Today you’ll learn how AI might be able to read your mind, what we can learn from a woman who doesn’t feel any physical pain, and a new app that aims to change our approach to mental health treatment.
Today you’ll learn how AI might be able to read your mind, what we can learn from a woman who doesn’t feel any physical pain, and a new app that aims to change our approach to mental health treatment.
Find episode transcripts here: https://curiosity-daily-4e53644e.simplecast.com/episodes/brain-activity-decoder-no-pain-molecules-and-psych-tech
Brain Activity Decoder
No Pain Molecules
Psychiatry Tech
Follow Curiosity Daily on your favorite podcast app to get smarter with Calli and Nate— for free! Still curious? Get exclusive science shows, nature documentaries, and more real-life entertainment on discovery+! Go to https://discoveryplus.com/curiosity to start your 7-day free trial. discovery+ is currently only available for US subscribers.
[SFX: INTRO MUSIC/WHOOSH]
NATE: Hi! You’re about to get smarter in just a few minutes with Curiosity Daily from Discovery. Time flies when you’re learnin’ super cool stuff. I’m Nate.
CALLI: And I’m Calli. If you’re dropping in for the first time, welcome to Curiosity, where we aim to blow your mind by helping you to grow your mind. If you’re a loyal listener, welcome back!
NATE: Today, you’ll learn how AI might be able to read your mind, what we can learn from a woman who doesn’t feel any physical pain, and a new app that aims to change our approach to mental health treatment.
CALLI: Without further ado, let’s satisfy some curiosity!
[SFX: WHOOSH]
NATE: Think of a number between 1 and 10.
CALLI: Okay… Got it.
NATE: Is it…seven?
CALLI: Whaaaa? How did you guess that?!
NATE: I did?! Yes! No way!
CALLI: Wait, are you a wizard?
NATE: I guess I am? Either I have ESP, or I’m an A.I. Because check this out: researchers at the University of Texas in Austin recently published a study in the journal Nature Neuroscience that explains how they used MRI scans and a large language model A.I. to read people’s minds.
CALLI: No.
NATE: Yes.
CALLI: Okay, so now robots know what we’re thinking?
NATE: Well…we’re not quite there yet, but this is actually super cool. Computer science doctoral student Jerry Tang and Alex Huth, an assistant professor of neuroscience and computer science designed the study using three willing participants.
CALLI: I mean, I sure hope they were willing.
NATE: They definitely had to be, because the study had them each listen to about 16 hours of podcasts while being scanned by an fMRI scanner.
CALLI: Think they listened to Curiosity Daily?
NATE: I mean, if they were cool, they would have. But as I understand it, they listened to narrative podcasts that focused mostly on storytelling.
CALLI: Maybe we’re just too smart for A.I.
NATE: Yeah. Let’s go with that. Anyway, as they listened, the fMRI measured blood flow in the brain. The words they listened to and the blood flow in their brains created a sort of puzzle that A.I. was able to piece together. So, when the participants were asked to imagine a story, the A.I. was able to mostly decode it.
CALLI: And, haven’t we done this before? Aren’t there programs that have been able to display a word that someone is thinking?
NATE: Yeah - but there’s a big difference here. The first is that this is totally non-intrusive. You don’t have to have a sensor implanted in your brain, in other words.
CALLI: I mean, I guess it’s the little things, isn’t it?
NATE: Right. The other thing that makes this different is that the A.I. isn’t exactly reading specific words - it’s decoding meaning. And if you think about it…that’s huge.
CALLI: Right. Because it’s one thing for a machine to simply copy a word from our mind, but it’s another thing for the machine to actually understand what we’re trying to say.
NATE: Exactly. In fact, the A.I. didn’t really reproduce sentences exactly the way the participants thought them. But it was able to reproduce the meaning. In other words, just by looking at the blood flow of their brains, it got the gist of what they were thinking, and was able to translate it. One example they gave is when a participant thought the sentence “I don’t have my driver’s license yet,” the AI wrote “She has not even started to learn to drive yet.” So it’s not quite the same, but it’s close enough.
CALLI: That’s both cool and kinda terrifying. So now can it just read anyone’s mind?
NATE: No, and this is why it was so important to have a willing participant. They trained the AI on one of their brain patterns, and then tried to use that AI to read the thoughts of someone else and it was just gibberish.
CALLI: Ah. Okay. So we all have our own thought patterns?
NATE: Yeah. And the participants were also able to throw off the scent of their thoughts by thinking about other things than the things they were supposed to be thinking about.
