A horse science roundup and using dubious brain scans as evidence of crimes
31:27 Share

This podcast is supported by the Icahn School of Medicine at Mount Sinai, one of America's leading research medical schools.

The school is the academic arm of the Mount Sinai Health System in New York City. It's consistently among the top recipients of NIH funding. Researchers at the Icahn School of Medicine in Mount Sinai have made breakthrough discoveries in many fields vital to advancing the health of patients, including cardiology, cancer, immunology, neuroscience, and artificial intelligence. The Icahn School of Medicine at Mount Sinai. We find a way.

This is a science podcast for May 15th, 2025. I'm Sarah Crespi. First up, freelance journalist Jonathan Mowens talks with me about a forensic brain profiling test, which police in India are using along with other techniques to try to tell whether a suspect participated in a crime, despite these technologies extremely shaky scientific grounding. Next on the show, scientists have been recently making strides in our understanding of horses.

from identifying the mutations that make horses amazing athletes to showing how climate shaped intercontinental horse migrations 50,000 years ago. Life Sciences editor, Sasha Vignieri, joins us to discuss new horse-related studies published in Science and how equine research has broader implications.

In India, police have been using dubious technologies for evidence of innocence or guilt in serious crimes, including a brainwave reader that has been declared inadmissible in courts. Despite the controversy, the results of these tests appear to continue to influence certain judicial decisions. Freelance journalist Jonathan Mowens wrote about this questionable crime-combating technique and others this week in Science. Hi, Jonathan. Welcome to the Science Podcast. Hi. Great to be here. I've seen a lot of cop shows.

And not so much lately, but in my past. Lie detectors used to feature, you know, on all these different shows. But most people today in the U.S. know that they're not considered court-worthy evidence. And I actually haven't heard of the other type of scan that's mentioned in the story. How do you say this? B-E-O-S? Yeah, so B-E-O-S stands for brain electrical oscillation signature, which is a mouthful, I know, but that's what it is. Brain electrical oscillation signal. So yeah, it's reading oscillations in the brain.

How is it supposed to work? What oscillations would this device be looking for?

It's a bit of a dystopian like technology, if you will. It sounds a bit like a Black Mirror episode and that's how I got interested in it. So basically it uses brain signals to try to determine if you've actually experienced a particular event and the events in question in these cases are criminal events. So whether you stabbed someone, whether you've murdered someone, raped someone, committed a terrorist attack,

So basically you are sat down on a comfortable armchair and you have these electrodes which are on your head. There's about 30 of them. These electrodes are connected to a machine and this machine has this proprietary software that is able to basically determine through its analysis and processing of these signals, whether you have actually done this particular thing, which you are suspected of doing.

How do you elicit that information? Yeah, exactly. Are you watching a movie? Yeah, exactly. So while you're comfortably sat down, you're also being played these short sentences, which they call probes. And it's essentially just like first person, very short sentences, things like I picked up the gun. I pointed it at Thomas.

whoever it is, I pull the trigger. And every time one of these sentences come up, the machine is basically analyzing these brain signals to try to determine whether you did that particular thing. So did he actually pull the gun? Did he actually pull the trigger? Did he actually do all those things? And when you combine those things, allegedly they're able to determine, well, yes, it seems like this person, Jonathan or Sarah, whoever actually did murder this person.

So there's supposed to be some distinguishable difference in your brainwaves if you are just hearing about something that happened or you're hearing about something that you actually did.

This is supposed to be detectable in the brain. Yeah. So basically there's a particular signature, which they claim to be not just one signal, but multiple signals all sort of coming together, which ultimately give you what they dub experiential knowledge, which is basically saying that you have experienced that particular event. It's a true experience. It's not something that you're imagining or confabulation or anything like that. You've actually been there, actively participated in an event.

And they claim that that signal is near infallible. It's hypersensitive. So you were sounding very sciencey up until that point where you said near infallible, but still like how much research is there for this kind of signal in the brain or this integrating process that you can see with a machine? Yes, that's essentially my, our investigation. It's been looking at what the science behind this technology has been or hasn't been. And it seems like there's

Basically, very, very little scientific evidence suggests that BIOS is a thorough technique validated by experts, not just in India, but abroad. Essentially, they rely on this one study, which was conducted between 2006 and 2008.

which has the government's stamp of approval. It looks at 110 people who conduct a mock crime. And this mock crime is kind of funny in a way. You're entering a room and you have to break open a piggy bank and steal some bit of money. And then you have to clean up the mess and then you leave. So that's what half of the people do. The other half recounted the crime, but don't actually have to perform it. The idea is to try to differentiate between the people who actually did it versus those who just know about it.

