WEBVTT - From the Vault: The Illusion of Control, Part 2 0:00:06.080 --> 0:00:08.920 Hello, and welcome to Stuff to Blow Your Mind. My 0:00:09.039 --> 0:00:12.760 name is Joe McCormick. This week Rob and I are out, 0:00:12.800 --> 0:00:16.080 so in our Tuesday and Thursday slots, we are bringing 0:00:16.160 --> 0:00:18.360 you a couple of older episodes of Stuff to Blow 0:00:18.400 --> 0:00:21.680 Your Mind, episodes from the vault, continuing the series we 0:00:21.760 --> 0:00:25.279 started this past Saturday. Today's episode is part two of 0:00:25.320 --> 0:00:30.120 our series on the illusion of control. This originally aired 0:00:30.200 --> 0:00:32.560 on February tenth, twenty twenty four. 0:00:32.800 --> 0:00:41.640 We hope you enjoy. 0:00:37.600 --> 0:00:41.360 Welcome to Stuff to Blow Your Mind, a production of iHeartRadio. 0:00:47.320 --> 0:00:49.680 Hey, welcome to Stuff to Blow Your Mind. My name 0:00:49.720 --> 0:00:50.639 is Robert Lamb. 0:00:50.600 --> 0:00:53.640 And I am Joe McCormick, and today we are back 0:00:53.680 --> 0:00:57.720 to continue our series on the psychology concept known as 0:00:57.760 --> 0:01:03.080 the illusion of control. This is a cognitive illusion, or 0:01:03.120 --> 0:01:06.160 a common error in thinking and judgment, in which we 0:01:06.520 --> 0:01:11.080 overestimate the amount of control we have over outcomes in 0:01:11.120 --> 0:01:15.400 the world, even outcomes that are in no way determined 0:01:15.440 --> 0:01:19.039 by our actions. So if you haven't heard part one, 0:01:19.080 --> 0:01:20.959 you should probably go back and listen to that first. 0:01:21.040 --> 0:01:24.120 But for a brief recap, we talked about some examples 0:01:24.240 --> 0:01:27.720 last time of the illusion of control. One would be 0:01:27.800 --> 0:01:31.120 the belief that you can control your chances of winning 0:01:31.200 --> 0:01:34.400 at a slot machine based on you know, who presses 0:01:34.480 --> 0:01:37.160 the button and how, and actually you know, it's a 0:01:37.200 --> 0:01:38.480 purely random process. 0:01:39.040 --> 0:01:39.640 There's no like. 0:01:39.920 --> 0:01:42.280 You know, you can't be like better at working a 0:01:42.319 --> 0:01:45.760 slot machine. But other examples would include like the belief 0:01:45.760 --> 0:01:48.800 that you can improve your chances of hitting a desired 0:01:48.880 --> 0:01:52.559 number on a dice throw by concentrating before the throw. 0:01:52.800 --> 0:01:56.360 I do this, yeah, yeah, and yeah. I was going 0:01:56.400 --> 0:01:57.720 to save this for a listener mail, but I go 0:01:57.760 --> 0:02:01.000 ahead mention it now heard from a listener on Discord 0:02:01.440 --> 0:02:05.240 who pointed out, this is I believe it's passy cish. 0:02:05.680 --> 0:02:08.440 I'm not sure if I'm pronouncing that correctly, so you 0:02:08.560 --> 0:02:10.840 use your name. But anyway. They pointed out that in 0:02:10.880 --> 0:02:14.560 Dungeons and Dragons there's an additional element here that we 0:02:14.600 --> 0:02:17.840 didn't touch on, and that's the drama of rolling your dice, 0:02:17.960 --> 0:02:21.000 of rolling that d twenty, doing that saving throw. You 0:02:21.040 --> 0:02:23.600 may put some concentration into it, not because not as 0:02:23.680 --> 0:02:26.320 much because you're hoping to influence the role, but because 0:02:26.800 --> 0:02:30.639 this matters, This is an important role. Perhaps the life 0:02:31.000 --> 0:02:34.160 or death of your character may hinge on the outcome, 0:02:34.200 --> 0:02:36.160 and you're going to play it up a little bit. Yeah, 0:02:36.200 --> 0:02:40.160 it's a socially performative drum roll, but I think undeniably 0:02:40.200 --> 0:02:43.480 there's off also that sense of like, all right, NAT twenty, 0:02:43.560 --> 0:02:44.080 let's do it. 0:02:44.400 --> 0:02:45.200 I can do this. 0:02:46.080 --> 0:02:47.360 Yeah. 0:02:47.720 --> 0:02:50.079 Other examples would be like the belief that you can 0:02:50.160 --> 0:02:53.280 influence the outcome of a sporting event hundreds of miles 0:02:53.320 --> 0:02:56.600 away by wearing a lucky charm. We talked about the 0:02:56.760 --> 