The Invisible Gorilla

• how the illusion of attention can cause us to miss important information
• why we’re not as good at multi-tasking as we’d like to think
• how the illusion of confidence can influence the judgements we make about ourselves
• how this illusion affects the way we interpret feedback
• what influence the illusion of confidence has on group performance
• how the illusion of knowledge can impact the way we approach our work
• what we can do to avoid these biases and prevent them from affecting our day-to-day decision-making
• why knowing about these illusions can help us to make better decisions

If you have access to video, it is a good idea to watch Chabris and Simon’s one-minute experiment before you listen to the interview.

Transcript

Female interviewer: Relying on our intuition is a method of decision-making that many of us often choose to adopt. But is our intuition really that trustworthy?

This is the question that psychologists Christopher Chabris and Daniel Simons seek to answer in their book "The Invisible Gorilla and Other Ways Our Intuition Deceives Us." In the book, Chris and Dan present a number of everyday illusions that can affect our ability to make decisions and remain self-aware. The inspiration for the book was an experiment that Chris and Dan conducted in the 1990s. If you have access to video, it would be a good idea to press pause now and watch the one minute experiments by clicking on the link provided in the written introduction.

When I spoke to Chris and Dan, I began by asking them to provide me with an overview of the experiment and its results.

Daniel Simons: This is Dan. The experiment that we did about twelve years ago was looking at the nature of focused attention, so what people can pay attention to and what they ignore and how effectively they can do that. And it was based on some much earlier work by our colleague Ulric Neisser from the 1970s.

So, here’s what people did. We brought them in and we had them watch a videotape of people passing basketballs and there were three people wearing white shirts and three people wearing black shirts and the task we gave them was to count how many times the players wearing white passed the ball and to ignore passes made by players wearing black. And about half way through the video, after about 30 seconds, we had a person wearing a full body gorilla suit walk into the middle of the scene, turn to face the camera, thump its chest and then walk off the other side a total of 9 seconds later and what we found was that 50% of people had no idea that a gorilla had walked through. And if you asked them afterwards, ‘Did you notice anything?’ they said no. ‘Did you notice the gorilla?’ ‘The what?!’ So they were shocked that they could have missed it. And the people who saw it were stunned that anybody could miss it. So that’s really what we are interested in here. Both the people missed something that’s fully obvious right there in front of them, fully visible, but also that their intuition about what they would notice and what they wouldn’t notice tends to be wrong. And that was kind of the starting off point for our book.

Female interviewer: So, what did that initial research start to tell you about what you call the illusion of attention?

Daniel Simons: The illusion of attention is this belief we have that if something important or visually distinctive or really salient is right in front of us and our eyes are directed at it, we will automatically see it, that important things will capture our attention. And the reality is that important things like that often don’t capture our attention, they don’t grab our awareness; instead, we often miss them. What’s interesting is this illusion that we notice everything, and we think that important things will grab us when they don’t. And that illusion has all sorts of interesting consequences, but it’s a natural result of our daily experiences. We are only aware of those times when something has grabbed our attention and awareness. We are not aware of those things we have missed. So we tend to mistakenly believe that we notice everything and that can have all sorts of practical consequences in our daily lives.

Female interviewer: Can you give us some examples of how the illusion of attention might affect us on a day-to-day basis?

Daniel Simons: The most straightforward example comes from driving. When you are driving on the road, you assume that as long as your eyes are looking at the road, if anything unexpected or important happens, you will automatically notice it and the result of that is that we do things like talk on the cell phone while we are driving because we feel that as long as we are looking at the road, if something important happens we will see it. The reality is that whether or not you are aware of something depends both on your eyes and also on your mind and if you are using up attention talking on a cell phone, you are that much less likely to notice something unexpected. So the idea that we can drive and talk on the cell phone at the same time is basically following from this illusion that we have that anything important will automatically capture us, we don’t have to use any of our mind to spot it.

Female interviewer: What does your research into the illusion of attention tell us about multi-tasking? Is it really possible for us to pay attention to several different things at the same time?

