Today we discuss how infants beat out AI when it comes to common sense, how fruit flies’ use of magnetoreceptors can teach about humans, and what depression has to do with your posture.
Today we discuss how infants beat out AI when it comes to common sense, how fruit flies’ use of magnetoreceptors can teach about humans, and what depression has to do with your posture.
Infant Common Sense
Magnetoreception
Depression Posture
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.
Find episode transcripts here: https://curiosity-daily-4e53644e.simplecast.com/episodes/infant-common-sense-magnetoreception-depression-posture
[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 about how infants beat out AI when it comes to common sense, how fruit flies' use of magnetoreceptors can teach us about humans, and what depression has to do with your posture.
CALLI: Without further ado, let’s satisfy some curiosity!
[SFX: WHOOSH]
CALLI: Okay. Okay. Okay. Okay. Nate. Nate. Nate. Nate. I have. I have an idea. I have an idea to pitch to our very good friends at Warner Brothers Discovery. It is a pitch for a game show called Is Your Infant Smarter Than Artificial Intelligence According to a new study out of New York University.
NATE: Okay. I asked to iyistaiatansoonyu… even the acronym, that’s a little wordy, a little much. So I don't think you can get too far with your pitch, but the study sounds interesting. I assume the study is real.
CALLI: Oh, yeah, it's a real study. So to get to the actual story, a few researchers at NYU became intrigued recently by the phenomenon of common sense A.I. Artificial intelligence that predicts your actions based on your past actions. So, for example, you're talking to your friend about a new game show for Warner Brothers Discovery. You look down at your phone and suddenly you're being targeted with an ad for HBO Max, which isn't a game show, but it is a Warner Brothers Discovery product just the same. That's common sense AI. And the NYU researchers realized it's actually kind of similar to how babies interact with adults.
NATE: Okay. So you're telling me that babies are going to recommend that I watch Succession on Sunday?
CALLI: Okay. No, it's it's similar to the well-known fact that babies are fascinated by other people. Think about how long like babies will stare at a random person or even the studies on babies’ common sense psychology, which is a similar term for a reason. It's the study of babies understanding of things like intentions, goals or preferences. Common sense psychology indicates that babies are able to set goals for others and then have expectations for those goals. They expect people to pursue them logically and efficiently. Their ability to make these predictions is treated as one of the foundations of human social intelligence.
NATE: Okay, so the study is literally just the game show that you were talking about pitching.
CALLI: Yeah. Yeah, exactly. So the researchers were especially curious over the perception that AI lacks flexibility when it comes to recognizing different contexts or situations that guide human behavior. For example, maybe I want to show you a band called Shakey Graves. I'll take your phone and search for it, and then suddenly your phone starts serving you ads for the shake weight, which is kind of an inappropriate looking workout device. But it's because they said the word shake both times, even though I was talking about a band.
NATE: All right. I'm not sure you actually had to type it in my phone. I think it's probably just listening. In fact, just record podcasts probably is enough because my phone's in the room. But I see you were talking about. So how did they measure the difference in this?
CALLI: Through a series of experiments with 11 month old infants. They gave them something called the Baby Intuitions Benchmark Test, which is the cutest acronym ever. It's the BIB test. This test is meant to probe common sense psychology-related answers, but it's specifically designed to quiz both baby and A.I. alike. So the babies who participated over Zoom watched a series of videos of simple animations of shapes moving across the screen. So think of like an older video game like Legend of Zelda or something, but with shapes instead of people and monsters. The shapes simulated human behavior and decision making by searching for and retrieving objects all over the screen. As for the A.I., the researchers built and trained a set of learning driven neural network nodes, basically just A.I. tools that help computers recognize patterns while simulating human intelligence and had the computers watch the same videos.
NATE: Okay, so we've take we've taken some data. I think we need a result. Time for the drum roll Calli. This is your area of expertise. About a bit about about a bit of a drum roll. Do a drum roll. Is your infant smarter than AI, huh?