And from this study, they're able to say that they have something along the lines of 90% of the time they'll be able to correctly identify criminals, but something around 5% of the time they'll wrongly misclassify people who haven't committed the crime. And from this, they're sort of saying that this is actually sensitive enough

to be used in courtrooms, which is a whole ethical debate. But that's kind of the big scientific study that they seem to rely on. How big? Big in the sense that they claim to be big, but it's 110 people, which is not a significant number. And the experts have spoken to say that that is nothing, especially if you're considering that this is being used for very sensitive cases where people are potentially going to be incriminated partially based on this evidence. So yeah, it's also about what it's being used for. I would...

I was really curious. I want to know more about what the claims are about this device. Like, how long ago can the crime be? How good are people at fighting this? Have they done the kind of testing where they're like, OK, definitely lie on purpose? You know, that kind of stuff. As in trying to beat the test? Yeah. Yeah, that hasn't happened. And that's part of why there isn't much trust in the technology outside of India. And I think one thing maybe we haven't gotten into is also just like I've spoken to some leading neuroscientists

memory scientists included, who are saying that they're just like, memories are so complicated. It's so dynamic and there's so many different processes within memory. It's not just about your autobiographical memories. It's also about memories that you simulate or memories that are implanted, memories that maybe just go in multiple directions that can be basically provoked if someone interviews you and shaped.

that it seems almost impossible that this technology is able to tease out autobiographically produced and also truly experienced memories over all these other kinds of memory processes. I mean, it would be kind of exciting if this actually worked. Yeah, exactly. And he says, you know, if they've done this, then amazing. But I just don't believe that's the case. One key point is also that it's

considered a bit of a black box technology. So no one really knows how it operates, except the ones who are the software developers of the technology. So this is a proprietary kind of algorithm or process that's taking the brainwaves and spitting out the yes, the thumbs up, thumbs down. Yeah, no. Yeah, you're right. How is it being used in the courts? I mentioned in my introduction that it's been deemed inadmissible as direct evidence of guilt or innocence, but there's some other ways that it's still being integrated into the

into the judicial system in India? It was considered to be inadmissible in court since 2010 after this landmark Supreme Court ruling decided to create these limitations around it. But from the investigation I've conducted, I found that it's still relied upon by police significantly, especially for the worst kinds of crimes. So it's used by police, suspects of rape, of murder and all these other horrible things.

And then the reports that are being produced from these tests, while inadmissible, is a kind of loophole where bail applications, if I'm suspected of a crime and I'm brought to police or judicial custody, the rules seem to be a bit lower in terms of what can and cannot be admitted. And I found a couple of court documents where people who have been suspected of raping young girls were released on bail, in part because of Bios' evidence pointing to their innocence. The Supreme Court ruling in 2010 that said Bios is not

something that you can admit into court as evidence. What was the reasoning behind that? Was it the scientific lack or was there something else going on? Well, there was the scientific lack, but it was also the fact that it seemed to go against people's rights, against self-incrimination. It basically discusses how...

Now, BIOS as a tool, especially when it doesn't involve consent, you don't have control over what's happening in the brain. And therefore, you're automatically incriminating yourself by letting information sort of spill out from your mind and be recorded by these machines.

You say you found some evidence in different court documents describing how this technique has kind of popped up in different places. Do we know how widespread its adoption is on the police side? Separately from the courts, do we know what nationwide in India the police are doing?

It's a tricky question to answer in large part because we filed information requests to various states, to particular departments, the Ministry of Home Affairs, for instance, but they don't provide that information and they haven't provided it until now. So it's hard to know exactly how many cases have undergone BIOS tests. What we know for sure, based on publicly available information from researchers who have already done this work, is that there are at least 700 or so cases involving rape, murder, kidnapping, etc., where BIOS was used.