Christopher Chabris: This is Chris. The illusion of attention is in effect the reason why we think we are good at multi-tasking. When we are doing several things at once, we don’t really have the perception that we are doing them very badly, we have the perception that we are noticing the important things in our environment, we are responding to them, we are keeping up with several tasks at once, in part because we believe we are paying attention to more than we are and that we are noticing all the important things around us. So this leads us to do more multi-tasking than we probably should, especially in potentially dangerous situations like driving or in other safety related circumstances.

It’s true that there, our research shows that there are some people who might be able to do two tasks at once without doing either of them much worse than they would do them if they did them one after the other but if those kind of people exist, they are a tiny, tiny minority of all of us and we shouldn’t fool ourselves into thinking that because there are some people who can do that, I must be one of those people. In fact, most people do tasks worse when they do them at the same time, not better; they do them substantially worse than when they do them one after the other. And the illusion really makes us not realise that and realise how little attention we are actually paying to things when we are focused on something else.

Female interviewer: And is there any pattern in terms of who is good at multi-tasking? You say it’s a small minority, is it a certain type of person?

Daniel Simons: This is Dan. There is very little evidence that anybody can do multi-tasking at all. There is one paper that just came out a couple of months ago that provides the first evidence that anybody can and that was a study of a couple of hundred people that found that 2% were able to do multi-tasking better without much cost. But I think there is a need for some replication for that claim. It is an interesting claim, there is a possibility that maybe some people are particularly good at the tasks they had, they used in this study, but there is very little evidence that anybody can do it well and there is not that much evidence for individual differences, and nobody can do it particularly well. Almost everybody is going to be better off doing one thing at a time rather than trying to do two things that rely on the same resources. You can think of it this way, it is pretty much impossible to whistle, talk and chew gum at the same time because all three use your mouth and your tongue in different ways. Well, anything that requires you to use your mind in the same way or use the same parts of your brain for different sorts of tasks at the same time, is going to cause you problems and we often don’t realise those problems because when we are doing that we are distracted.

Female interviewer: So, the illusion of attention makes us believe that we will always notice everything around us, especially seemingly obvious information, when in reality we won’t. This leads us to falsely assume that we can competently perform several different tasks at the same time.

Moving onto the illusion of confidence, the Dunning-Kruger effect is a bias that affects people’s awareness of and their confidence in their true capabilities, particularly in areas in which they have little experience. With this in mind, what can we do to become more aware of our real abilities and recognise when we need to improve?

Christopher Chabris: This is Chris. The Dunning-Kruger effect is what happens when people evaluate their own abilities and it refers to the fact that the people who are often the least competent or the least skilled in some particular task or domain of knowledge, are the most over-confident about their own abilities. There is an interesting study in which people were measured on their sense of humour by testing how well their decisions about what jokes were funny matched up with expert comedians’ decisions. And it turns out that the people who have the worst senses of humour actually thought they were pretty funny and that demonstrates this unskilled and unaware of it effect, people who are the least skilled are unfortunately also unaware that their skill is that low.

When it comes to figuring out what you can do about it to be aware of it, the first thing is to know that this happens and to realise that your own self-perception of your own abilities and your own level of expertise and your own knowledge may not be accurate and in fact it is probably inflated. You probably think better of your abilities than you really should if you were being objective. And this can be especially apparent when you are just learning something new or you are just involved in a new position or learning a new skill, I mean, that’s the time when your actual performance ability is lowest because you are just getting started, you haven’t learned the job yet or you haven’t acquired the knowledge yet, I mean, that’s when you are most in danger of thinking too much of yourself. It turns out that when you get to the higher levels of performance, you are getting more accurate self-perception, you are less over-confident and the way to really erase this effect is to actually acquire skill and to learn and the more you learn, not only the better will you be at the task, but the better you will be at understanding how good you are at the task. It comes with additional skill and expertise, comes additional, comes better self- perception, self-awareness of your own skill levels.