CALLI: Ha! Okay! So it’s like, okay Robot, you don’t know me!
NATE: Yeah. Totally. And while this might seem a little scary - robots reading our minds - this could actually have really exciting implications for people with disabilities, or who have trouble speaking. But only with their permission…
CALLI: So who gave you permission to know I was thinking about the number 7?
NATE: hmm… it’s one of life’s little mysteries
CALLI: "You are a wizard, Harry.”
[SFX: WHOOSH]
CALLI: Chronic pain is a total…chronic pain.
NATE: Yeah. Good one. That’s true.
CALLI: I mean… Thanks. But, no, seriously. It’s… it’s awful. According to the Centers for Disease Control and Prevention, over 20% of Americans suffer from chronic pain - that’s over 50 million of us. It’s defined as pain that lasts for at least 3 months, but if you know someone who suffers from chronic pain, you know it often lasts a lot longer than that. It can be totally debilitating. Feeling constant pain leads to depression, loss of work, and even addiction to pain medication.
NATE: So if my Curiosity Daily radar is working properly, this is where you say there’s been a really cool breakthrough, right?
CALLI: It’s not just cool…it’s FAAH-OUT.
NATE: Are you a…What are you, a surfer all the sudden? Far out?
CALLI: No, no, no! I said FAAH-OUT, it’s spelled F-A-A-H O-U-T. It’s a gene that just might hold the key to treating chronic pain, depression, anxiety, and more.
NATE: Ohh! Okay, that is FAAH OUT!
CALLI: See! Anyways, in a study published in the journal Brain, researchers tell a wild story about a Scottish woman named Jo Cameron, who, it turns out, feels virtually no pain, has never felt anxious or afraid, and who heals from injuries really quickly.
NATE: You’re sure her name’s not Wolverine?!
CALLI: I know, right? Doctors only realized all this after she had surgery on her hand and her hip and felt no pain after surgery and healed more quickly than normal people.
NATE: Ooh! Okay, yeah, gimme some of that!
CALLI: They sent her to pain geneticists who look at the underlying genetic mechanisms behind pain, and their findings were published in this study.
NATE: And what did they find out?
CALLI: That’s where FAAH-OUT comes in. FAAH-OUT is a gene that scientists thought was a junk gene - that just means a gene that doesn’t create any proteins.
NATE: Okay so…a gene that just sits around looking pretty.
CALLI: Kinda. But it turns out that junk genes might not be so junky after all. Jo has a rare mutation in FAAH-OUT that causes insensitivity to pain. FAAH-OUT and another gene just called FAAH are part of a system called the endo canna binoid system that has a lot to do with pain, mood, and memory.
NATE: So the mutation in those genes causes her superpowers?
CALLI: Yep. That’s the idea.
NATE: So how do researchers use this information to treat chronic pain?
CALLI: Okay, so they’re still at the very beginning of this journey. One of the senior authors of the study - Professor James Cox - said that this was like a “eureka moment.” Understanding where and how pain starts at a molecular level - and not to mention anxiety and fear, and the mechanisms that allow us to heal more quickly - that’s a huge leap forward that can lead to the development of new drugs that could help millions upon millions of people.
NATE: I’m imagining non-addictive pain medication.
CALLI: And how about new treatments for mood disorders and anxiety? Or medications that speed up recovery and healing time from wounds?
NATE: And all of this from a junk gene?
CALLI: And, I’m glad you said that, because that’s one of the bigger parts of this. If these ‘junk genes’ actually turn out to be so important, what other genes have scientists been overlooking?
NATE: The implications are pretty crazy…
CALLI: It’s pretty impressive!
[SFX: WHOOSH]
NATE: A first year medical student at Carle Illinois College of Medicine has developed something that could revolutionize mental health care around the world.
CALLI: That is a pretty big claim. Um…What’s this development?
NATE: Well… okay, basically…it’s an app.
CALLI: Okay… Is this another one of those NFT, like Crypto-miracle, AI, tech guy stories?
NATE: No, no that’s not it, different story.
CALLI: Ahh. That’s too bad. I like those stories.
NATE: Ha! K, well I don’t think you’re going to be disappointed. What medical student Aditya Vaidyam and a team at Harvard have created is called the LAMP Platform - that stands for Learn, Assess, Manage, Prevent - and it could mean life or death for anyone who might need mental health care. So is it tech? Yes. But this has very real world implications.
CALLI: So what does this tech do?
