WEBVTT - Ep69 "Why do you see something everywhere after you've seen it once?" 0:00:04.600 --> 0:00:07.880 Why am I going to start today's brain science episode 0:00:07.920 --> 0:00:12.360 by telling you about the beater Mainhoff group, which was 0:00:12.480 --> 0:00:16.920 a West German far left terrorist group in the nineteen seventies. 0:00:17.680 --> 0:00:22.480 This political group is fully irrelevant to any modern discussions 0:00:22.560 --> 0:00:27.280 and their political positions are totally irrelevant to neuroscience. So 0:00:27.320 --> 0:00:31.160 what do these domestic terrorists have to do with attention 0:00:31.880 --> 0:00:35.600 and salience and the front of your brain and synchronicity? 0:00:36.120 --> 0:00:40.520 And why might you soon hear about the better Minehoff group? 0:00:40.640 --> 0:00:44.440 Again not for political reasons, but for reasons to do 0:00:44.720 --> 0:00:51.000 with your own brain. Welcome to Inner Cosmos with me 0:00:51.159 --> 0:00:54.960 David Eagleman. I'm a neuroscientist and author at Stanford and 0:00:55.040 --> 0:00:58.320 in these episodes we sail deeply into our three pound 0:00:58.440 --> 0:01:02.480 universe to understand the intersection between how the brain works 0:01:02.560 --> 0:01:14.800 and how we experience life. So let's start in nineteen 0:01:15.000 --> 0:01:19.200 seventy when a far left wing militant group called beater 0:01:19.319 --> 0:01:23.080 Mainhoff also known as the Red Army Faction, they took 0:01:23.160 --> 0:01:27.040 to the streets in Germany. Now they define themselves as 0:01:27.200 --> 0:01:31.160 communist and anti imperialist, and they launched a string of 0:01:31.840 --> 0:01:37.440 robberies and then bombings, and eventually assassinations, and this all 0:01:37.920 --> 0:01:41.480 brought a lot of bloodshed and pain to Germany throughout 0:01:41.600 --> 0:01:46.520 the decade. Okay, now, who cares about the beater Minhoff Gang. Well, 0:01:46.640 --> 0:01:51.080 unless this touched your life in the nineteen seventies, probably nobody. 0:01:51.160 --> 0:01:53.520 So why am I telling you about this gang? Are 0:01:53.520 --> 0:01:58.240 they coming back? No, happily they are long defunct. But 0:01:58.360 --> 0:02:01.280 the joy of today's podcast is that I get to 0:02:01.320 --> 0:02:06.240 make a prediction that you will hear about Bader Mainehoff 0:02:06.400 --> 0:02:09.320 again and it probably won't be too long. And then 0:02:09.360 --> 0:02:13.440 you'll think, Wow, I just heard about that on Eagelman's podcast, 0:02:13.480 --> 0:02:16.240 and now I'm hearing about it again. Now, how can 0:02:16.280 --> 0:02:19.760 I make a prediction like that? Well, the important part 0:02:19.760 --> 0:02:24.399 of our story began two decades after the beater Minehoff Kang. 0:02:24.600 --> 0:02:27.760 It began with a man named Terry Mullen in nineteen 0:02:27.840 --> 0:02:30.960 ninety four. Now, Terry, like probably most people, had never 0:02:31.000 --> 0:02:34.520 heard of the batter Minehoff Gang, and then he heard 0:02:34.520 --> 0:02:38.320 about them. He heard their name, and then he was 0:02:38.400 --> 0:02:42.800 surprised to hear the name a second time within twenty 0:02:42.800 --> 0:02:45.560 four hours, and he thought that was just sort of 0:02:45.600 --> 0:02:48.160 weird and amazing that he had just learned that name 0:02:48.520 --> 0:02:51.040 and then he heard it a second time, so he 0:02:51.240 --> 0:02:55.600 informally posted about that on the Saint Paul Pioneer Press 0:02:55.720 --> 0:02:59.720 online discussion board, and others immediately chimed in that they 0:02:59.760 --> 0:03:03.720 had noticed this general pattern before. There had been times 0:03:04.240 --> 0:03:08.000 when they had learned a new concept or name, and 0:03:08.040 --> 0:03:11.800 then they began to see it everywhere, and this phenomenon 0:03:11.960 --> 0:03:15.519 of learning something new and then seeing it again shortly after, 0:03:15.720 --> 0:03:19.240 or seeing it all over the place. Everyone started chiming 0:03:19.280 --> 0:03:22.240 in on the discussion board that they had noticed this 0:03:22.360 --> 0:03:25.920 kind of thing before. So the original poster named this 0:03:26.200 --> 0:03:31.200 the Beaiter Meinhoff phenomenon, and it quickly blossomed into a meme, 0:03:31.680 --> 0:03:36.520 first on the discussion board, then nationally, then