SIO510: Digging Into the Science of the Placebo Effect

Transcript

This transcript was generated automatically and may contain errors — the audio is the record.

0:00:13 Thomas: Hello and welcome to Serious Inquiries Only. This is episode 510. I'm Thomas and I'm joined today by Jenessa. How are you doing? I stumped you. I stump you with that question. It snowed? That's crazy.

0:00:21 Jenessa: I'm doing pretty good. Yeah, I don't have anything for you today. Things are good. You did. I'm trying not to complain about the fact that it snowed today. It snowed. This winter is ruining my life, but I'm trying not complain about weather for the hundredth time. It's supposed to warm up tomorrow and then I will be out of my cranky winter blues. I am going to insist on wearing a dress in flip-flops even if it freezes again. I don't care. Probably not based on

0:00:49 Thomas: Well, anyway, over here I can experience no snow, but if I think it's snowing, will it have the same effect? Probably not.

0:01:02 Jenessa: What we're going to talk about today. Still probably no. Yeah, I wanted to build off of

0:01:03 Thomas: Oh wow. But what if, okay, what if I take these sugar pills and do that? Well, then we'll be better. Yeah, we're going to talk about the placebo effect. I'm sure a lot of people listening have probably heard a bit on this. Recently, there was a citation needed on it, but we're gonna go a little more, I think, in depth into the science of it. What's your plan for today? Right. Interesting. Wow. Well, it is pretty crazy if you are someone who hasn't heard

0:01:28 Jenessa: That. I think that episode they did a great job talking about some of it can be debunked very simply as this is just made up. You know, some of the things you've heard about like, oh, if it's a red pill versus a blue pill, you might get this or like take two. And And some of that seems to be either just completely invented or like not replicable, whatever, you know, the ways that charlatans have latched on to this, the way that it can really harm people to believe in this. What I want to build on to kind of yes and is what is happening in a placebo group then? Like we do see something. So like why? If we're going to say placebos don't work, well, then why did I bother with a placebo group? What's happening there statistically? What's happening there with our scientific methods? And then I got a bit for you at the end about there might be just a tiny little bit that kind of is having some experience with meds could make a placebo do something. Not the way of like, you know, mind over matter, magic belief, but like kind of a little bit almost that thing might a little but be there. So what is that? Yeah, yeah, yeah. It's a really widespread belief, and it's just completely not supported.

0:02:46 Thomas: Anything like this before. I think virtually the entire normie population just thinks there's this thing called the placebo effect that makes you have the same results as medicine if you think you're getting medicine like I think that's like 90 of all people think that or more so interesting so I'm always curious to learn more about that and I wonder how this took off so much you know like do you think it's a failure of science you think its people is the charlatans trying to sell stuff? Do you think it's just a misconception that is really sticky? Boy, whoever invented

0:03:24 Jenessa: You know, well, I'm always going to blame science for not communicating well enough, for not getting out there in the public, for having everything behind a paywall. And even if you can get behind that paywall, it's not written in language that people can understand. I think that's a huge part of the problem. I also think, yeah, it's super tempting for charlatans to take advantage of intentionally. And in a more unconscious way, if it seems like if you read these papers and come away with the impression of like, oh, wow, so like taking a sugar pill did something. Somebody who's not acting nefariously could easily build on that and think they're doing good by trying to take advantage of that and use it to do something and like it's such an attractive belief that like oh I just mind over matter I just gotta believe in it yeah yeah yeah there

0:04:22 Thomas: That mind over matter phrase. I think so much of our reasoning is just based on what sounds a little better word wise. All right. Well, we'll take our first break and then we're going to dig

0:04:31 Jenessa: There's really something to learning to be a good communicator to come up with those phrases that stick in people's heads? We love it. We love a good pithy phrase. And instead of fighting it, why don't we just embrace it and try to use that for good, you know? Yeah. So let's, you

0:04:53 Thomas: Into it. I'm excited. Okay. Where do you want to start? Wow. Really? And I can make it work for me.

0:05:09 Jenessa: Know, very briefly, what is this placebo effect thing? If you ask your average scientist, you know, especially somebody who's designing, usually it'll be medical drug studies. It's really just any effect in the control group. Most people doing these studies are not really interested in why is something happening in that group. Like it just is and I need to control for the fact that stuff will happen in this other group. I pulled some headlines on the other hand putting the placebo effect to work. That's from Harvard Medical School, Harvard Health Publications. Unlocking the healing power of you why placebos work and how you can use the placebo effect to accomplish your goals this is so much more than effect in the control group right like this is implying a direct effect that they work that they do things that you can choose to yeah

0:06:12 Thomas: Yeah can I like get it to edit my podcast what can I do with this placebo effects I need help