It's often used based on my conversations with some of these people who actually do these tests as a sort of, we have no other option. Like we don't know what else to do, but here's this tool in concert with these other technologies that maybe you want to go into. If we don't know where to look, can you give us a sense of which direction to take?

So that's how it's being implemented by police who often find it very exciting. And I think increasingly so because it's being advertised and campaigned for, if you will, by these forensic scientists and universities who adopt it. Yeah, I was surprised at how many people you found to say, well, this works, it's 99% and I have complete confidence in it in these academic institutions. Yeah, so we have spoke to a bunch of both interns and also senior scientists who

who uses technology, BIOS testers, and they seem to say that they're basically, if not infallible, then very close to that. So they're big proponents of this technology and they believe it's also like a way to save information

India from what is a widespread practice of torture. Because Bios doesn't require you to do anything other than to listen to these sentences being played at you and sit in an empty room. It's touted as this golden bullet against the widespread use of torture and other harsh interrogation techniques. And you mentioned in your story, it's being used hand in hand with these two other approaches, polygraphs or lie detectors. And

narcoanalysis. So we talked about polygraphs. Those are not considered extremely reliable for detecting truth from lies, but narcoanalysis I hadn't heard of before. How does that work? Yeah, it's even more bizarre, if you will. I didn't realize this actually happened. And when I found out, I wasn't even sure which direction to take the story because it seemed like narcoanalysis was maybe even more of a strange thing to be doing. Essentially, it's a truth serum.

Suspects are told to lie down and then they're injected with a bunch of chemicals, which put them in a daze and makes them more likely to speak about what they know, allegedly. And so when they are in this daze, these experts are just there listening and asking them questions and interrogating them and hoping to tease out information from these suspects with the idea that this information will be reliable and safe.

potentially be used. So BIOS is used both with lie detection and in concert with a narco analysis. And all three of them are then almost like a package, a trio, where you can help police take the investigation in a preferred direction. It does seem like from your story that there's a push to continue doing this though, from both the companies and institutions that carry it out, but also from the government itself. Yeah, it's very much...

Like I said, used because it seems to be seen as this very, like a truth obtaining machine and it's being pushed in various places, not just in India and domestically where it's being sold in new forensic science laboratories, but also advertised to police forces in different countries. So I found that there's one university called National Forensic Science University, but also the company are basically reaching out to Tanzania police or Singapore or

or Guyana police force to basically tell them about how these technologies work, how they could be implemented in their own police investigations. That's kind of an eerie thing because if it's widely accepted in India and then you see it spread beyond India, but also to South America, to Asia, to Africa, that seems to be suddenly not just a domestic problem, but also one that will become potentially global. Well, hopefully this story will reach some of the right ears and maybe we'll see

a slowing to that global rollout. Thanks, Jonathan. Thank you. Jonathan Mowens is a freelance journalist based in Rome. You can find a link to the story we discussed at science.org slash podcast. Stay tuned for a conversation about the latest in equine research. This week in science, we're doing a roundup of horse science. The Journal has published a number of fascinating equine papers recently.

And Life Sciences editor Sasha Vignieri is here to talk about some of those findings. Hi, Sasha. Glad to have you on the show. Hi, Sarah. Good to see you or talk to you. Okay, so I'm going to start with the most recent works and then go backwards. This week, we have a paper coming out on Beringia, which is where the Bering Strait is today, and climate. This is research looking at the paleo record going back tens of thousands of years. Yeah, that's right.

Well, so why is Beringia such an important part of the horse story? Well, Beringia is a really important part of the story because horses actually evolved in North America. So they have a very long, many millions of years evolutionary history in North America. There were many, many, many different species and including the modern genus Equus, also many, many different equus species in North America evolved.

early on. And then in the Pleistocene, and actually before the Pleistocene, animals were moving across Beringia. So the fact that horses originated in North America, but then were later domesticated in Eurasia, that Beringian connection really, really interesting and important. Right. And so this study uses ancient DNA where researchers looked at genomes on both sides of this land bridge that coincided

that comes and goes with climate to talk about how the populations have diverged over time. So can you talk a little bit about some of their big findings from this work?

