Cognitive biases: The bugs, features and zero‑days of the human mind - EMF 2026

All right. Hello everyone. Welcome to this talk on cognitive biases. Uh my name is Karolina Chara. It's my great pleasure to be here. Um I'm a behavior scientist and as behavior scientists we try to understand human behavior and why people behave the way they do um and also predict their behavior and change their behavior um at times. And today we'll focus on cognitive biases, which is just one topic within behavioral science. We're gonna explore what they are. Um, >> yeah. Oh, that's better. [laughter] So, we're going to explore what cognitive biases are um and how they they can get in the way and what we can do to um use them to our benefit. And we're going to start with a short activity. So, for those of you who can, if you could take out your phones and I'm going to ask you to scan a QR code and that's I'm going to ask you to scan a QR code a few times and and get engaged in some activities. First, I would like for this one I will split you into two groups. So, let's this side of the room be group A and this will be group B. Can group A raise their hand? Beautiful. And group B. Amazing. So, now you have two QR codes. if you can scan one of them. Group A, group B. Um, for those joining now, hello. If you can scan one of the QR codes, just one, please. And there'll be two questions. Um, and don't worry if you don't know the answers. Um, it's expected that you may not um just don't overthink it. Try to answer the best you can. Try not to Google the answers. And I promise it's going to make more sense later on. Um the second act the second question if you could put digits that would be ideal. Uh for those joining please scan one of the QR codes and if you can answer the questions that would be much appreciated. Um I'm going to give a few more seconds because I can still see people working. Okay. Right. Let's forget about the activity for a second. Okay. For those joining, please scan one of the QR codes and answer the questions. But I'm going to move on now. Um, let's forget about the activity for a second. I promise it's going to make more sense later. But now I just want to ask to explore what a cognitive bias even is. And I will ask you to scan another QR code. No groups anymore. And it's the last time I'm asking you to scan. I promise they should work for all other questions. And if you could put just a short description, a few associations that come to mind what a behav what a cognitive bias might be. They should be appearing on screen if everything is working. Perfect. Okay. Okay. They're coming in. Prejudice is the main one. I need to be aware. Scam. Okay. Inaccuracy. Implied. Okay. Oh gosh, there are loads of them. That's great. But prejudice is the main one. anchoring subconscious assumption preconceptions. Okay, that's super interesting. So, I'm going to tell you what my definition is. Um, it's a systematic error in judgment that would impact our um decisions and how we interpret information. And the most important thing, I guess, is that it's a systematic error. So if I had thought that this talk was at 11 rather than 10, well this would have been a judgment, an error in my judgment, but it wouldn't be a cognitive bias because it's not a systematic error hopefully. Um and for something to be a cognitive bias, we expect it to be kind of predictable and to um present itself in various circumstances. So now I'm curious if you think you are more or less biased than the average person. Um, if you could answer. Okay. Equally biased. A few people are less biased. Also, Mo. Okay. Okay. Most people will say they're equally biased, but which is fair, but I just wonder if that's going to change at all after the talk. Okay, that's fair enough. A few people are less biased. Um but most people are equally biased as the average person and some people said more biased. Right? So let's dive into specific cognitive biases and let's consider two scenarios. So imagine I'm asking you to put your hands in cold water for 60 seconds. 14 Celsius degrees is cold water is going to be painful. So imagine I'm asking you to do this for 60 seconds. And then another scenario is I ask you to put your hands in the same water for 60 seconds, but then for another 30 seconds in water that is slightly warmer, warmed up gradually to 15°, which is also painful and unpleasant, but a bit less. So, which scenario would you prefer? And again, if you can vote, would you rather keep your hand in 14 Celsius degrees um for 60 seconds or 14 Celsius degrees for 60 seconds and then for another 30 seconds gradually warmed up to 15 Celsius degrees. Okay. Well, the majority say 14 Celsius degrees for 60 seconds, which is fair because why would we want to keep our hand in cold water for longer if we don't have to? Well, luckily this has been tested and in a similar experiment. Uh participants were asked to do exactly that. There was a short trial where they would put their hand in cold water for 60 seconds. Then they would also put their hand in water for 60 seconds and then another 30 seconds of s uh less cold water. And as you can maybe figure out from this slide title and the published paper as well title is that most participants preferred to repeat the longer trial when they were asked afterwards which one was better. And if you ask yourselves why that's the case, well we call it the peak end rule. So when we think about past experiences, we don't evaluate them based on some average of our experiences or kind of the duration. It's more about the most intense moment whether it's positive or negative and how the experience