0:06:13 Jenessa: Yeah. Yeah. It's all implying that the placebo itself, you know, There's a picture of a pill on this one. Use this inert pill to do something. And like, that is just not what is meant in your average study, right? So what's really going on? There's a couple of things. I would divide this into two categories, really. There's some basic statistical explanations for what's happening. And then what I would call like human effects. So for statistics, some examples: spontaneous improvement sometimes people just get better on their own tumors shrink and we just don't know why sometimes people are misdiagnosed and miraculously on their second appointment they are healed what are you gonna do we expect that even if I did nothing in this study at all if I just tracked a group of people some of them would get better so I gotta make sure in my study I have a group of people I track to account for that. We've also got the issue, why did I give a placebo pill? Why did I give a placebo injection? Whatever. We're assuming it's a placebo. We're assuming it's inert, where there's the active substance, the drug I'm going to give you, and then there's this other thing that does nothing. I want to double check that it does nothing? The drug, the pill I give you has an active ingredient. It's also got binders, so it's pill instead of a powder. Maybe it's got things in it to make it less bitter or something that makes it digest easier and you just never know if that thing is not inactive yeah who knows right so certainly to justify giving you this new drug that may be expensive it may have side effects it should certainly be better than just the binders in this pill that doesn't mean the bindings in that pill magic. It could mean they're actually drug candidates, like you said. Highly unlikely that that is happening, but if we're going to be extremely diligent, control for everything, I'm going to give you everything everyone else got except the active drug. Another thing that can happen is a regression of the mean. Sometimes this is portrayed as like just any time you see a high score or a low score in an individual, you should expect it to get closer to the general average, that it should get less extreme. It's not exactly right. Regression of the mean is a very specific thing that will happen when you make a group based on scoring high or low on one metric and then you measure them again on that same metric at a second time point. That's what causes is this problem. So think about like batting averages. Take all the rookies in Major League Baseball and here's the top 25%. Here's the bottom 25%. Who's in there? The bottom 25% is going to contain some people who are just not good batters, you know, for Major League Baseball, right? They'd

0:06:17 Thomas: Yeah right yeah oh yeah hmm you accidentally came up with a new penicillin just in your sure yeah they just happen to be in a bad you know in any probabilistic thing if it's 50 50

0:09:21 Jenessa: Kick my ass. Some of them, they're not that great and they're going to stay not that great. Some of them are not that good. They'll get better over time, but they're never going to be the best. Some of them had a bad year. Their head wasn't in the game. Something was going on in their personal life. They had a bad coach. They had an injury, whatever. Essentially, they shouldn't have been in this group. It was an accident. And the next time I measure them, they're going to perform more representative of their actual ability? Yeah. Yes.

0:09:56 Thomas: It doesn't mean it happens like heads, tails, heads, tails, heads, tails, heads, tails. And so like you could, you could switch that. I mean, it's almost maybe more illustrative to take something that's not skill based at all. Right. Where it's just like, take a group of a hundred coin flippers and be like, all right, we've got the people got a nice distribution. They all flipped it a hundred times. We're going to take the most tailsy people and we're going to put them in this group because they're super tailsy.

0:10:22 Thomas: And then you measure again, let everybody do it again. And then you're like holy shit they've lost their abilities they're no longer super tailsy as tailsy yeah yeah yeah there's a famous thing that I don't know if it's exactly this but

0:10:32 Jenessa: Right. And what this would not tell us is that they were tailsy and now they're not or that anything happened. And same here. It would not Tell us like, it's good luck to do bad your rookie year. You'll do better next year. Like, no, no. No. Sure. Yeah. Right. Right? It could be you're

0:10:52 Thomas: It's one of the things people think it is there's something called like the madden curse have you heard of that it's where whoever gets I mean it happens in other games too but back in the Madden days I think they're still probably making those games but I haven't paid attention in a while but But whoever got on the cover of the Madden video game for the previous year would have a terrible year the next year, like happened like all the time. Sometimes it was like, oh, they got hurt. You know, sometimes it's just other bad luck stuff. But it was interesting because it seemed really, really consistent. At least some amount of it is probably that. It's like you got on The Cover because you had an extremely good year. Maybe you're good, but also got a bit lucky in that year and had some statistical fortune, you know. So like returning down to Earth.

0:11:37 Jenessa: The best in the league. And, you know, maybe they just don't want to put you on the cover every single year. But a lot of people, it could be that was not representative of your typical performance. You were the best in the league because everything came together, right? So it's the same kind of concept here. If I created groups that were the high pain and the low pain group, some of them are in the high-pain group because they consistently have high pain. Some of them are in the high pain group because they had high pain today. So when I measure them a week from now, a month from now I should expect some of those people will still have very high pain because that is their consistent I caught them on a representative day. Some of them will have lower pain because I did not catch them on a representative day I shouldn't have put them in the high pain group it was a mistake. They will pull the average down of that high pain group. And it looks like, oh, pain went down on the next month. It's totally a statistical artifact. Now think about what if I didn't make that group on purpose? Who joins a clinical study? Who takes experimental drugs that are not yet proven that, you know, we don't know what the side effects are? It's not the person who's like, well, you're a little kind of painful. Things are not great. It's almost exclusively people who come to the doctor like god whatever you've given me is not working I'm really struggling and if you've ever been in a lot of pain you know that if it's happening this week and it's really really high pain you may have forgotten what it's like to be in lower pain so it's not even necessarily a screw-up or a nefarious thing like somebody's trying to get in there they may have come to their doctor on a bad week and the doctor was was doing signups that week and was like, I have good news for you. This is expected. This is kind of unavoidable. And especially if we don't pull into this study a low pain group, we can't even really see it happening, this convergence towards the mean of both groups. We just have the high pain group. Those are some examples of things where this is just statistics. This would happen if you were testing, like, cockroaches. They don't have to have a belief that medicine works. Like, some of them would just get better.