0:03:00.800 childhood belief that you can control gameplay on video game 0:03:00.919 --> 0:03:03.639 with like a controller that's not plugged in, or by 0:03:03.680 --> 0:03:06.680 moving the joystick on a on an arcade cabinet you 0:03:06.720 --> 0:03:10.120 haven't put any quarters in. Yeah, yeah, And we also 0:03:10.240 --> 0:03:13.320 ended up talking about an influential early paper on the 0:03:13.360 --> 0:03:17.239 illusion of control from nineteen seventy five called the Illusion 0:03:17.240 --> 0:03:20.440 of Control in the Journal of Personality and Social Psychology 0:03:20.480 --> 0:03:25.320 by the American psychologist Ellen J. Langer. And for a 0:03:25.440 --> 0:03:29.600 quick summary of this paper, it used experiments involving games 0:03:29.639 --> 0:03:34.480 of chance with superficial elements inserted from games of skill 0:03:35.240 --> 0:03:38.360 to see if people would behave consistent with a belief 0:03:38.440 --> 0:03:42.440 that they had impossible levels of control over chance outcomes, 0:03:42.960 --> 0:03:46.440 and this study found that yes. In its experiments people 0:03:46.520 --> 0:03:50.640 did behave in a way that was consistent with overestimating 0:03:50.680 --> 0:03:54.720 their level of control over chance determined outcomes. However, the 0:03:54.760 --> 0:03:58.680 thing about this paper was the experiments did use indirect 0:03:58.800 --> 0:04:03.360 methods of studying the phenomenon, so these results came with 0:04:03.400 --> 0:04:06.840 some limitations that I'll describe in just a minute. I 0:04:06.880 --> 0:04:09.360 wanted to learn some more about the history of how 0:04:09.400 --> 0:04:12.080 the illusion of control has been studied, because there have 0:04:12.160 --> 0:04:15.760 been tons of papers on this, tons of experiments, and 0:04:15.800 --> 0:04:18.560 I wanted to kind of general overview, so I turned 0:04:18.640 --> 0:04:22.200 to a very helpful book chapter by a psychologist named 0:04:22.240 --> 0:04:27.360 Suzanne C. Thompson. The chapter is called Illusions of Control 0:04:27.560 --> 0:04:31.680 and it appears in a book called Cognitive Illusions, edited 0:04:31.720 --> 0:04:35.920 by Rudiger F. Pohl, published by Psychology Press twenty sixteen, 0:04:36.720 --> 0:04:38.640 though the version I read seems to have been an 0:04:38.720 --> 0:04:43.200 updated edition because it included references to more recent studies, 0:04:43.240 --> 0:04:45.960 such as one paper from twenty twenty one. So in 0:04:46.000 --> 0:04:50.200 this overview, Thompson uses a broader definition of the illusion 0:04:50.200 --> 0:04:54.320 of control than Langer did. Langer's definition was specifically about 0:04:54.960 --> 0:05:01.080 seeking desired outcomes in chance determined events, to Thompson says, instead, 0:05:01.160 --> 0:05:06.360 quote illusions of control occur when individuals overestimate their personal 0:05:06.520 --> 0:05:11.280 influence over an outcome, So that that's a more general 0:05:11.320 --> 0:05:13.440 way of stating it. You know, maybe your influence could 0:05:13.440 --> 0:05:15.839 be good or bad. It could be in getting something 0:05:15.839 --> 0:05:17.400 you want or in something you don't want. 0:05:17.720 --> 0:05:20.800 Yeah, so this broader definition could apply to like various 0:05:20.839 --> 0:05:22.920 games that have some sort of random element that you 0:05:22.960 --> 0:05:25.520 truly can't control. For you may be really great at 0:05:25.560 --> 0:05:28.080 the game, but you have this added level of illusion 0:05:28.120 --> 0:05:30.120 of control that thinks that you can you can definitely 0:05:30.200 --> 0:05:33.039 navigate any random occurrence. And I guess you could also 0:05:33.080 --> 0:05:37.320 apply it even to interpersonal relationships, you know, thinking that 0:05:37.360 --> 0:05:39.880 you have more control over other people in your circle 0:05:39.920 --> 0:05:40.400 than you do. 0:05:40.880 --> 0:05:43.560 Right, because this idea would also apply to things where 0:05:43.600 --> 0:05:46.520 you do have some control, but you're imagining you have 0:05:46.760 --> 0:05:50.840 more control than you actually do. And Thompson says that 0:05:50.960 --> 0:05:53.719 since the origins of this research in the nineteen seventies, 0:05:53.760 --> 0:05:58.440 there have been basically three different ways of experimentally demonstrating 0:05:58.480 --> 0:06:03.200 that people experience or control. There's like three branches of 0:06:03.279 --> 0:06:07.080 experiments on this tree. So approach number one that she 0:06:07.240 --> 0:06:11.760 outlines is the main example. Here is the original research 0:06:11.800 --> 0:06:15.200 by Ellen Langer, which we already described in the last episode. 