said, essentially, "Mate, put the violin down. What if, instead of playing Mozart or something, you ate a whole family bucket of KFC while having a breakdown?" So, what do you think happened next? >> This video's going to be really interesting, but it is what it is. This is my life. And my life is what it is. If people don't like it, they don't have to click it. You clicked it, so it means you're somehow interested. Why? I don't know. Over 8 years, Nicholas Perry gained over 250 lb. He developed sleep apnea. He fractured his ribs. He became disabled and bought a mobility scooter. He filmed himself crying and eating 20,000 calorie meals three times a day across five separate YouTube channels. >> [crying and screaming] >> Subscribe right now. I'm watching you, and I'll I'm literally I'll get notified when people subscribe here. So, I'll be able to see Oh my gosh, this is a not enough food. This is not enough food. I'm going to starve. And here's a direct quote from him that should keep every single creator awake at night. "They like when I'm sick. They like when I'm upset, and they like when I'm hyper, so I just give them that." He knew. He understood the mechanism perfectly. He could describe it in plain English, and he still could not stop. Because knowing you're in a Skinner box doesn't get you out of the Skinner box, as anyone with a phone will tell you. Sorry. Oh, sorry. I'll put that down. I won't be sick. Oh, sorry. Yeah, this one. I I won't touch it. Um yeah, I won't be sick. Yeah, but sorry. Well, I think it's Yeah, >> [snorts] >> all good. Oh, uh actually, Yeah. Okay, cool. Got that. I'll put it away. Yeah, something like that. I'll probably put it away. I actually there's one more thing I'll say. Yep. Yep, cool. Yep. Sorry. Sorry, yeah. Uh oh, sorry. Yeah, I should um Cuz the reinforcement schedule, it doesn't care whether you understand it. It works anyway. That's what made Skinner so absolutely petrified of his own research. Like, it's not that people were weak or that easy to condition. It works absolutely regardless of intelligence, regardless of insight, and regardless of self-awareness. That's what makes it terrifying. So, why can't creators see the change happening in themselves? If the Skinner box is making them seem more and more extreme, or more and more chasing the algorithm, or getting overweight, or just literally, "Oh, suddenly I'm in a mobility scooter." Am I making these choices? So, what is happening to themselves? In 2012, Diana Tamir and Jason Mitchell at Harvard ran five studies published in the Proceedings of National Academy of Sciences to answer a deceptively simple question about why do humans spend so much time talking about themselves. And we really do. Research shows that we devote 30 to 40% of everything we say to telling other people about our own experiences. Like on social media it's even higher. Upwards of 80% of posts are just announcement about what people are thinking or feeling right now. Awks, which if you think about it is remarkable, right? Cuz nobody asked. Nobody needs to know and yet we can't stop. So Tamir and Mitchell put people in an fMRI brain scanner and gave them a choice. Answer questions about yourself, your opinions, your personality, what you think about things, or answer questions about someone else. And they watched what happened in the brain. When people talked about themselves, two specific brain regions lit up. The nucleus accumbens and the ventral tegmental area. And those aren't just random structures. They're the core of the mesolimbic dopamine system. The very same reward pathway activated by food and sex and by addictive drugs. The brain was treating self-disclosure, the act of sharing what you think and feel as a reward on the same neurological level as eating when you're hungry. But here's the detail that really makes it interesting. In a follow-up experiment, they gave participants a choice. Answer questions about yourself or answer questions about someone else, but they attached different amounts of money to each option. Sometimes you get paid more to answer about someone else. And what they found was that people were willing to forego money for the opportunity to talk about themselves instead. The brain valued self-disclosure so highly that it would sacrifice financial reward to get it. Now think about what happens when you post a video and a million people watch you share your thoughts. Every signal to the brain is saying people are listening to you talk about yourself. It's It's applause without eye contact, right? Which is psychologically confusing but quite addictive. That is the nucleus accumbens firing not once but thousands of times across days and weeks. You're not just getting attention, you are getting a neurochemical reward on a scale that Tamir and Mitchell's participants in a quiet lab in Harvard could never have imagined. But the dopamine doesn't just make you feel good. It actually makes you less able to detect change in your own brain. Phil Reed, a professor of psychology at Swansea University, has written about a concept he calls dopamine overdose. Because here's what most people think dopamine does, right? It makes you feel good, reward, pleasure. And that's the popular version, but it's not actually its primary function. Dopamine's core job in the brain is to signal change. It helps you detect something in your environment that's shifted, something new that's happening, novelty, something that's different from what you expected. It's less about pleasure and more about noticing. That detection system works by contrast. You notice small increase in dopamine against a low baseline because not that many things in your environment are meant to spark dopamine. The way that you notice a candle being lit in a dark room. But when the baseline is already high, when you're already flooded with dopamine from constant notifications, reward signals, all the small things that come out as the small changes become invisible. It's like trying to spot a candle in the middle of day. The signal's there, but you can't see it against the noise. And Reed points to studies showing that artificially increased dopamine levels during discrimination tasks, tasks where you have to unlearn something old and learn something new, actually reduces your ability to learn the new thing. Too much dopamine makes you worse at detecting that something has changed. Now, apply that to a content creator in the middle of a growth spike. You're getting dopamine hits from every direction, views, subscribers, comments, shares. Your baseline is elevated and the very thing that elevated baseline impairs is your