Yeah, so the main finding of this work was that over the past 50,000 years, horse populations have gone back and forth across Beringia multiple times. And they also looked in more depth at how that related to climate. And so we know that the ice sheets were covering most of North America for a good portion of that time. But then at some point, an ice-free corridor occurred.

cleared and animals moved into that. And this paper really looked at how that change in climate also related to movement of the horses

particularly in North America, where the habitat really changed. So horses evolved in climates that were sort of arid and had a lot of grasses. And then post-glaciation, those regions were wet and not great for horses, really good for things like moose. And so the authors looked at this change over time. And it's not...

hugely surprising that these animals were moving back and forth across Beringia all that time. But it's important, especially for indigenous North Americans who have really seen the horse nation as a partner for, they say, time immemorial. So it was a very important finding for them to show this back and forth movement between North America and Eurasia. Did people do this too? Is there some evidence in gene flow in both directions across the land bridge in humans?

Yeah, so I know less about people, but yes, I believe there is. I mean, of course, it makes total sense that there would be. One of the big takeaways from this paper, actually one of their big messages, is that we really need to be thinking about how animals and people, because people are animals, move as that relates to their habitat.

And so instead of just thinking all the time about, let's say, oh, a horse has this range or an ancient hominid has this range, we need to start thinking about the fact that these animals were associated with habitats and those habitats change over time and that leads to movement. And if we want to think about future and preserving animals besides ourselves, we need to really think about maintaining this ability for these animals to follow their habitat.

Oh, really interesting. So a big lesson from horses. And the other horse paper I want to talk about came out at the end of March. And it's about key genetic changes in horses that increase their ability to get energy from food and oxygen and become such amazing athletes. And from this paper, I learned that horses have the highest level of oxygen consumption by weight.

This big shift in energy production happened in their mitochondria. Big surprise. What are horses doing differently at the cellular level? Yeah, so just to say, I'm not the biochemical expert. And so my description of that part is going to be pretty general, but that's OK. We had a lot of biochemists look at this paper. But basically, horses diverged from rhinoceroses or those lineages about 55 million years ago.

And sometime after that diversification, this change in the KEEP and the NRF proteins occurred, or the genes that underlie these protein translations occurred. And so we know that this happened a long time ago because we don't see the change in rhinos, but we do see it in all extant horses. So horses, zebras, asses, they all have this change. And so it happened

very early in evolution. And so the general finding is that there was a change in the stop codon that led to a change in this protein that leads to increased ability to reduce oxidation. And so let's just slow down. So basically the stop codon, it basically allowed the protein to be translated slightly differently than earlier versions at this change in the gene there. And it increased its sensitivity. So basically,

They're just a little bit more sensitive to free radical species and they clean it up so they can kind of use energy more intensely because a byproduct of energy production in the mitochondria is this reactive oxygen species. And so they're more sensitive to that and they have a better cleanup crew, a better cleanup response.

Exactly. And it also facilitated increased efficiency within the mitochondria. So the horses both have more mitochondria in their muscles. They work better and they're better able to get rid of the oxidation effects. And it's interesting because the stop codon you would normally think reduces function.

But in this case, it just altered it. And so that part is, again, less of my purview than other editors we have, because that part has implications for a lot of things like biomedicine and how we think about how these genetic changes can lead to differential expression and differential function.

And I think the big exciting part was that the change led to a replacement of a cysteine. And that actually is something that hasn't been seen in eukaryotes and has only been seen in bacteriophages. So it was a pretty exciting finding with regards to potential opportunity for future research.

Yeah. Okay. One more, Anne. Yeah. These proteins that are different in the horses might actually be important in people and some of the mitochondrial or energy metabolism disorders. And so like they kind of make a new therapeutic area of interest. Exactly. Right. So cool. Horses. Yeah. Yeah. Horses, who knew? But in terms of horse evolution, it's interesting. Like we think about horses in terms of sort of these extant

ones that we know about, right? But I mean, there are hundreds of species of horses over millions of years, and it's possible that most of them had this change because all we know is that it happened after they split from rhinos 55 million years ago. And it's too bad we don't have high speed, high energy rhinos, isn't it? Yeah. Well, I love a rhino, but I'm not sure people would appreciate it.

If a rhino can chase you down, you might be in trouble. Yeah. So the last little bit I want to touch on is a few papers that we've published on science that kind of have to do with the domestication of horses. So how they've interacted with people over thousands and thousands of years. And so back in 2009, there was this idea that horses were first domesticated by the bowtie people. This is in 3500 BCE.