0:06:15.640 --> 0:06:20.960 This general strategy involves gauging people's guesses about their likelihood 0:06:21.000 --> 0:06:25.200 of success in chance games that have superficial elements of 0:06:25.240 --> 0:06:29.359 skill games introduced. So to emphasize again, this approach does 0:06:29.400 --> 0:06:34.920 not actually directly measure people's perceptions of control. Instead, these 0:06:34.960 --> 0:06:37.920 experiments would kind of infer it from their behavior in 0:06:37.960 --> 0:06:41.000 a game. So you see that people bet more money. 0:06:41.400 --> 0:06:45.400 That suggests they think they have more control over the outcome, 0:06:45.839 --> 0:06:49.159 But it's possible there's another factor operating there, so there's 0:06:49.279 --> 0:06:53.520 less certainty that you're testing for the variable you're actually 0:06:53.560 --> 0:06:58.280 looking for. And Thompson explains some other ways of doing 0:06:58.279 --> 0:07:02.359 these kind of tests apart from like Langer's original experimental design. 0:07:02.440 --> 0:07:06.159 One thing she talks about is a type of study 0:07:06.240 --> 0:07:11.320 that you could call observer participant discrepancies. So an example 0:07:11.360 --> 0:07:13.440 of this would be you get a test group, you know, 0:07:13.480 --> 0:07:16.400 maybe a classroom of students or whatever, and you split 0:07:16.440 --> 0:07:19.400 them up into pairs, and you give each pair of 0:07:19.520 --> 0:07:24.160 subjects a random number generating apparatus, maybe a die that's 0:07:24.160 --> 0:07:27.920 a simple one. So in each pair, there's one person 0:07:28.160 --> 0:07:30.920 who gets to roll the die and the other person 0:07:31.200 --> 0:07:34.800 records all the numbers that they roll, and participants do 0:07:34.880 --> 0:07:39.360 this like twenty times, and then across the whole test group, 0:07:39.400 --> 0:07:43.280 whichever pair in the group has the highest total sum 0:07:43.440 --> 0:07:48.040 of rolls wins a cash prize, and both subjects in 0:07:48.200 --> 0:07:52.080 each pair guess their likelihood of winning before the game. 0:07:52.880 --> 0:07:55.960 Thompson says that if you try to replicate this sort 0:07:55.960 --> 0:07:59.320 of experiment with students, you will usually find that subjects, 0:07:59.360 --> 0:08:02.280 on average rate their chance of winning a little bit 0:08:02.400 --> 0:08:05.560 higher if they're the one rolling the die than if 0:08:05.560 --> 0:08:08.640 they're the one recording the roles. Again, that should not 0:08:08.680 --> 0:08:12.040 make a difference. So even though we both rationally know 0:08:12.160 --> 0:08:15.000 that the outcome is random, it just feels a little 0:08:15.080 --> 0:08:19.760 luckier if I'm the one doing it. However, and I 0:08:19.760 --> 0:08:23.440 thought this was interesting. Thompson says that some research has 0:08:23.480 --> 0:08:26.800 found that this effect can be reduced or even neutralized 0:08:26.840 --> 0:08:31.320 completely by the context of the game, for example, if 0:08:31.320 --> 0:08:34.920 it takes place in a classroom that has previously discussed 0:08:34.960 --> 0:08:39.240 the correct way to estimate probability on games like this. 0:08:40.280 --> 0:08:43.960 And that was interesting to me because it made me 0:08:44.040 --> 0:08:48.959 think about how people overcome cognitive biases and cognitive illusions. 