0:14:04 Thomas: There's not a Deepak Chopra, I don't know, I almost had it, saying like, hey, cockroaches, if you take this, the quantum effects are going to, etc., etc. No, they're just, okay.

0:14:17 Jenessa: Right. Regression to the mean will happen. Spontaneous improvement will happen, and so we expect some people to be looking like they got better. Maybe they did get better but like it would have happened if they weren't in the study and has nothing to do with their beliefs human effects it's kind of just a label I've assigned to the things that do require human subjects as your participants and it's got a little bit to do with them having kind of thoughts about studies but it's still not about believing in meds So, for example, the Hawthorne effect, this is just when people know they're being observed, they act different. Let's say you're coming in for a study about a new heart medication. I am not going to tell you to stop taking your blood pressure meds because that would be highly unethical. So I'm going to say, okay, everybody, I need you to write down every day when you took your meds and what your dose was. You're probably going to increase compliance with the med they already take. People forget sometimes. They don't always take it at the same time of day. Now they know they're supposed to, but they almost don't do it. And now they're, consciously or unconsciously, they're going to comply better with the current medication and get more effective treatment. This can happen with like, oh, diet and exercise. I need you to, you're all going to try this drug and I need you write down how much you ate and and how much you exercise every day. I don't want to recommend this. This can lead to severely

0:15:23 Thomas: Interesting. Yeah. Right yeah yeah good point yeah you're like ah it snowed today god damn it

0:15:54 Jenessa: Disordered behaviors, but it does tend to work. People who, without any goal to change anything, just tend to write down their behaviors tend to change their behaviors. Oh my god, I haven't had a vegetable all week. Like, you might do that, you know? So this will not happen to our cockroaches, but it will happen to our humans. And we might see the placebo group and the experimental group show these changes to better their behavior, to comply better with medications, that kind of thing. You can also get demand characteristics. This could be really big with things like pain. So already we've got like, oh, rate your pain on a scale of one to 10. And like, what is this? You know oh 10 is the worst pain you've ever experienced like some people have had a baby and some people Have Never Had Stitches 10 is The Worst Pain You Can Imagine Who Knows What I'm Imagining Like That's Helping No One Sometimes They Do The Like The Little Smiley Face And It's Like Oh The 10 Is Like Crying And Like Some People In A Lot Of Pain Laugh Or Go Stony Faced like none of this works so it's already very fuzzy but at least within a person you could imagine like I could tell you over time it got better or worse hopefully even so let's imagine I have like a nice I have a good relationship with this doctor I'm coming in you know I'm doing all the things and they come in and hey how you feeling well not better but maybe not that much much better but I like you so 6.5 like consciously or unconsciously people may be attempting to help by reporting less severe symptoms like yeah I think it's working totally yeah obviously if the relationship is poor we might not expect this but if you are consistently coming in this can start start to happen for people. Put this in with selection issues. So this is kind of like a combination of the stats and the human subjects problems. Okay. You come into this study. It's already hard for you to get out of bed, get out OF the house. You're having severe symptoms. This really sucks. And you got to come in here once a week and you get blood draws. You are giving stool samples and just, ugh, this is terrible, but it's working. You feel better. I'm going to tough it out, right? Like, this is great. If it's not working, if I am doing all of that, and I just, I don't, I still don't feel good, I might drop out of this study. And I can't force you to stay in. If it was an animal study, like I just you know, we keep doing what we're doing. But humans, they leave if they want to leave. Obviously, if you're being responsible about this,

0:18:44 Thomas: Yeah cockroaches are like, I think I should go home. I mean, I'm gonna get out of here. Yeah,

0:18:51 Jenessa: You're going to make sure that in your initial data you don't include people who dropped out right you're gonna say of the people who stuck all the way through here's what their symptom ratings looked like at the first and the second and the third time point and yada yada even so of people in the placebo group who happened to start feeling better because of spontaneous improvement or because you recruited them on a bad day and then regression to the mean they started to just have a better day over time or demand characteristics. They're starting to tell you they feel better and maybe they're almost convincing themselves that they feel better. They might be way more likely to stick around. The people in the placebo group who are not feeling better because they're not getting an experimental drug and they're are not spontaneously improving, they're way more likely to drop out. If your experimental drug works, the people who are feeling better are more likely to be in the experimental group. They're just less likely to drop out, so people with severe symptoms but seeing improvements stick around. So essentially, it's going to magnify that change you might see in a placebo group. It's already selecting for a group of people who are more likely to be spontaneously improving or recruited on a bad day. Put all of that together and you can essentially say, yes, I saw a change in my placebo control group. Some of them got better for God knows what reason, but it had nothing to do with me. Some of them say they feel better, and that could be for a lot of reasons that have nothing to do with actually feeling better. Some of it is a statistical artifact. It has really nothing to do with anything. It was expected. It's kind of inevitable. For the objective symptoms, objectively measurable, size of a tumor, survival rate, healing time of wounds. Yeah, we're not going to get like demand characteristics there. Like I just believe so hard in my doctor or that I just like toughed it out. But things like pain or ratings of like how well you're doing with whatever, it's gonna be a lot more likely to be influenced by this. Okay, if we could somehow measure,