Daniel Simons: This is Dan. Let me just come in really quickly. The real problem with the Dunning-Kruger effect is that those people who are the least competent and the least skilled confuse themselves for people who are actually the most skilled. So people think it is really easy at times to acquire a skill, to just kind of quickly have enough talent that you could do it as well as much more experienced people and any time you find yourself thinking that you are doing it as well as somebody with a lot more experience, that’s the time to be on the look out for over-confidence. If you have got a lot of skill, meaning you have spent a lot of time practising a task or you have got a lot of experience with it, your confidence is more likely to be more in line with your actual ability but when you are just starting something out, if you think you are as good as the people who have been doing it for a while, that’s when you are likely to be over-confident in your abilities.

Female interviewer: And are there any times where we might be likely to be under- confident and rate ourselves less favourably?

Christopher Chabris: This is Chris. In the humour study that Dunning and Kruger actually did, it turned out that the people with the best senses of humour, people who actually judge jokes to be funny in accordance with the way expert comedians did, actually under-rate their own sense of humour. They thought they were less funny than they really were. So, it’s possible that sometimes people at the high end of the ability scale or professional standing might be under-confident but that seems to happen much less often than the problem of people being over-confident and I wouldn’t want to advise people to be too much worried about under-confidence because over-confidence I think is a much bigger thing to be wary of.

Daniel Simons: This is Dan. Just one thing to add to that is there are big individual differences in confidence. Some people tend to be generally more confident than others and there are cases in which people can feel like they are less skilled than they actually are. This often happens in areas of, you know, tremendous expertise so a lot of academics when they are first starting out are convinced that they are fooling everybody, that they just don’t know as much as all the people around them and that’s a case where you are differentiating within levels of expertise and having a hard time judging how expert you actually are compared to the novices. So, there are some cases in which maybe people are less confident in their own abilities than they justifiably should be but it’s just because of the comparison group that they have. So, somebody who has got a lot of expertise but are around people who are a lot more expert might be under- confident in their abilities.

Female interviewer: You point out in the book that we generally interpret feedback about our abilities in the most positive light possible. With this in mind, what’s the best thing we can do to interpret feedback appropriately when we do receive it?

Daniel Simons: This is Dan. So that phenomenon of interpreting feedback in the way that is most favourable to us is actually known as the self- serving bias and it is something that we all do. We tend to attribute our own successes to our abilities, our failures to the situation and to do the reverse for other people. We tend to basically attribute people’s successes to the situation and their failures to their personalities.

And these are subtle biases, this isn’t an all or none sort of thing. It is not the case that you always interpret everything in the best possible light for yourself but it’s a bias that we tend to have. It explains why, you know, if you have a successful, let’s say you are a chess player and you beat somebody who is your own skill level, you might attribute that to your skill whereas if you make a blunder and lose to somebody of your own skill level, you will interpret that blunder to something that really wasn’t inherent to you as a person, you kind of attribute it, just some bone headed mistake that doesn’t really matter, it doesn’t reflect your true ability. We tend to discount those sorts of things that go wrong more readily than we discount those things that go right.

One way to think about that is just to kind of be aware of those cases in which you are assuming that your successes are due to your ability or that other people’s successes are due to their situation. It is just a slight bias to kind of give ourselves the benefit of the doubt much of the time and that in many ways is a positive thing. It is something that kind of boosts your self-esteem and your view of yourself. The danger comes in assuming that all of those successes are inherently due to your own skill and not due to occasionally luck or other factors that you didn’t have any control over, and in the same way interpreting failures, some of those might be due to your lack of skill and not just due to the situation. So being aware of this sort of self-serving bias allows you to kind of reassess whether what you are doing is actually due to your abilities or due to the situation.

Female interviewer: How can the illusion of confidence influence group performance?

Christopher Chabris: This is Chris. Confidence is a trait of individuals but it is also a trait that groups can express when groups have to make a decision or a judgement or an estimate, which happens a lot in situations in government and business, a group is called upon to decide something rather than one individual being given the power.