ends. So in this scenario, the experience ended on a slightly more positive note. It was slightly less painful. So people pref remember it as more pleasant and were more willing to repeat the longer trial. So how can we use it to our advantage? Well, we can for instance avoid scheduling unpleasant meetings or unpleasant events at the end of our days because we're going to remember those days as better. Uh we can try to focus on positive elements of the experience. So if you think back and there is something negative that happened, you can try to focus on a positive peak because the peaks are going to determine how you remember the experience. Well, think beyond the middle. So think how the experience going to end and if there's going to be an um an intense moment, try to make it positive rather than negative. And try to save good parts to the end. And maybe there is a reason why desserts are served after the main meal and so on and so forth. So try to add something at the end that's going to be pleasant. Okay, another voting. So now imagine that I'm asking you to choose between guaranteed gain of 900 or 90% of chance of gaining a,000 pounds and a t% of chance of gaining zero. Which one would you go for? should be the same QR code you used before. It seems to be working, but you can still scan it if you're joining now. Okay, so almost everyone prefers a guaranteed gain of 900. What if I ask you about losses? If I tell you you have a choice between a guaranteed loss of 900 and 90% of chance of losing a,000 and 10% of chance of losing zero pounds, which one would you prefer? Well, it seems to be reversed now. Okay. So, this is a summary of what I've just asked you. There are very two very similar scenarios, but one is about gains and one is about losses. And when I ask you about gains, you were going for a certain choice. Why would you risk not gaining 900 pounds if I was telling you you had a guaranteed option of getting 900? But when I was telling you about losses, you were not so willing to take a guaranteed loss of 900. And this is kind of what um I would expect. And also note that within each of these examples one and two that the expected value of one and two is identical mathematically and I'm trying to be careful what I say here because I know there are some mathematicians in the room. Um but that's to be expected and we call it the reflection um effect. So this basically describes people's attitudes towards risks and they are reversed depending on whether the outcome is framed as a gain or a loss. So kind of as you saw and this has been uh researched uh um many many times when people are facing potential gains they are risk averse they would rather go for a certain outcome and when they are facing potential losses they would rather gamble and are risk seeking. And you may wonder why it's important. Well, it's important because if you come across certain information like yogurt that is 90% fat-free, that of course means it has 10% of fat. Or if I tell you that a treatment works in 95% of cases, it doesn't work in 5% of cases. So if your decisions depend on how I frame it, well that's problematic. So what can be done about it? Well, you can simply ask yourself if you would make the same decision if I frame something as a loss or a gain depending on what you are facing and then see if you would make the same decisions. Okay, now it's time to look at your answers um the the activity that you participated in in ad in like at the very beginning. So you may already know that I asked you whether the population of Iran was more or less than four thou four million people. What you may not know is that I asked half of you uh if the population was more or less than 190 million people. So thank you so much. Okay, these are the results. Um so I was actually interested much more in the second question. So what the population was your best estimate and as you can imagine the group A estimated 30 million and group B estimated 97 million. I didn't run any statistical analysis, but it seems like the first number I presented made some impact on your decision and estimate in the second question. Well, as I said, it wasn't statistically compared, but don't worry because it has been tested. So, in a similar study where people were asked exactly same two questions across multiple experiments, we can see this trend. When you were presented with a lower number, the lighter color, the estimates were lower compared to the higher number. And not all of them were significantly different statistically um different. You can see the mark with asterisks, but you can see that there is um a trend. So we call it call it the anchoring bias. So this is our tendency to make decisions and be influenced by based on the first piece of information that we encounter. So, as it's illustrated in this example, if you see um headphones that cost £60, but they have been discounted for from £100, that's a slightly different scenario to when you just see 60 because in the first case, the first piece of information that you encounter is that they are worth £100. So £60 in comparison looks like a good deal but not necessarily in when in the second scenario when you have £60. So you can reflect and think if you can think of any other examples. I'm not going to make you uh use slido for this one but a few that come um to mind for me are black Friday prices which is similar to what we see in the image. So we are often made believe that something cost more and then was discounted. So the first piece of information that we encounter is a higher price and we think like something is worth more. Many of you may know that the