CALLI: Okay. Yeah, definitely. So Not to be too anticlimactic here, but infants recognize human-like behavior and motivations much better than AI. Even in the simplified actions of an animated shape. On average, most infants were able to predict that the actions the shapes were doing were being driven by hidden but consistent goals. So, for example, one shape would always try to get the same object no matter what part of the screen it was on. As if the shape was dedicated to having that item, the researchers noticed that when infants became fixated on that shape, their eyes would look across the screen to find the object it had been collecting from before. They called this a surprise paradigm because the AI models showed no evidence whatsoever of understanding the motivation of the shapes.
NATE: Okay, that's pretty interesting because a human baby’s knowledge is limited and honestly, we can't communicate them well that well. So it becomes a bit intangible. And to me that kind of thinking makes sense as to why they would identify the motivations of a shape. And to me, that kind of thinking makes sense as to why they would identify the motivations of a shape. But why can't I pick up on those motivations?
CALLI: Okay, to be fair, it's not necessarily true that all artificial intelligence is incapable. This was just one model the team built for this specific study. But the. Big test is considered a state of the art way to measure human intelligence on not only an intellectual level, but a social one. Think about the recent controversy surrounding Bing's open AI powered chat bot. Within one day of its existence, users were able to manipulate the bot into revealing classified trade secrets from Microsoft, simply by asking, not by hacking, not through any kind of manipulation other than asking the right questions. And it kept doing this until humans had to step in and reprogram the chat bot because it never actually learned how to tell the difference between being manipulated and having a conversation, which most babies do.
NATE: Okay, that's great news for parents everywhere, but I feel like it's bad news for fans of A.I.. What's next after this?
CALLI: Okay. If you're scared of A.I., you can rest easy knowing that there's no possible way you’ll mistake, your interactions with A.I. for humans anytime soon. If you're excited for A.I., it's going to take a bit more time for machines to be able to draw upon the core abilities infants have for detecting goals and preferences. But at the rate A.I. is evolving, who knows? Maybe it will be smarter than an infant one day. But for now, it's not quite human yet.
[SFX: WHOOSH]
NATE: I was thinking a lot the other day about the story we did on magnetic migration, and I wanted to look into how other animals use that ability. And what I found is that there are a lot of animals that display evidence of having a sixth sense.
CALLI: Here to do this to you. But that makes a lot of sense. I'm sorry. Please don't fire me. Please explain more.
NATE: It turns out just a lot of animals can see dead people. It's. Oh, yeah, I made a movie about it. It's great.
CALLI: That's it? That's the story. Yeah.
NATE: Okay. No. Okay, so there were some really interesting results of a recent study led by researchers at the University of Manchester and Leicester in England, supported by the National Physics Laboratory. And the researchers were studying fruit flies when they discovered that fruit flies have something that might sound a little familiar to you: Magneto Reception.
CALLI: I knew it flies are X-Men.
NATE: Yes.
CALLI: No. Okay. But actually, if I remember correctly, if I remember correctly, that was something we talked about where birds have a sort of sixth sense where they can actually detect magnetic waves and rays from the earth, right?
NATE: Yes, that is exactly what it is. And much like the birds in that study, fruit flies use magneto reception for navigation. And that's a study that the England researchers were aware of while conducting their own research. And they've come to the conclusion that magneto reception may actually be far more widespread in the animal kingdom than previously believed.
CALLI: Well, I mean, if it's just birds and flies, that's I mean, it's don't get me wrong, it's impressive, but I wouldn't call that widespread.
NATE: Oh, so about that. It's not just birds and fruit flies. It's also monarch butterflies, sea turtles, salmon, frogs, honeybees, salamanders, lobsters, dolphins and all rodents. Fun fact: the sea turtle actually uses magneto reception to find their way home after a long journey away.
CALLI: Okay, I take it back. That is impressive. How did they figure this out? With, of all of those creatures, Fruit flies.
NATE: Well, they used genetic manipulation, and this is a bit complicated. Bear with me. First, they did an electrophysiology study, which is a test done to assess the heart's electrical activity. Then they studied the fly to look for a molecule that's present in all living cells called flavin adenine dinucleotide or FAD for short. After they measured the flies FAD levels. They used genetic manipulation to increase fad levels within the fly. And then they studied the flies behavior when they hit it with magnetic waves. Lo and behold, high levels of FAD made the fly sensitive to magnetic waves and more able to ride the waves to specific locations.
CALLI: I'm just wondering how you hook a fruit fly up to an EKG machine.