We did an experiment where we found that if you take some people who have sort of inherently low confidence to start with, people who tend not to be confident people, but put them together to work in a group to make judgements as a group, then they will become more confident. It is as though they’re each individually having little confidence in their opinions but when they aggregate their opinions together through discussion, they gain greater confidence in the outcome than they would as individuals. This doesn’t happen as much to people who go in with high confidence to start with. Their confidence is not really in play when they are in a group. The people who have low confidence, their confidence gets greater. So the result of that is that you might have a situation where you have a bunch of individuals who don’t really hold an opinion very strongly and could go either way on some decision but just because they are all a little bit leaning in one direction, then they put together their opinions into a very confident conclusion that none of them would really have subscribed to as an individual. And this can happen in the context of government officials who have to meet together to make a decision or a Board of Directors or any other kind of group that has to not only make a decision, but also express how confident they are in that decision.

Female interviewer: What might be a more effective way to approach group working then if we are all susceptible to this bias in groups? Is there anything that we can do about it?

Christopher Chabris: This is Chris again. There are some decision procedures that groups can follow that leave the group less susceptible to bias than others. For example, a bad decision procedure would be to have the most senior member of the group start out by giving his/her opinion and then have the more junior members give their opinions because what’s obviously going to happen is that the junior people will be heavily influenced by what the senior person says just because that person had more influence over their career and their promotions and you just might tend to think that they know what they are talking about and you should follow along with them. A better procedure would be to go in the opposite order and let the most junior person start so that person would get a chance to express his/her opinion and then move up until the most senior person finally tries to describe a consensus or comment on what everybody else said. This seems so obvious once I say it, but surprisingly there are many case studies where groups of important individuals in government organisations or even at Board of Directors level did it the opposite way. And when you do it the opposite way, when you start with the most senior person and everybody else just says yes, then you are going to have an impression of greater confidence, we were all in agreement on this decision, but in reality you weren’t always in agreement, you just followed along the most influential person who spoke first, and that is a bad recipe for effective group decisions.

Daniel Simons: This is Dan. Let me just add in, one other aspect of group decisions like that is that whoever speaks first tends to influence other people much more, they tend to be viewed more as the leader of a group.

So if you really want to get a true sense of what everybody in the group says, the best way to do it is not to say, state your opinions out loud when other people haven’t actually written down their opinions, but to actually have everybody write down their opinions independently and then look at what the consensus is and you might find that there is going to be more disagreement when that happens and more discussion of, you know, what the appropriate steps should be.

The problem comes that we are socially influenced by other people, so whoever speaks first, tends to be viewed as the most knowledgeable and as the leader of the group, regardless of whether they actually know more than other people in the group. Again, it comes back to people who are confident tend to be the ones who speak first and we tend to trust confidence as a good sign of knowledge or skill or accuracy.

Female interviewer: We have heard that the illusion of confidence can often lead us to feel more confident that we should do in our abilities. This can affect the decision-making capabilities of both individuals and groups. It also impacts on the way in which we interpret our successes, mistakes and the feedback we receive from others.

Moving on to the illusion of knowledge, you talk in the book about how this illusion can affect the way that we approach projects at work, how we estimate how long different pieces of work might take. Can you talk to us a bit more about that?

Daniel Simons: Sure, this is Dan. The illusion of knowledge is in many ways related to the topic we have just been discussing in that it has to do with how we assess our own understanding. So we tend to think that we understand things in a much deeper way than we actually do.

So, for example, if I ask you, ‘Do you know how a toilet works?’ You might say, well, you know you think about it for a moment and say, ‘Yes, of course I know how a toilet works.’ The reality is you probably don’t. So, if I just ask you a couple of simple diagnostic questions like ‘Why does the water leave the bowl, or what’s the tank for?’, you quickly realise that you don’t really know how a toilet works, you know how to work a toilet. And we mistake that surface familiarity for a deep understanding. That actually plays out in many contexts. So one of the more obvious cases is when you are doing long range planning for big projects, there’s a general rule that you should at least double the length of time you think something big will take. And it is interesting that we have to do that, and we still tend to underestimate it because we tend to think we can anticipate all of the possible consequences and all the possible things that could wrong and all of the possible delays in the project, but in reality we don’t. We tend to be more likely to assume that we know what is going to happen and we don’t actually know what’s going to happen most of the time.