0:21:33 Jenessa: And people have done this, they've done a lot of studies to try to measure the impacts of these. If we could say, here's the difference between the placebo and the experimental group. So that's our basic placebo effect. In a more complex way, let's break out how much of this was due to spontaneous improvement how much this was due to regression of the mean it's really hard to actually figure that out but let's say we could subtract, subtract, subtract, what is left could you call that the power of positive thinking probably not but like what is it is it maybe a real placebo not really maybe I have a really really cool study for you about this. Not necessarily. So in some studies, it's entirely

0:21:54 Thomas: Hmm oh wait do we know that there would be something left mm-hmm wait are there studies about rats liking food I figure they just like food

0:22:20 Jenessa: It's possible there would be nothing left. Maybe after we get rid of spontaneous improvement, regression of the mean, selection bias, the fact that people drop out weirdly, get rid of demand characteristics, maybe there would be nothing left. There would be no unexplained variability still hanging out in my data. I think that's much more likely when we're talking about the objective symptoms. It's almost certain, I would say there is just no mechanism out there to imagine how you could shrink a tumor by just believing something. I can't even begin to think what that would be. We're also often going to have unexplained variability left just because practicality? What are the odds? I can explain all of the stuff, you know? There's probably always something I'm not thinking about. Humans are so weird. We just do weird stuff. Off and I ask people to rate something on that 10 point scale and like God knows what they're telling me so like I expect some unexplained variability at the end of this and then we can speculate about what it could be could it ever be because very specifically because I took this inert substance and we'll assume it is truly inert it's not an active ingredient because I I took it I felt better maybe maybe a teeny tiny bit the majority is going to be explained by that other stuff this study what they did they pulled in a bunch of rats and I love this study because they have so many controls built in they're going to bring the rats in and first uh well first First, they're going to have them fasting, so they're very hungry, and they're going to give them sweetened condensed milk in like one of those little dropper bottles. You come over and you like lick the thing. They love it. They're hungry. It's very sweet and yummy. They love sugar. As a side note, that is something anytime you read a rat study about like how much they loved whatever food, check the methods. They have very often fasted the animal or even been like, oh yeah, we starved them down to like 80% of their body weight. Yeah, they were hungry, man. Of course they did whatever they had to do to get the food.

0:24:45 Thomas: Right yeah or it's just my kids and they're like yeah they'll do anything they'll they'll kill a

0:24:45 Jenessa: Well, sometimes like a lot of those studies that are like, oh, addiction to sugar, it activated, blah, blah. They did this, this, and this to get a hold of sugar. And it was rats that were at like 80%, of their baseline body weight, of course.

0:25:03 Thomas: Guy. That's Remy. That was Remy Rat. Wow. Wow, Rat, what else am I doing? I miss Gay Rat. Gay

0:25:03 Jenessa: Children are a whole other beast with their dietary desires they are strange but yeah the vast majority of those I think all of those studies used rats that were like starving so of course they were like oh sugar so always check that but yeah these rats were hungry so that it would you know incentivize them to they're very curious little creatures they're gonna go explore explore the cage. They're going to do this. They'RE going to do that. And they're like, no, I want them to go over there and eat the food. So they got 20 minutes in this cage. It's not their normal cage. They go in, they're going to explore, they're going to, you know, toss the bedding around, whatever. And then they'RE going to go over and they're going to find this little food source and they're going to count how many times they lick it as like, okay, they liked it and we know that they are pursuing it. They kicked one rat out for licking it 649% more than the other rats. And they were like, this is going to be a problem to try to assess your data. So I don't... Man, I did some rat research in grad school very briefly, and they really have personalities. They've got some unique ways of behaving. I definitely had a gay rat, for sure. He would like never mount the females and occasionally tried to mount the male rat that he was housed with. Like, I don't have an explanation other than gay rat for sure. There was one female rat who was like the horniest rat on earth and she would mount him sometimes to be like, this is what you do. Try it. And he just couldn't get into it. Anyway, they pull the rats in. They go over. They go to lick the thing. And when they put their nose in this area to get to the little bottle with the food, their nose is going to touch a little thermal thing. And at baseline, it's 37 degrees Celsius, which I am told by this paper is a comfortable temperature. It's just warm and cozy. I don't know. So they figure out a baseline for each rat. How much do they go over and lick this thing and enjoy it? Take them out. Come back later. And now they're going to go in and it's 48 degrees Celsius, which is noxious. It's painful. So don't think like burning, but like, ouch, like not comfortable. They made sure it was below what would be like, you know, obviously if you damage the skin, the next time you touch it would be even worse. But also like you know trying to be reasonable about this and like the thing they might you might put up with it if you were like oh but I'm so hungry so they get a baseline for how much will this rat go over there and lick and tolerate the discomfort the the ouch of this then really great control. Bring them back in and bring it back to 37 degrees Celsius. Retraining. We're going to double check that we have not created a conditioned fear response. They're not going to go over there and go, I don't want to touch that. It's too hot. I don' like it. No, no, no. They go right back to their baseline. So again, it's not so painful that we've created some problem where in later trials they're not going to go to it. It's just enough that during the trial, they're going to react as if it's painful, but later they'll be like, I'll try again. It might be nice again. Who knows? I'm hungry. Yeah. I'm in a relatively small cage. I don't have a fellow rat with me to hang out with. You know, they tend to be pretty, yeah, they're very friendly. They like, there was a ton of

