How to spot and fight misinformation

Hello, I'm Alex Hughes and this is the Insingenious Podcast. The Bite Size Masterclass from the BBC Science Focus magazine. The internet is a flood of misinformation as friends, family and strangers pitch to you the latest scientific findings or events that have happened around the world. How do you decipher what is real or not? We spoke to Alex Edmunds, the author of the new book May Contain Life to Find Out. He explains how our biases lead us to full misinformation, how to overcome it and what to look out for with news articles, books and any new information. So in your book you address both how we typically fall for misinformation and how to overcome it as well. This falls into three sections starting by addressing our own biases and how these can be activated. Could you explain a little bit more about these? You state them as the confirmation bias and black and white thinking. Certainly, so let me start with the confirmation bias. So this is the idea that we have a view of the world and if we see evidence that confirms that view, then we will lap it up uncritically even without checking whether it's rigorous. For example, if I were a climate change denier and there is a study which comes out saying climate change is a myth, I will shout it from the rooftops, I will share this even without looking at the actual methodology. And the flip side to this is if I see something that contradicts my worldview, I will either not bother to read it to begin with or I will read it with the mindset of trying to pick it apart. So again, if there was a study saying that climate change is man made, if I choose to open it, I would go through it line by line and think about all the potential holes in it but not apply the same scrutiny to something that I want to be true. So confirmation bias applies where we have a pre-existing viewpoint, but there might be many things where we don't have a pre-existing viewpoint and that's where the second bias comes in which is black and white thinking. So this is the idea that we can view something as either always good or always bad. So one example might be carbs. So when you eat, you think that protein is good, you learn at school that it builds muscle, you think that fat is bad, it's called that because it makes you fat. But carbs aren't so clear-cut. But black and white thinking means that even if you're not sure, you think it's either one way or beyond that. So we're aware of our biases and how they could be triggered and we know misinformation typically tends to play on our emotions. How do we then break down this information to better understand what is, I guess, true and what's not true or what has a bit of added spice to it? In such an information heavy world especially, you detail what you call the ladder of misinfurrence in this. Could you sort of touch on that a little bit more as well? Yeah, so absolutely. So what I'm trying to encourage is for you to apply the same discernment to a study that you do like as one that you don't. And so this is really important because you might think, well, how do I check the facts? I'm not a statistician, I'm not going to suddenly do a PhD in statistics. But what I'm trying to highlight is that the knowledge is already within you because people are ready to exercise discernment if they see a study they don't like. Let's ask the same questions and apply the same skepticism to a result that we do like. And so one practical tip I give in the book is if there's a conclusion you were willing to lap up, let's imagine it had the opposite result and see how you would tear it apart. So let's say I would love an excuse to drink lots of red wine this evening and I find a study which claims that red wine leads to longer life. So if I imagine the opposite, how would I react if the study said red wine leads to shorter life? I would say well, there could be a lot of alternative explanations for this. Perhaps the people who buy red wine are poor people. They can't afford champagne, they can only afford red wine and it's that poverty that leads to shorter life. Now that I've alerted myself to the possibility of poverty being an alternative explanation, I should ask myself does this continue to apply even though the result is in the direction that I want. Red wine is correlated with the longer life. Well maybe the people who are drinking red wine are wealthier to begin with, whereas everybody else can only afford beer and it's that wealth which is leading to the long life. So the idea of imagining the opposite is to trigger you, to ask the same questions as you would do for a study whose conclusions you don't like the sound of. And how do we go about, you know, when we see something that's maybe just a headline that says the same sort of thing as you're mentioning there, how do you go about checking this sort of information? Is it a case of following something back to its root and finding out the information where it began? Yeah, so that is important. So sometimes when a statement is quoted, we take this as gospel truth. Sometimes it could be there's a footnote at the end of the statement and we think, oh, if there's a reference, it must be true. But there are many places in which people are misquoted and I have been misquoted myself in something as formal and official as a government inquiry. So from time to time, there's government inquiries where they invite people to submit some evidence to the inquiry. And there was one inquiry into executive pay where I saw the final report and the final report claimed that chief executives add very little value to a company