And there was evidence for husbandry and milking.

But it turns out that those are not the ancestors of the domesticated horses of today. So where do we think that domestication of the horses that, you know, I can go and pet, where did that happen? Yeah, so my understanding is we, I mean, still sort of that middle part of Eurasia. So that's what the current thinking is, that it came later and that this new lineage of horses had some unique changes, particularly having to do with temperament.

as well as genes that are related to back function and in humans and pain. And so they think that this lineage of horses was really selected for these traits. And at that point, they call it the domestication two lineage. That's when horses really took off in terms of

being widely used by people to ride and pull things and in all the ways we use them now. Yeah. And these papers really showed the importance of ancient DNA because the earlier papers were focusing on marks on bone and evidence of milk. But once you get the DNA from very, very old samples, you can say, oh, no, these are what is it? Prezbaluskis? Prezwalski. Prezwalski's ancestors as opposed to the domestic horse.

Just to go back to kind of what we talked about at the beginning about the horses moving back and forth across the land bridge, the presence of domesticated horses in the Americas was something we published on, I think, in 2023. This was by a lot of the same people as are doing this land bridge study this year.

And they kind of wanted to talk about where we put it this way. Our horse is invasive in the Americas, right? Right. And I mean, I think the easy answer to that, even before these papers, is no. I mean, we know. So it depends on who you talk to with regards to when they may have gone extinct in North America. But archaeological evidence suggests 12,000 years and some DNA evidence suggests as recently as 5,000. So that's like a blink of an eye in terms of evolutionary history.

But the paper that you're talking about, they actually looked at sample horses within North America and Eurasia to try to figure out, is there any remnant of Pleistocene horse DNA in modern horses? And in part, because many of the indigenous peoples have this traditional knowledge that they've been in association with these horses for a long time. And so that paper was interesting because although it didn't find

evidence of that at this point. That would be the idea that horses evolved in the Americas, went across the strait, mostly extinct in the Americas. They came back in through European colonization and then there was some crossing with like extant...

very small populations of North American horses, but this is not what they found in the people. Yeah, so that was a hypothesis that they put out there that maybe there was some persistent lineages of Pleistocene horses that were actually in continued

with these human groups. And I'd say, although they didn't find that there, we just don't really know. I mean, we haven't seen... We're still finding ancient humans that we didn't know about. Right. Ancient hominins that you didn't know about. Yeah. I mean, we have so much to learn. We know just a little scratch of the surface of so many things. But yeah, so they didn't find that in the samples that they looked at. But what they did find was that there was a lot of evidence that

European-derived lineages actually were in association with indigenous peoples well before colonial testimonies say. And so what it seems like happened, at least with the samples that they looked at in that paper, they came over with the Spanish and they actually moved all the way through North America, probably through indigenous routes. And rather than most Western interpretations say that they

to be distributed in the late 1600s after this revolt in New Mexico. But what that paper showed is that they were actually well integrated before then. Hundreds of years, probably, right? Exactly. Are there...

There is feels like home. Well, and also, you know, if you think about it for the people, too, I mean, humans were definitely in the area 5000 years ago. So if they were associated with those horses before they went extinct in Pleistocene, North America, those lifeways may have been ready for them to reintegrate. How amazing would it have been to have known of horses but never seen one? Yeah. Can you imagine?

All right, Sasha, any other horse science we should touch on before we break it up?

What I've noticed is that horse science is just a continuing burgeoning area of research. I mean, people love horses. Scientists also love horses. Like there's all these new things to find out. So I guess I'd just say stay tuned. That's wonderful. Thank you so much, Sasha. Thank you, Sarah. Sasha Vignieri is the life sciences editor at Science. You can find links to the papers we discussed at science.org slash podcast.

And that concludes this edition of the Science Podcast. If you have any comments or suggestions, write to us at sciencepodcast at aaaas.org. To find us on podcasting apps, search for Science Magazine or listen on our website, science.org slash podcast.

This show was edited by me, Sarah Crespi, and Kevin McLean. We had production help from Podigy. Our music is by Jeffrey Cook and Wenkui Wen. On behalf of Science and its publisher, AAAS, thanks for joining us.