NATE: Maybe the best way to explain that, is to start with the problem that it addresses.
CALLI: And I know the world’s mental health needs have exploded in the past few years.
NATE: In some sense, that’s absolutely true. But the needs were already there. The pandemic played a huge role in bringing those needs to the surface. In fact, by some estimates 1 in 5 Americans experience some form of mental illness - so that’s over 50 million of us. Over 16% of kids between the ages of 12 and 17 say they’ve suffered a major depressive episode over the past year alone. And almost 60% of kids with what is considered major depression do not receive any mental health treatment at all.
CALLI: So there’s a gap between mental health needs and treatment?
NATE: Right. Nearly a third of adults with mental illness say they don’t receive the treatment they need.
CALLI: Is this about healthcare costs?
NATE: Partly. Around 23% of adults with at least some mental health issues weren’t able to see a doctor because of costs. But there’s an even bigger number that might be more telling: there are around 350 people who need help for every single mental health provider in the nation.
CALLI: Holy cow! Okay…
NATE: Yeah, and that actually could be conservative, because it might also include providers who have become overbooked and have stopped taking on new patients.
CALLI: So there just aren’t enough doctors for everyone who needs help.
NATE: That is definitely part of the problem. Training a million new providers would certainly alleviate some of this pressure. But that’s not going to happen any time soon. So the team that developed the LAMP Platform decided to take another approach. Historically, mental health care has been an in-person experience, right?
CALLI: I mean yeah… The famous psychiatrist’s couch.
NATE: Exactly! That means that your care was dictated by your ability to get to an office at a specific time. But over the pandemic, people couldn’t make it to the office, so that care began to shift to what’s called synchronous care.
CALLI: Like tele-health, right?
NATE: Exactly. And it has been a boon. When you need care, you can just log in and talk to your doctor and get the help you need. But you are still at the mercy of the schedule. What the LAMP Platform does is take it one step further, to asynchronous care.
CALLI: So…the doctor and the patient never even meet?
NATE: They still meet. But LAMP is basically an app that you can put on your phone. A patient will see their provider for an initial diagnosis and to be guided through this app, which features cognitive games and other tools that will help your care providers monitor their mental health between appointments. This might seem like a minor shift, but it could actually be a game changer.
CALLI: I mean, I can imagine this means actually getting more care without having to wait for appointments.
NATE: That’s right. Providers can track changes in their patients’ behavioral health during the week, including biomarkers like heart rate, and sleep patterns, and so on. And patients can report symptoms as they happen, in real time. In the past, because providers only saw their patients during appointments, they relied on the patient’s memory of symptoms, or often imperfect retelling of events that occurred. But with LAMP, providers can see who needs the most immediate help in real time and prioritize their care.
CALLI: That seems like a big shift in the way we understand patient care.
NATE: It is. And with the ability to gather an enormous amount of data that they’ve never had access to, they think that they’ll actually be able to provide more preventative care by predicting when and why mental health episodes happen.
CALLI: Okay… that part feels like it could have some privacy issues.
NATE: Yeah and that is a concern. But they say that the LAMP system is fully compliant with U.S. health information privacy standards, and Vaidyam says that one of the problems with adopting a system like this is the trust gap. There are so many different mental health apps with a million different privacy standards, so patients don’t always know who to trust. But the LAMP team is taking steps to address that and to put patient privacy front and center. They even created a database with a list of apps where they use a system similar to peer reviewing to tell the consumer where apps live on the spectrum of privacy.
CALLI: Is this available now?
NATE: Not only is it available to providers, they made it free to anyone, as well as open source so that users can adapt it to their own needs. If you’re interested, you can download the app and explore for yourself! It’s in the app store and it’s called mindLAMP.
CALLI: Okay! So this is a tech story after all.
NATE: Yeah! Who’d have thought we could do a tech story and not even mention AI these days?
[SFX: WHOOSH]
NATE: Let’s recap what we learned today to wrap up.
NATE: A team of researchers from the University of Texas have used an MRI scanner and an AI large language model to accurately read the minds of study participants.
CALLI: By studying a genetic mutation in a Scottish woman who barely feels pain, who doesn’t feel fear or anxiety, and whose wounds heal very quickly, scientists think they may have stumbled onto a finding that could lead to revolutionary new treatments for pain, anxiety, and all sorts of mood disorders.
NATE: A new app developed by a first year medical student promises to expand access and quality of mental health care by helping users track their behaviors in between doctor’s visits, giving healthcare providers a more accurate and timely picture of their patients’ needs.