internationally. Now 0:03:36.800 --> 0:03:39.720 you've probably noticed this in your own life. You see 0:03:39.720 --> 0:03:42.600 some item or some new term, and then you see 0:03:42.640 --> 0:03:46.000 it again and again. It's not necessarily something really popular 0:03:46.080 --> 0:03:51.320 that everyone's talking about, but seemingly something more obscure. Once 0:03:51.360 --> 0:03:54.240 you've heard of something, you start noticing it everywhere, like 0:03:54.280 --> 0:03:57.720 you've just learned a new word when you're totally sure 0:03:57.800 --> 0:04:01.240 you've never heard before, and then seemingly out of nowhere, 0:04:01.280 --> 0:04:03.800 you hear it several times in a week where maybe 0:04:03.800 --> 0:04:08.040 you learn about some obscure piece of technology and suddenly 0:04:08.320 --> 0:04:12.520 it appears elsewhere on your news feed, or your friend 0:04:12.720 --> 0:04:15.760 mentions it out of the blue. If you chew on 0:04:15.840 --> 0:04:17.720 this for a second, you'll definitely be able to come 0:04:17.800 --> 0:04:21.520 up with examples where you learned some new piece of trivia, 0:04:22.000 --> 0:04:24.680 or you became interested in some new type of shoe 0:04:24.720 --> 0:04:28.080 and then suddenly it seems like you encounter that particular 0:04:28.160 --> 0:04:31.440 fact again, or you see that shoe everywhere you go. 0:04:31.920 --> 0:04:36.360 That is the beatter Minehoff phenomenon. A friend recently reported 0:04:36.360 --> 0:04:38.839 this to me. When his wife got pregnant, he started 0:04:38.920 --> 0:04:43.320 noticing pregnant women everywhere. Suddenly the world was filled with 0:04:43.480 --> 0:04:46.760 women about to give birth. Where were they all before? 0:04:46.960 --> 0:04:49.599 And the better Minehoff phenomenon happens to me a lot. 0:04:49.720 --> 0:04:52.560 It happens that I'm thinking about getting a new vehicle now, 0:04:52.880 --> 0:04:56.039 so I've been paying more attention to car models on 0:04:56.080 --> 0:04:59.000 the road around me, and I've spotted a couple of 0:04:59.040 --> 0:05:01.359 makes and models where I thought, hey, I kind of 0:05:01.400 --> 0:05:04.440 like that I've never noticed that vehicle before, and now 0:05:04.880 --> 0:05:08.119 I'm seeing it everywhere. It makes me feel that these 0:05:08.320 --> 0:05:12.320 models that I had never seen before, that they're proliferating, 0:05:12.360 --> 0:05:15.000 they're driving past me in the street, they're parked in 0:05:15.040 --> 0:05:18.560 my neighborhood, they're featured in a movie I'm watching. The 0:05:18.600 --> 0:05:23.760 impression is that the universe is echoing what I just learned. Now, 0:05:24.080 --> 0:05:27.680 as you've probably guessed because this is a neuroscience podcast, 0:05:28.160 --> 0:05:33.440 this phenomenon isn't about the universe making things appear more frequently. Instead, 0:05:34.160 --> 0:05:37.720 it's a quirk of your psychology. You are having the 0:05:37.960 --> 0:05:43.040 illusion that the frequency has increased to this end. In 0:05:43.200 --> 0:05:47.080 two thousand and five, the Stanford linguist ARNOLDSWICKI gave a 0:05:47.120 --> 0:05:51.040 more official name to the beter Mainhoff phenomenon. He coined 0:05:51.040 --> 0:05:55.720 this the frequency illusion, and that name captures the idea 0:05:55.880 --> 0:05:59.520 more precisely. Even though this remains more widely known by 0:05:59.560 --> 0:06:03.159 its origine meme name. I've also heard this referred to 0:06:03.279 --> 0:06:06.520 by the way as the red car illusion or the 0:06:06.880 --> 0:06:09.880 blue car effect, meaning that when you start paying attention 0:06:09.960 --> 0:06:12.080 to red or blue cars you see them everywhere. But 0:06:12.839 --> 0:06:17.240 whatever we call it, it's all about the illusion of 0:06:17.360 --> 0:06:22.000 the frequency of a previously unknown thing increasing, and you 0:06:22.040 --> 0:06:25.359 can see this in many guyses. For example, I've noticed 0:06:25.400 --> 0:06:31.240 a fascinating conflation between online ads and the beta Meinhoff effect. 0:06:31.279 --> 0:06:34.560 It goes like this. A friend is surfing around and 0:06:34.640 --> 0:06:37.800 spends a moment admiring a new band or a new 0:06:37.920 --> 0:06:40.840 kitchen item, and then she's surfing around the next day 0:06:41.240 --> 0:06:43.840 and sees more of these ads for the same thing. 