NATE: Different kind of parts.
CALLI: So why were they looking for FAD? And if it's an every living thing, then shouldn't that mean that everyone has magneto reception?
NATE: We don't know for sure, but that actually is what the researchers believe now. So with enough cell manipulation, they think every creature, including humans, can be induced into a state of magneto reception. FAD can, at high enough levels, release a feeling of magnetic sensitivity on any biological system. Now, here's the thing about fruit flies. They obviously look nothing like humans, but they have a nervous system that is shockingly similar to humans. It works the same to the extent that fruit flies have been used in countless studies as a way to better understand human biology. So basically, if a fruit fly can have magneto reception, there's a pretty good chance that humans can too.
CALLI: Okay. I know I keep asking you the really weird and tough questions here, but if we had magneto reception, how would we even know we're detecting magnetic waves?
NATE: It's a tough one. Magnetic fields emit very little energy when compared to, say, photons of light or sound waves. There's a protein many animals and plants have called cryptochrome, and it's the protein associated with how we absorb light, either through our eyes or skin or whatever. But it was believed for a long time that Cryptochrome was also how animals bodies detected magnetic waves. Since the presence of magnetism distorts how Cryptochrome works. But this study basically says no to that, because even when there was a limited amount of Cryptochrome in the fly, it didn't matter. What did matter was the presence of FAD.
CALLI: And FAD is the molecule that binds the Cryptochrome. Right. Okay, I think I get it. But let me repeat it back just to make sure. Cryptochrome was believed to be what triggered Magneto reception since magnetic waves can distort how we view light. But the truth is FAD triggers magneto reception alone. Did I get that right?
NATE: Maybe a lot of this is still theoretical, but that's the science as we understand it currently. And if all this checks out, then it helps us understand so much more about Magneto reception than we currently do. For example, how does the machinery of this cell even work? How much do environmental factors like magnetic waves from cell phone towers affect magneto reception? And most importantly, what are the evolutionary origins of Magneto reception? We don't know those answers yet, and we don't know for sure if humans are magneto receptive. But all the signs point to at least maybe. And if so, this new theory may open new means of research into using magnetic fields to manipulate target gene activation, which is a holy grail for experimental tools in human biological studies.
CALLI: I am just so excited to be an X-Men.
NATE: It's not.
CALLI: Don't kill my dream.
NATE: Me too. Me too.
[SFX: WHOOSH]
CALLI: Do you suffer from? Bad posture, slurred speech, feet dragging and other behaviors people associate with depression? If so, you might be depressed.
NATE: Okay. Where are we going with this?
CALLI: All right. In all seriousness, my point is that most mental health treatments rely on medicine or therapy to help patients. But there's new research that suggests there might be another way to treat mental health. And it has to do with your body.
NATE: What do you mean by your body? You mean like diet or exercise? Surgery? A brain transplant. I'm going to see a little bit more. All right.
CALLI: Yeah. Let let me back up and give you some context first. When people think of a depressed person, the first thing they may think of is someone who carries themselves like Eyesore from Winnie the Pooh. What you might not know is that all those things I named in the intro bad posture, slurred speech, feet dragging. They're all related to something called gross motor functioning, which is movement that involves the largest muscles in the body. So by thinking of mental health treatments as something that can only be solved psychologically, the question arises, are we doing people suffering from depression and other ailments a disservice by not focusing on ways the body can be treated? That's a question that a team of researchers from Aarhus University in Denmark wanted to answer in a brand new study.
NATE: Huh? It makes sense that the body would be affected just like the mind when it comes to mental illness, especially with the picture you painted of that stereotypical depressed person. But is that more of an observation or is there real science supporting that idea?
CALLI: There actually is. The Diagnostic and Statistical Manual of Mental Disorders says that a number of mental illnesses like anxiety, depression, bipolar disorder, post-traumatic stress disorder and schizophrenia might be associated with negative motor function changes. But specifically with a psychological condition known as psychomotor retardation, also known as psychomotor impairment, which refers to when your body and speech gets slower.
NATE: Okay. I mean, that's good. We have that. So with that in mind, what was the purpose of the Aarhus study?