0:28:56 Thomas: Rat was so entertaining. I'm still wondering what's going to happen. Okay, an injection?

0:29:02 Jenessa: Of studies that showed basically a lot of early rat research had to get tossed because they are very social they really need stimulation and so when they were housed alone without stuff to do they're just kind of depressed and shitty and their behavior is not normal so the recommendation is you always house them two to a cage or with stimulation like stuff to play with or both So that they're more behaving the way they would. I mean, they're still, you know, basically domesticated at this point. So it's not like what you'd expect a New York City rat to be up to. But it's really important that they have like stimulation and socialization. They're very social critters. Okay so now we're going to get into we've figured out a baseline for this animal we've figured out that it's not creating a conditioned fear response we've taught them that sometimes it's painful and sometimes it isn't and you don't have to be scared and you can go back and forth and now we are going to have this branching tree of like seven control groups and one experimental group. We're going to try to control for every possible alternative explanation for what's going to happen here. Yeah. Okay. A bunch of them will just repeatedly get PBS phosphate buffered saline. It's a placebo. It is a saline injection and it doesn't really do anything. It just, you know. Yeah, yeah. I believe it's peritoneal. It's just a little,

0:30:35 Thomas: Injection huh yeah nah it's a myth yeah yeah yeah it's like adrenaline and stuff

0:30:39 Jenessa: Very small volume of a little bit of liquid that's saline. It's just basically matched to the fluid that's already in the body. So it's not going to affect anything or do anything. But we're making sure we're still giving something so that they still get a needle, they still get a little extra fluid in the body. Just not a drug. And when that happens, if you just give that and then they go in there and it's the 48 degrees Celsius, it's painful. Yeah, they react as if it's painful. They go over there less often. And that group, if we just keep doing that, nothing changes. They're rats. They don't have beliefs about injections. They don't change their behavior because of it. Nothing happens. Then we have a group that's going to get morphine in that 48 degrees, that painful condition. They are going to go lick it more often. So it's a pain suppressant, right? The way morphine works. It's a central nervous system function. It gets into, so central being the brain and the spine. It basically suppresses the pain response. So like the receptors in the skin are still sending the like, uh-oh, uh‑oh signal. Compared to like if we did lidocaine, you may have had that like in a spray or a patch on your skin before. That would actually numb the skin. The pain signal would never go up. With morphine, the signal goes up and then your brain's like, I don't know. Is it really that bad? Maybe we don't need this. Essentially, you have a endogenous opioid system. You have receptors for opioids you make, and they are your natural pain suppressant. If you've ever gotten runners high, I have run my ass off and never gotten Not for me. We've certainly, you know, I'm sure you've seen examples of like somebody is injured and at the time during the emergency, they kind of push through. After the emergency is over, the pain is perceived. Exactly. It's that system. So technically, adrenaline is a different system in the sense of like it binds different receptors, but they're interacting. So that adrenaline system kicks in in the sympathetic nervous system. I always remember this one is like it is sympathetic to the fact that you are going through something really bad. You have the parasympathetic is what you're in most of the time you're chilling out. The sympathetic is going to say, eyes dilate, blood vessels dilate. Your lungs open up so you can breathe better. Run, run, run. Shut off that digestive system. This is why runners are pooping themselves during a marathon. Marathon it's just go go go get you out of the bad situation that can also interact to trigger the opioid system the endogenous so endo inside genus generated you make your own opioids you got this already morphine the exogenous exo outside chemical it binds the same receptors Comes in and says, hey, pain relief, extra super duper pain relief way more than your internal system could ever do for you. No amount of sitting in the hospital and convincing yourself it's an emergency. Please give me pain relief. My own body. It'll never come close to what morphine can do for him. So we give the rat the morphine. It goes into the central nervous system. It binds those endogenous opioid receptors, and it says, pain is maybe not the worst thing in the world. I am hungry. I'm going to lick, lick, flick this thing. Great. Right. Then we can give naloxone. I talked about this a while ago. Narcan is the drug name. So this is our miracle drug to rescue you from an overdose. And essentially what it'll do, it will block opioid receptors. And it'll block them so much so that your endogenous opioids cannot bind,