and they had a footnote and a footnote referred to me. And I was surprised. I thought I would never say something like this. I was a bit flustered. I wondered whether I put a typo in my original submission. Maybe I'd added the word not in by mistake. But no, I had written what I believed to be correct and I had just been misquoted. Sometimes the misquoting might be just cutting out what the original source was. So there was the famous book Why We Sleep by Matthew Walker, which has been a huge bicella. And this is trying to claim that sleeping more always leads to better outcomes. And so there's a graph in that book which shows that the more hours you sleep, the fewer injuries you get in a study on teenagers. So nine hours of sleep led to fewer than eight, which led to fewer than seven, which led to fewer than six. But if you go to the original graph, the original graph had also a bar for five hours of sleep. And five hours of sleep leads to fewer injuries than six and fewer injuries than even seven hours. So often we want to go back to the underlying source. Now you might see well this is impractical. We would be at a whitsend if every time we had to check back to the original source. But if it's something which is particularly accurate and also something which triggers our biases, which might mean that the writer knew that he or she could play fast and loose because people wouldn't check it, then it's particularly useful to be sure. So is there almost a case of, you know, if it's too good to be true, do the deeper research on something? It is. If it's too good to be true and you want it to be true, then you might want to be skeptical. And obviously we're talking about this very much in connection to the digital age, you know, headlines, articles online, all of this sort of stuff. Is this misinformation a new experienced cause by the era? Or is it just more that we're seeing such a rapid expansion of it and it's so heavily in front of us all the time? Yeah, so I think the issue is that we have so much information out there that it's easy to be misinformed. Now that might seem a paradox. You might think isn't it the opposite, the more information that's out there, the more likely you are to be informed. But it's not because again of confirmation bias. So when I mentioned confirmation bias earlier, I talked about the biased interpretation of data once we receive information, how we interpret it depends on whether we like what it says. But there's also bias search, what information do we search for to begin with? We will only search for evidence that confirms our point of view. So again, if I was a climate change denier, I could do Google for why climate change is a hoax. I'm sure I could find a lot of studies, some claiming to be scientific to state that where so I was to Google the opposite why climate change is real, I'd get quite different studies. So the accessibility of information actually can make us less informed because it means that it's much easier to find evidence that supports our viewpoint. We often hear this phrase banning around like evidence shows that. A study finds that research proves that. But this is all quite meaningless because you can almost always find a study to show whatever you wanted to show. There's a weird argument of do you want to find truth because for a lot of people there might be a slight awareness that there isn't a complete level of truth or the argument they're pitching for isn't always correct and they'll track down the angles that work for them. Why do we need to find truth? Because these are things which affect our daily lives. So you might think isn't this just academic? I'm an academic, I'm a research professor, I spend time scrutinising academic papers but the average person doesn't do this in their job so does this really matter, is this not academic point scoring. But it does matter because research is everywhere. You're acting on the basis of research. So what is the best food eat or food to avoid and if the research claiming that something is a superfood is actually not correct? then you're just spending a lot of money on an overpriced superfood which might be unpalitable you're a slave to a diet which is not even working. If you want to train for a London marathon because you're inspired by the marathon last month, what is the best way to do that by finding the most efficient training regime then you can put in far few hours and also achieve the same results. One genre of books which is really popular right now is the self-help or self-improvement so does waking up at 5am every day transform your life well if it actually doesn't then knowing this is really important because otherwise you'd be needlessly waking up so early. And we're talking about this a lot from the angle of I guess how to identify this information or false facts but as a society as a group is there a way for us to simply end it or is this just sort of wishful thinking an impossible task? Unfortunately I think it's impossible to try to end it so when you hear about misinformation people might think why can't you regulate this? Can't you just have prosecutions for people who are releasing untruths just a bit like fraud? But this is difficult because one of the things I stress in the book is even if something is 100% accurate it could still be misleading because there could be multiple interpretations. So if indeed people who drink red wine live longer if that is a fact then anybody who states that fact is not being untruthful however if people infer from that that drinking red wine causes you to live longer then that is incorrect