0:08:49.480 --> 0:08:54.160 You know, sometimes the unfortunate fact is that simply being 0:08:54.360 --> 0:08:58.160 aware of a cognitive illusion, like knowing that sometimes our 0:08:58.200 --> 0:09:02.160 brains have a certain kind of bias, is not sufficient 0:09:02.240 --> 0:09:05.520 to keep us from falling for that bias. So you 0:09:05.559 --> 0:09:08.760 can know about the tricks your brain plays, and you 0:09:08.800 --> 0:09:11.079 can fall for them anyway. It happens to all of us, 0:09:11.480 --> 0:09:12.560 but in a case. 0:09:12.400 --> 0:09:15.360 And this obviously applies to many other aspects of the 0:09:15.400 --> 0:09:19.680 human psyche as well. I mean awareness, self awareness is 0:09:19.800 --> 0:09:23.079 often the first step, but that doesn't mean you've completely 0:09:23.679 --> 0:09:27.040 defeated the illusion or illusion that you are having to 0:09:27.040 --> 0:09:27.520 deal with. 0:09:28.160 --> 0:09:31.240 Exactly. This is true for everybody, but in other cases, 0:09:31.559 --> 0:09:34.440 and it varies from case to case. So in some cases, 0:09:34.880 --> 0:09:39.160 research has shown that we can be successfully inoculated mentally 0:09:39.240 --> 0:09:42.680 from certain irrational tendencies by being made aware of them, 0:09:43.120 --> 0:09:45.120 and this seems to be one of those cases. You 0:09:45.200 --> 0:09:49.880 can sometimes neutralize illusions of control just by like having 0:09:49.960 --> 0:09:53.480 a context in which people have already been reminded about 0:09:53.480 --> 0:09:58.480 how probabilities work. And I think that's interesting because you 0:09:58.600 --> 0:10:03.720 might naturally assume that the variable in resistance to cognitive 0:10:03.720 --> 0:10:07.240 illusions like the illusion of control is the person you know, 0:10:07.360 --> 0:10:11.600 like permanent features of a person's personality, and you might 0:10:11.640 --> 0:10:14.480 be inclined to think like well, a more rational person 0:10:14.640 --> 0:10:17.559 is better able to overcome their biases and think clearly. 0:10:18.160 --> 0:10:20.200 But I don't know if that's always the case. I 0:10:20.240 --> 0:10:24.760 wonder if it's really more about setting and context. Maybe 0:10:24.800 --> 0:10:29.760 setting and context are equally, if not more, powerful predictors 0:10:30.360 --> 0:10:33.640 of how well people overcome cognitive illusions. In other words, 0:10:33.720 --> 0:10:37.280 does like currently being in the setting of a statistics 0:10:37.360 --> 0:10:42.240 class inoculate you against the illusion of control better than 0:10:42.400 --> 0:10:45.720 being a person who is generally aware of cognitive illusions. 0:10:46.000 --> 0:10:48.200 I don't know the answer for sure there, but it 0:10:48.240 --> 0:10:52.640 seems worth considering rather than just defaulting to the explanation 0:10:52.760 --> 0:10:58.640 of permanent internal personality based differences. But anyway, so to 0:10:58.679 --> 0:11:02.200 move on, That was approach number Experimental approach number two 0:11:03.360 --> 0:11:07.600 is different. In this type of experiment, you give subjects 0:11:07.679 --> 0:11:13.000 a laboratory task where researchers can program exactly how much 0:11:13.040 --> 0:11:16.600 control the subject actually has. And in many of these 0:11:16.640 --> 0:11:20.760 experiments the subject has zero control. Sometimes they have more control, 0:11:21.360 --> 0:11:24.160 and then you ask the subject how much control they 0:11:24.240 --> 0:11:27.559 think they had. So an experiment typical of this type 0:11:27.640 --> 0:11:30.360 is one that was done by Alloy and Abrahamson in 0:11:30.440 --> 0:11:35.280 nineteen seventy nine, in which subjects would be given a 0:11:35.320 --> 0:11:39.040 button to press and they're told to see if they 0:11:39.040 --> 0:11:42.200 can use that button to control whether or not a 0:11:42.320 --> 0:11:45.360 light comes on, and then they're asked to judge at 0:11:45.360 --> 0:11:48.560 the end what amount of control they think the button 0:11:48.600 --> 0:11:52.560 had over the light. In reality, the light had no 0:11:52.720 --> 0:11:55.640 relationship to whether the button was pressed or not. It 0:11:55.679 --> 0:11:58.360 was simply programmed to come on at some fixed percentage 0:11:58.360 --> 0:12:02.120 of the trials with each