ability to notice change. Specifically, notice that you are changing. Your content is shifting. Your tone is shifting. Your opinions are shifting, drifting towards whatever the audience rewards. But the detection system that would normally flag this, "Wait, uh something's changing. I'm not sure you actually believe what you're saying there." The information about self is totally drowned out by the reward signal. "Ah, the personality alarm's gone off there. Something's changing." Because of dopamine, it doesn't just feel good. It makes you blind to your own drift. "Oh, this is awkward. Okay, you're just going to do it anyway. I guess the reward's big. Okay." Which means, right? That the very people that are most at risk from being changed by the algorithm are precisely the people that are experiencing the most success. The bigger the spike, the higher the baseline. Then the harder it is to see yourself clearly. Because the algorithm rewards certain content, right? Therefore, the creator makes more of that content and the dopamine reward makes the creator less able to notice that they're changing. And the audience reinforces that direction because that's what they get served. They don't know anything else. Engagement. Things that make them feel more engaged and more likely to stay on the platform. Hey, hey. And that reward structure then reshapes the creator's belief to match that behavior. Until the creator is suddenly, well, I don't know, way more extreme. Well, way more left-wing, way more right-wing. Just anything to get more views, right? But at no point does anyone feel manipulated. Because it all just feels like choice. It feels like free choice. It feels like I don't know, authenticity. Um it just feels like I'm just saying what I believe, man. I'm a I just love the system. Which is, when you think about it, right? Exactly what a perfectly designed conditioning system would look like from the inside, right? Which means, well, I think it means I need to be transparent with you. Because because this is what my analytics are showing me right now. In the last 28 days, my channel has received a million views and a lot of those views have come from the Epstein videos that I made. They're responsible for almost all of it. The algorithm's message is really clear. More Epstein, make it make more Epstein, more more big big bang bang. And I can feel that working on me. I genuinely can. I check my analytics more than I should. I think in titles in terms of what will get clicks rather than what's true. I'm not saying I act on it, but I do. I'm a rat. You are a rat, mate. That's right. You are a rat. That's why I act like a rat. So how do I not become Nicholas Perry? And this is the only way I think is possible and that's by stating clearly why I'm doing these videos so that you keep me honest to them. I'm not posting them because my echo chamber agrees with me. Posting them because of what the peer-reviewed evidence actually says. And I don't see that evidence out there. And what is the evidence that I think is most important to get out there? For example, a study published in the Journal of Experimental Psychology in 2014 showing that experience of power literally damages the brain's ability to feel empathy. And yet we're building structures where people have power. Or the study by Kraus, Piff, Mendoza, Denton, and Keltner that showed in Psychological Review that wealth creates what they called solipsism, a first-person-only view of the world. The richer you get, the less you see other people. Not because you're evil, but because your environment stops you requiring to see other people's point of view. That is baked into our system and nobody's talking about it. Wilkinson and Pickett in Spirit Level in 2009 analyzed data across dozens of countries and showed that the more unequal the society, the worse the outcomes on virtually every single measure of health, crime, trust, mental illness, life expectancy, social mobility. And not just for the poor, for everyone including the rich. Inequality doesn't just hurt the people at the bottom, it degrades the entire system. And Piketty showed in Capital in 21st Century clearly that when returns on capital exceed economic growth, which they almost always do at the moment, wealth concentrates inevitably. Not because of individual greed, but because of mathematics. And these views, they're not left-wing, they're not right-wing. This is arithmetic and neuroscience. A wealth tax is not a political position. It's what the evidence says is necessary if you want to live in a society that doesn't keep manufacturing the conditions that produce Epsteins, Epsteins, and more Epsteins and the systems that protect them. Staying quiet about provable, peer-reviewed, repeatable facts because I might be afraid of being called political or because I'm afraid of the algorithm or because I'm afraid of the negative comments. Staying silent is a political act. Silence protects the system every time. So yeah, the algorithm is trying to change me and I can feel it working and I'm going to keep posting anyway, not because of the algorithm, but because of despite it. Not because my echo chamber agrees with me, but because the evidence is the evidence is the evidence is the evidence whether anyone agrees with it or not. And I hope the difference between me and the rat is not that I'm immune to the lever. I'm not. Nobody is. The difference I'm hoping is that I know why I'm pressing it and it's not for the food. So if you find value in the channel, you can support the community at Angel Comedy and there's a link to Patreon in the description. Okay, and I'll see you next Sunday. Thanks for watching. It's undoubtedly true that what we might call the literature of freedom has aggrandized the individual. You are free, your values are the only values and anyone who wants to change them is your enemy and so on. Well, I think we are in real trouble because the whole notion of individual freedom is miscarrying very badly and is doing great damage to the individual. I I think to put it very in very broad terms, I think what a culture does for the individual is to bring him under the control of the remoter consequences of his behavior. What what will happen 10 years from now because of what he does today. And if he then rejects the culture and said I am the authority as to what I am going to do today, he loses contact with with the future. I think the great danger in America right now is that we are not behaving with respect to the remoter consequences of our behavior and other cultures are and that could be a very serious thing. Well, great. I love chicken. Great. Oh.