if in truth what's going on is wealthier people drink red wine and wealthier people live longer. Another example is the exception that doesn't prove the rule so Simon Sinek has the third most viewed Ted talk of all time on how starting with why leads to success and he says well Apple is extremely successful and Apple started with why it has a purpose it has a mission but that is one single hand picked example there could be hundreds of other examples of companies that start with why and failed but you never hear about them because Simon Sinek will never mention them because they don't support his thesis. So what he does say is completely truthful you can never prosecute people for not saying something and not giving the full picture so this is why in the book I highlight that the only this is on us to try to address misinformation by being aware of our own biases and doing our own research. What happens when something has become I guess so cemented as a belief that it's almost become a fact in itself you know how do you reverse something like that the example I think my head is the idea of MSG and it's like health concerns something that's become so much so cemented is basically a fact to a lot of people. Yeah so this is really interesting and this was covered in a great podcast by Katie Milkman a professor at Wharton I used to be a professor then myself. Her podcast is called Troy Solagy where she talked about the idea that if something is repeated many many times then people think it's true just because of repetition and so what the podcast highlighted was the importance of going through to the original source and so what the podcast interviewer he did is they went to the original source and this original source would I think just an opinion piece it was not even a scientific study suggesting that MSG was bad when there was very little evidence behind this but because of this Chinese whispers phenomenon people have thought that this was actually true and you see this not just an MSG but something even more serious so the opioid epidemic within the US one of the most cited papers which apparently claims that opioids do not lead to addiction was something published in the new England Journal of Medicine you might think well that's a great scientific journal but this was never a study it was just a letter to the editor and so it was cited without people realizing it was just a letter they thought it was an article in the new England Journal of Medicine which claimed that there was no addiction lots of people cited this and this was an impactful article in encouraging the uptake of opioids with devastating consequences so 650,000 people have died in the US alone over 20 years this is millions when you look worldwide. We're talking about this idea of addressing a biases and trying to understand when you might feel like you've chosen to feel a certain way or that you might be tricked by some bit of information for a lot of people this sort comes down to a conversation with someone else where they don't feel like that person is right or that they feel like someone is misunderstood what do you do when you're speaking to someone and you feel like they've fallen for misinformation and they can't see that. I think it's to try to highlight the alternative viewpoint and this is often not best done by telling them the alternative why because if you want to recognise that that person is biased then they might not be listening so I saw this phrase once which is before you want to argue with somebody ask whether they are in the right headspace to consider the idea of an alternative viewpoint because if not sometimes there are certain people that you can't argue against and no matter how much you want to it's not actually worth trying so sometimes you can actually back people into a corner and make them even more defensive but if the person might be amenable you might just ask them questions so often actually asking questions and getting people with strong positions to explain them in detail might cause them to realise that their positions are not strong as they thought and so there was actually a scientific study which asked people to rate their knowledge on things such as how does a toilet flush work how do piano keys make sound and people rated their knowledge as like 9 out of 10 or 10 out of 10 and then they asked them to write an essay describing this and actually many people were not able to describe this and after failing to write the essay they were asked to re-rate their knowledge again and they rated themselves more lowly so if somebody is to say I think climate change is completely man-made all the evidence suggests that it's not happening just ask them to walk through what the evidence is not in a ha ha ha way to try and catch them out but be generally inquisitive what is the reason for why they think that particular way and sometimes you might realise that you are not always correct so in the Brexit referendum I was a very strong remainer I thought that every a Brexiter is xenophobic and uninformed but I went to a Brexit talk and the speaker I thought had really good arguments now I didn't agree with everything that he said but at least I could see that they were based on logic rather than just hyperbole and so even if listening to an alternative viewpoint you think 90% of what they say is wrong maybe 10% of what they say is right and maybe that 10% leads you to come away more informed than you started with and why is that some people are more likely to fall from this information is there some sort of predisposition to this kind of thinking? I think it's