0:34:52 Jenessa: Your morphine cannot bind. And, essentially, if we gave morphine and nalaxone at the same time, or just naloxone, we are going to knock out this response. They're going to behave as if they're right back at baseline or maybe even worse. They are in more pain because they can't block it. So this is all expected. Morphine is blocking a pain response so we get more pain tolerance and we're measuring it and instead of asking you to rate on a 1 to 10 or whatever, we're just just measuring it in a very objective way. They are clearly tolerating what was previously a noxious, painful experience. They're tolerating it more. I give naloxone, which would block the response, and they are not tolerating It. They Are back to, oh, this is terrible. I don't want to do this. If I give morphine and then saline, this Is when it finally happens. So the morphine condition. They got pain relief. They went over. They licked the thing more than baseline. They clearly are experiencing less pain. Come out of the cage, do a washout period, so let the drug get out of their system, bring them back in, give them an injection again, but it's just saline. Those rats behave like they got morphine. It's basically a placebo effect. They act like they They have some pain suppressant in them and they go over there and they lick more. The critical trial is take those rats. They got morphine. Then they got saline. They look like they have a placebo. Then give them naloxone and the placebo effect is gone. They go right back to acting like, I've never seen this in my life. I am in pain. This is terrible. We have a mechanism. We can actually identify what could be happening here. If they get the placebo, right, they got the morphine before they experienced the pain relief. How could it be that saline came in and somehow they still acted like they got morphine? They're rats. They don't believe in this. That's not what's happening. I cannot emphasize enough. This is not about just believing hard enough that

0:34:47 Thomas: Wow interesting yeah it's not Deepak Chopra or Oprah talking about the secret

0:37:14 Jenessa: But you're, it's not. It's definitely not. So what happened? Same problem. Okay. They have in a, not in a like higher order learning sort of way, but in a classical conditioning learning. When the morphine came in, the injection happened. Okay. I got picked up and and then somebody had a needle, and then a little ouch, and then, oh, and then I felt good, and then that thing, it was quite warm, but it didn't hurt. Cuz critically, this is gonna suppress pain, but not temperature perceptions. The level of heat is still perceived the same, but not as painful. The next time all the same rituals happen, essentially what we think is going on is that endogenous opioid system is gonna kick in with it. It's going to go, aha, I've seen this before. Now, not on like a cognitive level, but on a really base, deep in the brain kind of level, which is also why we'd expect this should translate from rats to humans. This is like a brainstem kind of thing that essentially it's saying, oh, this is the good thing. This isthe nice thing, the thing that makes me feel good. And the reward system can be part of this. The obvious classical conditioning is like, You're walking around town, you smell something yummy, and you go into that restaurant. And then the next thing you know, every time you walk past that corner, even if you don't smell the food, you salivate, you get hungry. It's your body saying, hey, even if you weren't hungry before, let's not miss an opportunity. I'm going to incentivize you to go eat. I'm gonna prep you to eat. I'm going to first do the salivating and dumping the stomach acid, but I'm also going to make you hungry. So you'll go eat it. And then when you engage in the behaviors to seek the food, I'm gonna reward you. Sometimes it's not even the first bite of the cake that's rewarding. It's the making the cake and the seeking, the going in. You walk, open the door, you're already getting the dopamine, the reward. Essentially, it might be kind of the same thing. It's reaching for the pill bottle could already start the, oh, thank you, God, for doing this

0:39:43 Thomas: Hmm. Wow. Hmm. That's interesting. Hmm, hmm. Wow, wow. Yeah. So the thinking is if that happens in rats, that could be some part of what's happening in humans. Hmm hmm. Yeah, yeah. It also doesn't seem like it would be that sustainable, right?

0:39:43 Jenessa: Nice thing for your body. What is weird about this? It might almost be the opposite of what we'd expect because sometimes you get a homeostatic mechanisms, homeostasis, this balance idea. So sometimes when your body knows like a stimulant is coming in, it can depress your nervous system to go, OK, my heart rate's about to go up. That can be bad. I should push it down. So you'll dump like the opposite chemical to try to save your life. This has been proposed actually as a mechanism for how an overdose can happen sometimes, that somebody who is a regular user of a central nervous system depressant, they know their dose, they've got some tolerance built up, and they also have a ritual. They go to the same place to the same dealer they get the drug they come home or maybe they use in that location they have all this set up that they do and their body has like learned the thing is coming dump the stimulant so that when the depressant comes in I counteract I have a dose that depends on that happening and if I go use in an unfamiliar condition and I don't dump my stimulant to counteract, it seems like people are more likely to overdose in unfamiliar situations. What's weird here, we might have expected the opposite. You knew morphine was coming and you might have dropped your endogenous opioids like, oh, I don't need it. But it kind of seems like the opposite, like your body is going, ooh, hooray, pain relief. I'll give myself even more. I'm going to reward myself for seeking it. This system seems to behave differently. You have homeostatic mechanisms to try to save your life if like stimulants or depressants are coming in. But when it's the reward happy stuff coming in, your body feeds it and goes, yes, even more. I will participate in this process. Where does the naloxone come in? The fact that we can block the placebo is how we are getting to that is the mechanism if your endogenous opioid dump I am going okay yes hooray injection this is I don't know what this drug is but this is going to make things less painful I will dump my endogenous opioids naloxone blocks that and I don'T get a behavior that looks like pain relief anymore so essentially it's not about some higher cognitive I believed in the pain relief and so I felt less pain now the drug stopped this thing your brain was classically conditioned to do and therefore stopped the placebo it stopped the pain relief. Super cool, just all the ways they controlled