subject, and unsurprisingly, even though 0:12:02.200 --> 0:12:05.320 it had nothing to do with whether the button was 0:12:05.320 --> 0:12:08.760 pushed or not or when subjects broadly thought they had 0:12:08.800 --> 0:12:12.360 some amount of control, and experiments when the light came 0:12:12.360 --> 0:12:16.840 on more frequently but again unconnected to the button, caused 0:12:16.840 --> 0:12:20.880 people to believe that they had more control over the light. So, 0:12:21.120 --> 0:12:24.280 at least in some cases, it seems like success at 0:12:24.360 --> 0:12:28.560 getting a desired outcome makes people more likely to believe 0:12:28.640 --> 0:12:31.480 they have control over that outcome, whether or not they do. 0:12:33.360 --> 0:12:36.680 And while at the risk of over extrapolating from a 0:12:36.800 --> 0:12:40.600 very contained laboratory outcome, this does sort of suggest to 0:12:40.640 --> 0:12:43.560 me connections to behaviors in the world, Like you know, 0:12:43.600 --> 0:12:47.200 when somebody has very good fortune at a particular juncture, 0:12:47.320 --> 0:12:52.400 they're like, yep, that was all me. Later, Thompson describes 0:12:52.440 --> 0:12:55.079 another version of this kind of test. This one is 0:12:55.120 --> 0:12:58.160 called the computer screen on set task. And so in 0:12:58.200 --> 0:13:00.000 this test, like you sit in front of a computer 0:13:00.920 --> 0:13:03.319 and you're looking at a screen, and the screen will 0:13:03.400 --> 0:13:07.040 sequentially produce a series of forty images, and all of 0:13:07.080 --> 0:13:10.400 these images are either a green X or a red O. 0:13:11.040 --> 0:13:13.640 And with each new screen, you can choose to press 0:13:13.640 --> 0:13:16.360 a button or not press a button, and your goal 0:13:16.520 --> 0:13:19.319 is to make the green X appear as many times 0:13:19.360 --> 0:13:22.040 as possible. So people will be trying to figure out 0:13:22.080 --> 0:13:24.480 if there's some pattern like pressing the button or not, 0:13:25.559 --> 0:13:27.680 you know, pressing it or not in what sequence, et 0:13:27.760 --> 0:13:31.360 cetera that'll make the green exes appear. Actually, once again, 0:13:31.400 --> 0:13:34.400 the button has no relation whatsoever to whether the symbols 0:13:34.400 --> 0:13:37.760 appear on the screen. The button doesn't do anything. And 0:13:37.800 --> 0:13:41.679 you can vary what percentage of each symbol the subjects get. 0:13:41.960 --> 0:13:43.760 At the end of the test, you have them rate, 0:13:43.840 --> 0:13:45.920 on a scale of zero to one hundred how much 0:13:45.960 --> 0:13:49.200 control they think they had over what appeared on the screen. 0:13:49.640 --> 0:13:52.480 People who got the green X seventy five percent of 0:13:52.520 --> 0:13:55.520 their random screens believed that they had a lot of 0:13:55.559 --> 0:13:56.960 control over the display. 0:13:57.520 --> 0:13:59.640 This is also interesting to think of in terms of 0:13:59.679 --> 0:14:02.480 the sample we discussed in the last episode about as 0:14:02.480 --> 0:14:05.600 a child thinking you had control over a video game. Yeah, 0:14:07.760 --> 0:14:10.280 maybe this doesn't play out. I'd be interested to hear 0:14:10.320 --> 0:14:13.800 from folks much younger than me. But looking back on 0:14:13.800 --> 0:14:16.440 the video games that I was doing this on, like 0:14:16.480 --> 0:14:21.800 these were the old school arcade games, where it was 0:14:21.960 --> 0:14:25.040 maybe a little more directly comparable to just pressing a 0:14:25.040 --> 0:14:27.560 button and seeing a random O or an AX on 0:14:27.640 --> 0:14:29.520 the screen, Like, there is a lot more room to 0:14:29.680 --> 0:14:32.440 ask the question, am I controlling it? I have fifty 0:14:32.440 --> 0:14:33.640 percent chance I am. 0:14:33.960 --> 0:14:36.920 In a way, I'm almost nostalgic for that mindset, Like 0:14:36.920 --> 0:14:40.880 there's something kind of beautiful about the ambiguity of wondering 0:14:40.920 --> 0:14:43.720 if you're controlling what's happening on the screen. I feel 0:14:43.720 --> 0:14:46.240 like maybe I'm wrong about this. I feel like I 0:14:46.240 --> 0:14:48.640 wouldn't fall for that now, but I kind of wish 0:14:48.680 --> 0:14:52.560 I could, because it suggests a more I don't know, 0:14:52.760 --> 0:14:56.560 just kind of like totally radically opened state of mind 0:14:56.680 --> 0:14:59.160 in which anything is possible, a more magical way of 0:14:59.200 --> 0:15:00.240 relating to the world. 