0:06:44.000 --> 0:06:48.960 So she blames cookies and invasive adwere and yes, of 0:06:48.960 --> 0:06:51.440 course this is part of why she's seeing that thing more. 0:06:52.080 --> 0:06:56.120 But this model of big brother can quickly get stretched 0:06:56.160 --> 0:07:00.200 beyond sensibility when the new item is next noticed on 0:07:00.400 --> 0:07:05.359 a billboard or overheard in a conversation. People will sometimes 0:07:05.360 --> 0:07:08.839 make crazy arguments that there's a more devious ad network 0:07:08.880 --> 0:07:12.560 at play here, But this is simply the Beater Minehoff 0:07:12.600 --> 0:07:17.240 phenomenon in action. Now, just before we move into understanding 0:07:17.320 --> 0:07:20.840 why this happens, I'll tell you a quick academic joke. 0:07:21.080 --> 0:07:23.960 A student says to her friend, have you ever heard 0:07:24.040 --> 0:07:27.760 of the beatter Minehoff phenomenon? And the other replies, that's 0:07:27.800 --> 0:07:31.680 so strange. I just read about that yesterday. Okay, So 0:07:31.720 --> 0:07:34.720 we have a sense of what the phenomenon is, and 0:07:34.760 --> 0:07:37.760 the question we're going to ask today is why does 0:07:37.800 --> 0:07:42.640 it occur? Because what we notice and when tells a 0:07:42.680 --> 0:07:46.920 deeper story about how we engage with the world around us. 0:07:47.560 --> 0:07:51.520 So with the beta Minehoff phenomenon, we're always tempted to say, no, 0:07:51.720 --> 0:07:54.760 I'm definitely sure that I've never seen that before. It's 0:07:54.800 --> 0:07:58.600 not that I'm having some perceptual illusion. It's that I 0:07:58.640 --> 0:08:01.840 really never heard that term in my life before yesterday, 0:08:01.880 --> 0:08:05.880 and now I've heard it twice. But that's generally not 0:08:06.160 --> 0:08:09.520 the case. It's simply that we don't notice those things 0:08:10.000 --> 0:08:13.560 that are not relevant to us. The beta Mindhoff phenomenon 0:08:13.760 --> 0:08:16.320 is not some trick of the universe. Instead, it is 0:08:16.400 --> 0:08:21.880 the love child of two psychological processes running under the hood. 0:08:22.120 --> 0:08:27.320 The first issue is selective attention. Now, people sometimes think 0:08:27.360 --> 0:08:30.600 that selective attention just means you're not listening to your 0:08:30.640 --> 0:08:34.240 spouse or kids, but in fact, selective attention is our 0:08:34.400 --> 0:08:39.080 brain's way of managing the enormous amount of information that 0:08:39.200 --> 0:08:43.640 it encounters. Remember that your brain lives in silence and darkness, 0:08:43.679 --> 0:08:45.520 and it's trying to make a model of the whole 0:08:45.559 --> 0:08:49.120 world out there. But the whole world is way too 0:08:49.120 --> 0:08:53.760 complex and it's overflowing with detail, almost all of which 0:08:53.880 --> 0:08:57.800 is meaningless to you. So your brain is selective about 0:08:57.880 --> 0:09:00.640 what it looks for and what it takes in. So 0:09:00.679 --> 0:09:04.760 think about the last time you walked through a crowded marketplace. 0:09:04.880 --> 0:09:09.960 The sites, the sounds, the smell. It's more details than 0:09:10.000 --> 0:09:13.240 you could ever encode or whatever want to encode. So 0:09:13.960 --> 0:09:18.360 your attention acts like a filter. Now, let's get straight 0:09:18.400 --> 0:09:21.920 what we mean by a filter. Sometimes you might think 0:09:22.160 --> 0:09:24.360 that what this means is that you're taking everything in 0:09:24.520 --> 0:09:27.079 and then you're throwing out the stuff you don't want. 0:09:27.760 --> 0:09:31.000 But filtering is generally harsher than that. You don't even 0:09:31.080 --> 0:09:35.200 take stuff in unless it matches a shape for you. 0:09:35.280 --> 0:09:38.400 So think about those children's toys where you have a 0:09:38.440 --> 0:09:41.360 box and you put in a square object into the 0:09:41.400 --> 0:09:44.720 square hole, or a triangular object into the triangle hole, 0:09:44.800 --> 0:09:48.800 or a circular object. Those are filters. And if I 0:09:48.800 --> 0:09:52.280 hand you a hexagon, it doesn't even go in. It's 0:09:52.280 --> 0:09:55.560 not that the box considers hexagons and then spits it 0:09:55.640 --> 0:09:58.920 back out, it's that it doesn't even make it into 0:09:58.960 --> 0:10:01.719 the box. Now, from the