Female interviewer: Is there anything else that we can do to stop falling into this trap when we approach a piece of work?

Christopher Chabris: This is Chris. There’s one solution that is adopted in planning projects in large scale projects sometimes which I think can be usefully applied in a wide range of situations and that’s to take what the psychologist, Danny Kahneman, called the outside view, which is to think about how your project would be viewed by other people. If other people were looking at your project, not as a project that they were going to do themselves, but looking at the question of how long will it take you to do your project, what estimate would they make? Would they think you could get it done as fast as you can? Would they think you can get it done as cheaply as you think you can? Would they think you can get it done with as few resources as you think you are going to need? Probably not.

You can even go one step further and you can try to analyse similar projects to your own that have been done by other people in the past. So, for example, in construction, if you have to build a building of a certain size and with certain materials and so on, you can look at other projects that have been conducted with those kinds of buildings and see how long they actually took. And looking at the projected times that it took other people to complete things is a much better way of figuring out how long you are going to take to complete it than thinking about your own process of thinking about how quickly you could do it.

It’s sort of paradoxical because we tend to think as individuals that we know the most about our own abilities and our own projects and our own time management and nobody else knows all the details of what I am trying to do, so how can they possibly estimate a project better than I can? But it turns out the illusion of knowledge is exactly what leads us to think that we do understand our projects better than we really do and we do understand ourselves better than we really do. Looking at it from an outside view, from other projects that other people have done or the opinions would have of your project, can lead to a much more objective forecast. Just for a simple example, when we signed our contracts to write The Invisible Gorilla, we had an idea how quickly we could write a book and our publisher had an idea how quickly we could write our book. Our publisher probably had a better idea how long it was going to take us to write this book because they deal with authors writing books like this all the time, whereas this was the first time we had written a book together and of course we are going to think we know what we are up against, but the publisher with all of their experience across hundreds or thousands of different projects, actually has a much better database to draw from in making those kinds of estimates.

Female interviewer: So, are we all naturally predisposed to thinking we are more capable than we actually are?

Daniel Simons: There’s a common thread running through many of the illusions that we discuss in the book which is that we tend to think of ourselves in slightly more positive light than we possibly should. So, we tend to think that our ability to focus attention and to take in all of what’s around us is better than it actually is. We tend to think we remember more and more accurately than we actually do. We tend to be somewhat more confident in our abilities and knowledge than we probably should be. All these are cases in which we have limitations on our mind that we are largely unaware of and the limitations have been there all along, that’s one of the important things here, these are not, you know, surprising limitations from a perspective of cognitive psychology, but what’s interesting is how little we are aware of them. And being aware of these sorts of limitations on our own mind allows us to be aware of the sorts of assumptions we make about our own minds and our own limitations and as a result of that you can actually compensate for some of these limitations.

So, the limitations have been there all along, the question is what we do about them, and remaining blissfully unaware of them, you know, might be a pleasant way to go through life, but it is actually going to make you less safe and less productive and less effective in what you do. So if you know, for example, that you don’t take in everything that’s around you all at once automatically, you can do things that will allow you to maximise your chances of noticing when something unexpected happens. You can, for example, lock up your cell phone in the trunk so that you aren’t tempted to talk on the phone while you are driving, which makes you that much better able to notice when that rare, unexpected event happens that can be catastrophic. Or, if you know that you can’t actually multi-task, you might figure out a way of not trying to do two things at once and making the tasks more sequential, doing one at a time, and being more productive as a result. So knowing about these limitations, knowing that confidence is also a personality variable so that just because somebody speaks with certainty, doesn’t mean that they are any more knowledgeable or accurate than somebody who doesn’t. Knowing those sorts of things allows you to take into account your own limits and other people’s limits in making good decisions.

Image Credit: Flickr user Saad Faruque (accessed 02 June 2014).