0:15:00.600 --> 0:15:03.120 It's a cheaper way to go about going to the arcade. 0:15:03.200 --> 0:15:05.360 You know, I wonder what they would think if there 0:15:05.400 --> 0:15:07.320 was an adult who regularly came into the arcade and 0:15:07.360 --> 0:15:09.560 they're like, oh, man, he never spends anything. He just 0:15:09.600 --> 0:15:11.520 stands at the machines and pretends to play. 0:15:12.040 --> 0:15:14.320 Yeah, just toggling the joystick at the demo. 0:15:14.560 --> 0:15:17.040 It's great. You gotta sell this guy's nachos or something. 0:15:18.720 --> 0:15:21.920 Okay, anyway, that's approach number two, these very tightly controlled 0:15:21.960 --> 0:15:26.040 laboratory experiments. Approach number three is different. Once again, you 0:15:26.120 --> 0:15:29.320 get people to report their judgments of control in real 0:15:29.400 --> 0:15:33.360 life scenarios. An example here is a study by McKenna 0:15:33.440 --> 0:15:36.240 in nineteen ninety three. Not that McKenna different, Chaya, I 0:15:36.240 --> 0:15:37.160 think this is Frank P. 0:15:37.400 --> 0:15:38.000 McKenna. 0:15:38.080 --> 0:15:43.040 Yeah. Ask participants to rate the likelihood that, compared to 0:15:43.200 --> 0:15:48.480 other drivers, they would experience an auto collision, and they 0:15:48.480 --> 0:15:51.800 were asked to judge this when imagining themself as the 0:15:51.920 --> 0:15:57.320 driver versus imagining themselves as the passenger. Perhaps unsurprisingly, most 0:15:57.320 --> 0:16:01.240 people thought that accidents would be relatively less likely if 0:16:01.280 --> 0:16:02.160 they were the driver. 0:16:02.760 --> 0:16:05.840 This absolutely matches up with my experience. You know, even 0:16:05.880 --> 0:16:07.680 when I'm in the car with a driver that I 0:16:07.840 --> 0:16:10.920 definitely trust and even know that they are a better 0:16:11.000 --> 0:16:14.760 driver than me. You know, maybe they have more experience 0:16:14.880 --> 0:16:18.000 or they've undergone training. They're still like that gut feeling 0:16:18.040 --> 0:16:21.440 of like I'm not actually in control. I can't hit 0:16:21.480 --> 0:16:24.440 the brake when I see the brake lights ahead getting closer, 0:16:24.880 --> 0:16:27.600 and therefore I feel like a little more anxious about 0:16:27.600 --> 0:16:31.600 the whole scenario oftentimes, like realizing that this is irrational, 0:16:32.040 --> 0:16:32.680 but feeling it. 0:16:32.720 --> 0:16:35.360 Nonetheless, I totally relate to that. I feel that too, 0:16:35.440 --> 0:16:37.880 the same thing. It's not like I actually think this 0:16:37.960 --> 0:16:40.400 other person is a more dangerous driver than me. I 0:16:40.560 --> 0:16:42.760 just it's just hard to get over that feeling. 0:16:43.360 --> 0:16:43.560 Yeah. 0:16:43.960 --> 0:16:48.200 In the second study in this mckinna paper, participants were 0:16:48.200 --> 0:16:51.920 asked about specific types of collisions, those that would seem 0:16:51.960 --> 0:16:55.120 to involve either more or less driver control. So they 0:16:55.120 --> 0:16:58.600 were talking about like rear ending someone versus being rear 0:16:58.720 --> 0:17:02.760 ended versus having a tire blowout, And the idea was 0:17:03.480 --> 0:17:07.440 rear ending someone is generally thought to be largely subject 0:17:07.480 --> 0:17:09.920 to driver control. Of course, we know that there are 0:17:09.920 --> 0:17:13.359 factors that other factors that can intervene breaks could fail whatever, 0:17:13.640 --> 0:17:16.159 whereas getting rear ended seems to be out of the 0:17:16.240 --> 0:17:20.760 driver's hands, and people were highly likely to say that 0:17:20.800 --> 0:17:24.520 they were less likely to have the type of collision 0:17:24.560 --> 0:17:27.960 in which the driver was in control. Has high control, 0:17:28.240 --> 0:17:30.879 So I am much less likely than other people to 0:17:30.960 --> 0:17:34.480 rear end someone. Whether it's me or someone else, makes 0:17:34.560 --> 0:17:39.280 less difference in getting rear ended. Quote. Thus, people show 0:17:39.359 --> 0:17:43.720 illusory control over avoiding an accident by assuming that they 0:17:43.840 --> 0:17:47.200 will be able to exert control that others cannot. 