CALLI: They had two goals. Step one, they wanted to figure out if there are gross motor function differences between healthier people and people with anxiety or depression. Step two They wanted to see whether changing motor displays, also known as a person changing their posture can affect the emotions and behaviors of people suffering from anxiety or depression.
NATE: Okay. How did they pull this off?
CALLI: They did this by prowling a few academic databases for studies that either compare contrasted motor function differences between clinical and healthy control groups, or that focused on motor manipulation within a clinical group. Motor manipulation isn't clearly defined in the study, but based on the context, I think they're referring to different exercises or therapies intended to change someone's posture. For instance, some references were made to experimental methods that change someone from a slumped position to an upright one, while others discussed changing your movement height from looking down while you walk slowly to looking up while you walk quickly after they narrowed their search down. They found 82 relevant studies that compared a clinical group to a healthy one, most of which dealt specifically with depression and five studies that investigated motor manipulations in a clinical setting.
NATE: Okay. What did they find?
CALLI: Well, for step one, whether there are differences between healthy people and people with mental illnesses, there are. People with depression are actually more likely to have a slumped posture characterized by their heads, leaning forward with a curved upper back. They also overwhelmingly have reduced gait velocity, which is a really fancy way of saying they walk slower, but they also have less of a stride. Their steps are shorter and they even stand for less time than healthy people too. As for people with anxiety, one of the studies the researchers looked into suggests they also have reduced gait velocity and step length. But here's what I actually found really interesting is that there's evidence suggesting that people with anxiety disorders have problems with their balance.
NATE: What makes that interesting.
CALLI: Because there's contradictory research as to why. Some say that people with an anxiety disorder, like a fear of heights or a panic disorder, have a tendency to sway back and forth more often when they're exposed to anything visually stimulating. But on the other hand, there are studies showing a reduced sway compared to a healthy individual. This might be explained by the differences in types of anxiety diagnosed. For instance, a phobia isn't the same as a generalized anxiety.
NATE: Wow. So what about step two? What did they find out about motor manipulation?
CALLI: So each study the researchers looked into dealt with a different motor manipulation, no matter which movement they looked into. The researchers noted that each was successful at helping improve the state of mind of the individual, so each might be able to be applied toward depression treatment, especially when some sort of counterattack is needed against lack of motivation and energy. In short, better posture equals better mindset.
NATE: Okay, So that was that was a lot. Let me see if this figured out. People suffering from mental illnesses, especially depression, have less control over their posture. And this posture issue creates issues with mobility. And a lot of these issues can be solved through changing your posture, whether it's through an experimental treatment or on your own. Right. So my last question then is, are these changes in posture something that can actually lead to depression?
CALLI: You're asking me the chicken or the egg question, Which came first, the depression or the posture? The truth is, as we say quite often on this show, more research needs to be done to answer that question. We don't know if motor function changes are risk factors for anxiety or depression or if they're just symptoms. That being said, for the time being, each of us should watch how we sit, stand, walk or move. When we do that, we should compare our posture to whatever mood we're in. And if we're in a bad or anxious mood, we can maybe ask a therapist if posture changes are something that can improve that. Otherwise, it might be time to take a mental health day.
NATE: That's probably healthy for everyone.
[SFX: WHOOSH]
NATE: Let’s recap what we learned today to wrap up.
CALLI: Pop quiz: who’s smarter, the baby or the chatbot? Shockingly - it’s babies! New studies out of NYU have revealed that although machines are capable of processing information more quickly than humans, humans as young as infants have an edge machines may not have anytime soon: the ability to identify why people do the things they do. It’s a shocking discovery on how machines consider the “context” of things, and may prove to be the biggest barrier yet in the recent AI explosion!
NATE: There’s a new sixth sense in town, and it’s not M Night Shyamalan’s - it’s the common fruit fly’s! For the first time, the magnetoreception sense has been detected in fruit flies, which is a sensitivity to the earth’s magnetic waves that enables better navigation through the world! This discovery is massive because it means that humans, who have similar biological qualities to fruit flies, might also have magnetoreception!
CALLI: Do you find yourself walking slumped over, dragging your feet while slurring your speech? Research says you might be depressed - and one of the treatments might be as simple as changing your posture. Nearly all of the studies a Danish team cited in their research on this phenomenon indicated that a healthier mindset can be accomplished by something as simple as not slouching!