price of an engagement ring should supposedly cost three times your salary which is again something that we were made believe as a starting value. And some restaurants would also place um overpriced items on the menu first within categories like mains because then it sets your expectations on how much a meal should cost. Right now let's think if you would prefer a chocolate bar or an apple and what would you choose? what you would choose now and what you would choose if I was to ask you to choose now for tomorrow and if you think the answer will be different where you are not alone we call it the present bias which is our tendency to place more value on immediate benefits over future benefits that's why it's hard to go to the gym or eat healthily or sleep more or save more for retirement because we need to put an effort and cost up front for something that's going to pay off in the future and we are wired not to. So as a result, we can often choose what feels good over what feels what is better for us later. And I'm choosing this one because it's a very common one and I think applicable and a lot of us struggle with that and there are things that can be done. So you can add immediate incentives. So rather than choosing the hardest workout at the gym, you can choose something that's maybe less effective but something you like and then you're going to have this immediate benefit in addition to the future benefit and you're going to be more motivated to go to the gym. Um you can also decide before the temptation arrives. So maybe it's better to choose your lunch for tomorrow today if you can. Um you can also try to reduce friction for good choices. If good choices feel hard, let's make them easier. And using defaults can also help. So once you sign up for something, just make it work. That should help as well. And we talked about the peak end rule. So I wouldn't be myself if I didn't try to end um with a positive uh on a positive note. And this is an initiative called save more uh tomorrow. So, I didn't want you to think that it's just the cognitive biases are just for us to explore and find quirky um effects, but they're are also very applicable not only on a personal level, but also on a wider level. So, it has been recognized that a lot of Americans don't save enough for their retirement. And this initiative um encouraged people to commit to saving more for their retirement but not from today but from a certain point in the future. And this helped overcome the present bias because people didn't have didn't feel like they were putting money um aside today but this was something they were doing for later. But this also took um advantage of this default and inertia because once you signed up, you would have to opt out. It wasn't something that they would ask you constantly to um to do. And we also talked about loss aversion. We don't like losses. So this um commitment to save more was always linked to people's uh raises in their salaries. So they never felt like they were suddenly having less money because they would save more only if they started earning more. And this helped a lot of people earn more which I think is a great example of how we can use um um overcome at least cognitive biases but also somewhat use them to our uh advantage. So a few uh takeaways. I don't want to leave you with a feeling that cognitive biases are bugs and we are somewhat flawed because they help us uh navigate the world. Huristics and shortcuts are useful otherwise it would have been very difficult um to navigate. So they are there for a reason but of course they can sometimes um get in the way as we saw. So if you want to make sure that you are not as susceptible to them as possible um well you can slow down important decisions. So we know that cognitive biases happen mostly when we are rushed or when we are under pressure or if we are tired. So for important decisions try to take your time. You can also try to use uh outside perspective. Um, so try to maybe ask someone else for opinion. If you're buying a house, maybe don't rely on the first price that you saw, but do your research. Um, seek evidence that kind of uh is um disisconfirming your beliefs. So maybe um try to think what would I need to observe to think that I was wrong. And if you have any checklists or systems in place, that helps as well because you think them through in advance and then you can use them and don't you don't have to decide in the moment where uh cognitive biases are more likely to happen. Right? So back to the initial question and again um there's not going to be um a a slideo to scan but I wonder at if um your answer to this question would have changed at all. Uh it's for you to decide if you are more or less biased than the average person but the reason why I'm asking is that there is something we called we call bias uh blind spot. So people commonly perceive themselves to be less susceptible to cognitive biases than general population. Um well interestingly you went for the option that you were equally biased most of you but that's if I was to ask you are you more biased or less than the general population statistically on average anyway people would think that if anyone they are less biased and I think after this talk you may start noticing cognitive biases everywhere. Well, and perhaps especially in other people. Um, thank you so much. If you have any questions or comments, I will be in the Q&A uh tent. You can also email me on info@unstackscience.com. And if you can spare a few minutes to uh give me a feedback using the same QR code as before, I would greatly appreciate it. Thank you so much. [applause]