just the strength of your existing belief so if you have a strong viewpoint then confirmation bias is at play so Leo Tolstoy wrote that the most difficult tasks cannot be explained to the most intelligent man if he has firmly decided that he knows already but the simplest thing can be explained to the least informed man if he is willing to learn so it's often not your intelligence but the strength of your biases and it may will be the case that more intelligent people are more biased so sometimes listeners might think I'm so smart I would never fall for misinformation but there are studies which find that actually sometimes intelligence makes the problem worse why because people deploy their intelligence selectively they use their smarts to argue against any evidence that contradicts their worldview but they suddenly switch off their critical thinking faculties when they see a conclusion they like the sound of. In your book you they are a pretty comprehensive guide to better preparing someone to deal with misinformation can you talk us through I guess a bit of an easy checklist for what you call smart of thinking? Certainly so the first is to ask do I want this to be true because if I do then my confirmation bias might yet play and so I might need to be a bit more discerning so I think this is a good place to start because we can't go around checking everything we need to be realistic so what are the cases where I particularly want to devote my energy then what I want to look at is well what am I trying to check if it's a statement then let me check is this a correct quote was there something which is cut out from a bar chart so to go to the original source when I've gone to the original source I want to look at is this a single example or is there large scale data behind it so if somebody is telling me Apple started with why an Apple is successful and I go to the original source and I look at Apple's market value and I see it's massive that is true but even if it's true it's misleading because it's only one single example let me look at large scale data if what I'm given is large scale data I wanted to look at and check is this actually evidence so what's the difference between data and evidence let's think about evidence in the trial so evidence in a criminal trial is evidence that supports one particular suspect but is inconsistent with other suspects so if the evidence suggests that Tom Dick or Harry could have killed Sarah it's not evidence because it's consistent with all terms of suspects and similarly date is not evidence if it's consistent with all turn to explanations. So go back to the example, people who drink red wine live longer. That is data. It could be gathered across thousands of people, and you find in large scale data people who drink red wine live longer, but this is not evidence because it could be that red wine causes long life, or it could be that wealth causes you to drink wine, and wealth also leads to longer life. So this is why I encourage you to imagine the opposite, to think of the alternative suspects, the rival exhalations for the same data. So the idea of misinformation, alternative facts, all of these different parts of this topic, they have become so widespread in recent years, but how do you see it looking at, and I guess the next five, ten years, especially with the rise of deep fakes, AI, all of this technology that is probably just going to make it a lot harder to identify what is the truth online. I think unfortunately misinformation is going to get even more serious an issue. And often people say, "Oh, AI will always make things worse." I think these are problems even if you did not have AI. So I'm going to come back to AI shortly, but even without AI, there's a lot of forces towards misinformation, which is just the scalability. So if you are somebody who just is giving a viewpoint, even if it's not based on data, it can go viral because of social media, and if it's something that people want to be true, then people will share this widely. So it could be something with very little scientific backing, such as 5G causes coronavirus. But if people want this to be true, if they like to believe in conspiracy theories, these things will spread quite widely. And also, how do you become famous? That people want to be true. There are many people who've managed to sell a lot of books or become Instagram influencers by peddling potential myths. But then with AI on top of this, this could work in two ways. So on the one hand, AI could be a solution to misinformation, because AI could try to check the underlying facts. But also AI could be a source of misinformation. Why? Because AI could be used to reduce, as you say, deepfakes. Maybe AI could write something in the style of a famous scientist or author. And also AI, I don't know this field particularly well, but how AI can think critically is not clear. So even if AI can find that data is completely conclusive and comprehensive and robust, and AI think about the multiple alternative explanations for that same data, it might not have got to the stage when it can do that. And if so, it might be focusing on data, which is 100% truthful, but it's not evidence because there's alternative explanations for that data. Thank you for listening to this episode of Instant Genius. That was Alex Edmunds on Misinformation. The Instant Genius podcast is brought to you by the team behind BBC Science Focus magazine, which you can find on Teo Now, in Supermarkets and news agents, as well as on your preferred app store. Alternatively, you can come and find us online at sciencefocus.com.