0:17:47.920 --> 0:17:49.440 And I guess this is what's in play when you 0:17:49.480 --> 0:17:54.240 see drivers, so many drivers just riding bumpers through terrifyingly 0:17:54.280 --> 0:17:56.840 fast traffic all the time, like they just maybe they 0:17:56.880 --> 0:18:01.040 have just heightened control over things. I would tend to 0:18:01.040 --> 0:18:01.399 doubt it. 0:18:01.760 --> 0:18:03.959 Yeah, that would be dangerous if someone else did it, 0:18:04.000 --> 0:18:05.640 But I can handle it. 0:18:05.880 --> 0:18:08.880 I alone can weave in and out of traffic and 0:18:09.000 --> 0:18:12.720 make it to my destination two minutes ahead of schedule. 0:18:13.000 --> 0:18:15.679 So this type of experiment is taken to show that 0:18:15.720 --> 0:18:19.120 people have an illusion of control when they consider themselves 0:18:19.200 --> 0:18:22.879 relative to other people. A driver has some degree of 0:18:22.920 --> 0:18:26.120 control over whether they end up in a collision. On average, 0:18:26.119 --> 0:18:28.800 people think that they are better able to avoid that 0:18:28.880 --> 0:18:31.960 outcome than other people are, and so looking back over 0:18:32.000 --> 0:18:34.800 these three methodologies, Thompson says, you know, each of them 0:18:34.840 --> 0:18:37.760 have strengths and weaknesses. So approach number one kind of 0:18:37.800 --> 0:18:41.480 the Langer approach. The pros are that it uses realistic 0:18:41.520 --> 0:18:45.160 situations that people engage in every day, like lottery drawings 0:18:45.200 --> 0:18:49.280 and games and stuff. And also it has the pro 0:18:49.440 --> 0:18:53.360 that the indirect measure can help detect an illusory belief 0:18:54.000 --> 0:18:58.120 in control that does in reality guide behavior, but which 0:18:58.160 --> 0:19:02.000 people might resist admits if they were asked directly, and 0:19:02.040 --> 0:19:05.000 that does seem big to me. It helps avoid like 0:19:05.440 --> 0:19:10.159 people tailoring their answers to avoid embarrassment. Cons on the 0:19:10.200 --> 0:19:14.280 other hand, are it's indirect, so it doesn't test whether 0:19:14.359 --> 0:19:17.439 control is really the deciding factor. You kind of have 0:19:17.520 --> 0:19:21.040 to infer that and wonder if other factors could be 0:19:21.040 --> 0:19:25.159 contributing as well. Approach number two the laboratory experiments like 0:19:25.200 --> 0:19:27.680 Alloy and Abramson with like you know, the light coming 0:19:27.680 --> 0:19:30.800 on or the greenexes and red o's on the computer screen. 0:19:31.200 --> 0:19:34.919 The pros of that are that the dependent variable is 0:19:35.040 --> 0:19:39.040 definitely judgment of control, like it's a very tightly controlled experiment. 0:19:39.680 --> 0:19:42.399 Cons would be that these tasks do not have what 0:19:42.440 --> 0:19:47.800 psychologists call external validity, So they're like weird tasks with 0:19:47.920 --> 0:19:50.919 no close analogy in our day to day lives, so 0:19:51.200 --> 0:19:53.680 they might not be telling us how people would actually 0:19:53.720 --> 0:19:56.719 behave in reality. They might just be like producing a 0:19:56.760 --> 0:19:59.639 weird kind of behavior that's specific to the lab task. 0:20:01.400 --> 0:20:04.960 Approach Number three the self reporting of control judgments about 0:20:04.960 --> 0:20:09.959 everyday activities like driving. Ala McKenna pros this does have 0:20:10.040 --> 0:20:15.000 external validity cons are it relies on reflective self reporting, 0:20:15.040 --> 0:20:17.639 which can be subject to all kinds of biases you 0:20:17.680 --> 0:20:19.480 know when you're trying to when you ask people to 0:20:19.640 --> 0:20:24.040 self report on their