0:42:53 Jenessa: Trolled for so many different possible alternative explanations for what happened and they got it in a rat they got a placebo in a rat it's so cool it's way beyond the things we expected in the cockroach this is not regression of the mean this is non-spontaneous improvement this is the thing we were looking for they reacted to a needle because they've learned that needles make them feel good. And then we blocked it with Narcan. Like, it's just so cool. Yeah, exactly. So that type of mechanism, you know, I'm always reluctant to be like, oh, rats, like, let's say this is anytime we've got like a drug treatment or a psychological thing that happened in a rat, I'm Always careful to be Like, is that going to really happen to a human? This mechanism, this exists in pretty sure all mammal brains um might even go past that um you always want to be careful when you get outside of mammals because like other animals will have the same receptors but they might not do the same things so like I don't know if opioid receptors if they exist in a fish do they do pain relief I'm not quite sure but certainly like mammals all have these same analogous structures because it's not up in like you know prefrontal cortex it's like a very like human thing it's where you do decision making and thinking about the future and impulse control like all the stuff that like teenagers and rats don't do the lower brain structures like this and these really simple circuits like rats have it and we have it narcan works on us it saves our lives. Probably most of what we see in a placebo control group is spontaneous improvement, regression of the mean, the Hawthorne effect. But if we subtract all of that out, if we just look at like you are just at home and you reach for that Advil bottle and God, you almost feel like sometimes you feel better faster than the mechanism of action. Like that should take, I have no idea how long Advil should take to work, but like you feel better in like 15 minutes and like that's too fast. That could be this. It could be your body rewarding you for like, thank you, Jesus. We have been in pain for so long. I've been trying to tell you to go take the thing that makes the pain go away. And you just kept sitting on the, I don't, this might not be everybody's experience. I do this where I sit there in pain and don't do the thing and walk 15 feet to my bathroom and take the medication that I know will make me feel better. Why would it reward you for that? Really, your brain shouldn't evolutionarily like care too much about that. But usually pain means injury and removing pain means healing. You can hijack that with like you know this pain it's healing on its own reducing the pain isn't necessarily helping the healing but like the mechanisms we've evolved for this do not understand that so it's still just going to be do stuff to make pain go away thank you keep doing that stop running on your broken leg you idiot unless there's an emergency and then we'll suppress it It's probably a small percent of what's happening. It's still not will yourself out of illness. It'S still not your fault if you're in pain.

0:46:46 Thomas: Right. I mean, how how long could this last? Like, how long does it take before your body's like, oh, wait, we're not actually getting anything from that, you know, sugar pill or whatever, that empty injection.

0:46:55 Jenessa: That's a good question like the general concept with classical conditioning there would be

0:46:57 Thomas: Yeah. Just you could just keep doing this rat design.

0:47:01 Jenessa: Extinction every time you raise your hand you get a cookie so you start raising your hand to get a cookie how many times you have to raise your hands to get the cookie how often do you have and not get a cookie before you stop the behavior sometimes it really depends on the scheduling so for example if it's like every third time you get the reward or a variable schedule where there's like a 60% chance as opposed to like instead of every third time it's it could happen twice in a row who knows those schedules are really hard to get extinction this is gambling sometimes you win a couple of times in a row oh my god breaking that habit is like crazy your brain will just keep latching on to repeat the behavior repeat the behaviour repeat the behavior so the dopamine will that persist depending on like how often you got a sugar pill versus an actual drug to still get the pain relief though like how separate is that dopamine reward that's making you repeat the behavior versus to get the placebo pain relief I don't know like when would that extinguish well you could test it with this rat design yeah I wonder this study I'd have to look at like later

0:48:21 Thomas: I'm surprised they didn't do that. Like, see how long it'll go. And this really only deals with like a slight pain improvement or is there other responses that this gives?