own judgments about their lives. However, 0:20:24.119 --> 0:20:27.120 Thompson says that a strength of illusion of control research 0:20:27.240 --> 0:20:30.639 is that even though these methodologies all have their strengths 0:20:30.680 --> 0:20:35.280 and weaknesses, they mostly point to a similar conclusion, which 0:20:35.320 --> 0:20:37.800 is the fact that on average, people believe we have 0:20:37.920 --> 0:20:42.440 more control over outcomes than we actually do. And there 0:20:42.440 --> 0:20:45.600 do seem to be some doubts about in exactly what 0:20:45.680 --> 0:20:49.840 scenarios this applies and what causes it, but the core 0:20:49.920 --> 0:20:52.800 finding seems fairly robust. Though I'm going to talk about 0:20:52.840 --> 0:20:56.600 one paper later in this episode that has some theoretical 0:20:56.600 --> 0:21:01.199 criticisms of how this research and how the experimental findings 0:21:01.200 --> 0:21:14.160 are framed. So it seems there probably is an illusion 0:21:14.200 --> 0:21:17.280 of control, especially for outcomes that we have very little 0:21:17.320 --> 0:21:20.720 control over. But it would be very surprising if people 0:21:20.760 --> 0:21:24.880 showed an illusory belief in control over all variables in 0:21:24.920 --> 0:21:28.200 all situations equally. So there has to be some more 0:21:28.240 --> 0:21:32.520 granular research on like when illusions of control happen, Like 0:21:32.880 --> 0:21:35.760 what are the kinds of things that we think we 0:21:35.840 --> 0:21:39.360 have more control of than others, more illusory control over 0:21:39.520 --> 0:21:43.200 than others, And what kind of situations or states can 0:21:43.240 --> 0:21:46.920 we be in that heighten this illusion? And to continue 0:21:46.920 --> 0:21:51.280 with Thompson's overview, Thompson highlights seven variables that have been 0:21:51.320 --> 0:21:54.879 studied and found to affect the illusion of control. This 0:21:54.960 --> 0:21:57.360 list does not mean that these are the only factors 0:21:57.400 --> 0:21:59.720 influencing it. It's just that these have been studied well 0:21:59.800 --> 0:22:02.919 enough to discuss in this book chapter. So the first 0:22:02.920 --> 0:22:07.240 one is skill related factors. Now, this was a major 0:22:07.280 --> 0:22:10.480 part of the original nineteen seventy five paper by Langer. 0:22:11.240 --> 0:22:14.280 A lot of studies have found that if a situation 0:22:14.480 --> 0:22:19.440 has features we associate with with dependence on skill, we're 0:22:19.480 --> 0:22:23.320 more likely to experience an illusion of control and Examples 0:22:23.400 --> 0:22:29.280 of these features could include quote, familiarity, making choices, active 0:22:29.359 --> 0:22:34.439 engagement with the material, competition, and four knowledge. So we 0:22:34.480 --> 0:22:39.080 talked about several of these in the previous episode. For example, familiarity, 0:22:39.160 --> 0:22:41.720 you know you might be more inclined to think you 0:22:41.880 --> 0:22:45.080 have control over the outcome of a chance game if 0:22:45.119 --> 0:22:47.720 you are familiar with the game, or if there are 0:22:47.800 --> 0:22:50.800 elements of the game that are familiar to you. And 0:22:50.880 --> 0:22:53.479 this is generally true of skill based games, but wouldn't 0:22:53.480 --> 0:22:57.520 affect chance based games. One of these variables though, Actually 0:22:57.520 --> 0:22:59.600 there was a twenty twenty one paper that casts some 0:22:59.720 --> 0:23:03.800 doubt over whether it affects illusions of control, and that 0:23:03.920 --> 0:23:08.080 variable is choice. So the original idea is that if 0:23:08.080 --> 0:23:11.440 you have a choice to make that gives you illusions 0:23:11.440 --> 0:23:14.560 of control. An example would be a lottery type game. 0:23:14.640 --> 0:23:17.439 So imagine a game where you buy a lottery ticket. 0:23:17.720 --> 0:23:20.199 The ticket has a random series of numbers on it, 0:23:20.400 --> 0:23:23.040 and you win a prize if the winning number matches 0:23:23.080 --> 0:23:26.199 your ticket. Now consider the same game, except you get 0:23:26.280 --> 0:23:29.439