0:48:29 Jenessa: If I go look at these authors, I'm sure if they did that, if they did that iteration, they will have cited this paper. So I could check that out. That would be interesting. Certainly, like I would expect if your doctor kept prescribing placebos to whatever, first of all ethically, do not do this, obviously. But let's say like a misfill, an accidental giving you the wrong thing, like that happens sometimes. How long could that go on? And to whatever tiny extent this was working, when would it stop? That is a really interesting question. Probably wouldn't last that long, you would think. Exactly. Yeah, this should be, this is very

0:49:18 Thomas: You know, it's not like we can harness this for anything. Thing, right? Like, it's constantly — yeah, that's right. It is the thing — you're all, you're

0:49:25 Jenessa: Specific to pain. When we are talking about using morphine and then naloxone to test this mechanism, that relies on the endogenous opioid system, which is a pain perception system that would would never address healing a tumor. It couldn't, possibly. I can't even think of a mechanism for that. So definitely not for any actual healing. Obviously, pain relief is extremely important, but we have things that actually do pain relief. So, like, I don't know why we would try to harness this little extra piece. Like, when I can give you morphine. Yeah, exactly. So

0:50:03 Thomas: Doing is kind of doing the thing that the medication is supposed to do anyway, right? It's like sort of doing it right. Hmm, it is interesting because I think it's

0:50:09 Jenessa: So like this is a scientific curiosity to start to understand how that little thing might, how could it, and why would it happen, without it being magic? It is absolutely not a route to a viable medical treatment of any kind. Just give the Advil. There's no reason not to. I can't think of a person who can't get any medicine. If they've never gotten medicine that is pain relief before, then this wouldn't work. This is classical conditioning. You have to start with something that works first. Maybe weaning, but usually like weaning off of drugs, one of the things that will work, for example, Suboxone. It's a mild agonist for the opioid system. So what it does is it binds and it slightly activates. It doesn't morphine or heroin-activate the system, but it a little activates. It gives you the, okay, I was craving, I feel a little better. It's got really high binding affinity, like kind of naloxone levels of binding. So some people have said when they're on suboxone, it's prescribed for addiction. They could go have a rough moment and shoot up and it doesn't work because it's blocking it. This is not medical advice. Don't do this. Some people have said this has broken habits for them, that they're like, wow, I like tried my drug and it didn't work and it kind of broke the connection of the craving of that because I did all the behaviors and I didn't get the reward. Obviously, it is not that simple. But generally, like when we think about giving something that blocks it, we're thinking about trying to break habits, not trying to form habits like that's the last thing I want to do is form an opioid habit for anybody so yeah this is meant to try to explain some of how this could happen to a brain it is not useful for any of that and it definitely like we got to get away from this idea of that somehow you're responsible I mean we have data on this like people who have been taught like you're a fighter are you can do this you're gonna fight your way out of this like that leads to shame and guilt for people who are like oh I'm still dying and it must be somehow my fault if I fought harder and it's like no sometimes diseases just kill people and there's nothing you could have you tried everything medically and belief would never have fixed it all that ends up with is somebody on their deathbed who feels bad and like that's not good so no it's still still still not about belief and not a treatment method yeah exactly there's the placebo effect is like 15 different effects

0:52:57 Thomas: Like the first you know the common belief is like there's this amazing placebo effect it can do all kinds of things you know and then there's the debunk which is no there isn't But there's still a little something. And most importantly, though, this is a method like placebo group is a method of science to get better results. Right. We're just trying to eliminate variables and make sure when we're testing something, we are only testing the thing. And this entire placebo effect is a result. Correct me if I'm wrong of like when you split off that group to make sure you're only testing the thing there's some other random stuff that can make it seem like that group got better in different ways yeah yeah right because you assume all those things could be or I guess I think all

0:53:43 Jenessa: And it's very hard to know which it even was and how much each of them contributed and most of the time we don't even care the point is to justify giving you the active substance it better be better than doing nothing. Yeah, exactly. Those things are almost certainly happening in both

0:54:05 Thomas: Those things should be going on in the other group as well it's just you've also got the treatment so it's you're kind of just making sure it's all things being equal positive thinking

0:54:16 Jenessa: Groups, hopefully at equal rates. Maybe, maybe not. But we should subtract out what would have happened if I just observed you over this time point. What would have happened if I brought you into the lab and made you write stuff down. I'm trying to justify charging you for a drug or giving you a drug with side effects, not trying to figure out the impacts of doing studies. Those are a separate issue. Yeah. Just a little bit. And not for the reasons the newspaper

0:54:48 Thomas: Yeah well that's so fascinating thanks for taking us through that yeah I guess at least if people pretend to listen to the podcast, they can get a tiny, maybe small benefit, but not like a ton. Yeah. Yeah. I don't know. I believe in Deepak Chopra. I think that he is the truth.

0:55:02 Jenessa: Told you. Oh, God. Oh no. Deepak Chopra is in my head for the rest of my life. Yeah, me either.

0:55:17 Thomas: Quantum. Where's that guy? I haven't seen Deepak chopra in a long time. Did people finally figure out the quantum thing? I think enough people were like, oh,

0:55:22 Jenessa: Did he just drop off oh no oh my god you're so right oh sugar I am in a— lick

0:55:25 Thomas: Oh, quantum. Now that's AI in a way. It's not really exactly the same thing, but now it's just AI is every single thing. Yeah. Sam Altman. No, that doesn't work. There's not a Deepak Chopra. Deepak Roach. Deepak Choproach. I don't know. I almost had it.

0:56:07 Jenessa: Lick— lick— this thing, I don't have an explanation other than— gay rat