brain's point of view, how 0:10:01.760 --> 0:10:05.480 does it know what things to filter in and out? Well, 0:10:05.920 --> 0:10:09.080 you pay attention to certain things because you have networks 0:10:09.080 --> 0:10:13.439 housed primarily in the prefrontal cortex for deciding what information 0:10:13.840 --> 0:10:18.600 is relevant or salient, and this is constantly updating from 0:10:18.720 --> 0:10:22.760 your experience. In other words, your brain is always asking, hey, 0:10:23.120 --> 0:10:25.920 what is the important stuff that I need to pay 0:10:25.960 --> 0:10:30.240 attention to? What stuff matters here? So when you encounter 0:10:30.400 --> 0:10:34.440 new information, like hearing that cool new band or seeing 0:10:34.440 --> 0:10:39.360 the cool new shoes, this network updates and subsequently your 0:10:39.400 --> 0:11:00.240 brain can take in this information. Now, let's flip this 0:11:00.360 --> 0:11:03.920 notion of selective attention and look at what can happen 0:11:04.040 --> 0:11:07.120 if you're simply not paying attention to something. This is 0:11:07.200 --> 0:11:12.600 called inattentional blindness, and it says though you never even 0:11:12.800 --> 0:11:15.840 see the thing at all. So I'll give you an example. 0:11:15.920 --> 0:11:19.640 I did an episode some weeks ago about why magic 0:11:19.679 --> 0:11:23.400 tricks work, and a lot pivots on exactly this issue 0:11:23.440 --> 0:11:27.959 of inattentional blindness. Something can happen right in front of you, 0:11:28.559 --> 0:11:30.959 but if you're not watching for it. In other words, 0:11:31.000 --> 0:11:35.600 your attention is elsewhere. The photons hit your eyes, but 0:11:35.679 --> 0:11:39.680 you don't see it. The magician moves his hand in 0:11:39.720 --> 0:11:42.680 an arc, or pretends he's throwing something in the air, 0:11:42.880 --> 0:11:45.640 or there's a snap of the fingers right over here, 0:11:46.000 --> 0:11:51.000 and suddenly the card slip happens. Right there, but it 0:11:51.080 --> 0:11:55.439 says though it didn't happen, why because your attention has 0:11:55.520 --> 0:11:59.040 been pulled elsewhere. Now, as an example of this that 0:11:59.080 --> 0:12:04.800 I'm guessing most people have seen, consider the Invisible Gorilla video. 0:12:05.320 --> 0:12:08.319 If you haven't seen this, search for it on YouTube. 0:12:08.480 --> 0:12:12.080 But pause this podcast here because what's coming next is 0:12:12.080 --> 0:12:15.840 a spoiler. The way this works is you see several 0:12:15.880 --> 0:12:18.720 people and half are wearing one shirt color and the 0:12:18.760 --> 0:12:21.920 other half is wearing a different color. And each team 0:12:22.320 --> 0:12:25.800 passes a basketball, but only two other members of their team, 0:12:26.120 --> 0:12:30.160 but they're all intermixed, so they're passing and moving between 0:12:30.200 --> 0:12:33.520 one another, and your job is to count the number 0:12:33.559 --> 0:12:37.400 of times the ball is passed by one of the teams. 0:12:37.840 --> 0:12:41.439 So you count, Okay, there's one pass, Then that person 0:12:41.520 --> 0:12:44.080 throws to another person on their team. Okay, that's two. 0:12:44.679 --> 0:12:47.400 Then that person throws to another team member. Okay, that's three. 0:12:48.120 --> 0:12:52.040 Like that, but it's challenging because to keep an accurate account, 0:12:52.120 --> 0:12:55.040 you have to ignore what the other team is doing. 0:12:55.640 --> 0:12:57.200 So you get to the end of the video and 0:12:57.240 --> 0:13:01.600 you say, great, I got it. This team made twelve passes, 0:13:02.240 --> 0:13:04.160 and you're told the answer, And let's say the answer 0:13:04.280 --> 0:13:06.760 is twelve you got it right. But then you're asked, 0:13:07.480 --> 0:13:11.520 but did you notice the gorilla? And then the video 0:13:11.640 --> 0:13:14.439 is shown again, the exact same video, and now you're 0:13:14.480 --> 0:13:17.000 not counting passes. You're just looking at the big picture. 0:13:17.320 --> 0:13:20.640 You're watching the video with a different context. And now 0:13:20.679 --> 0:13:22.800 you see something that blows your mind, which is that 0:13:22.960 --> 0:13:26.960 a guy in a gorilla suit walks right into the 0:13:26.960 --> 0:13:30.800 middle of the basketball players and he turns and faces 0:13:30.840 --> 0:13:34.560 the camera and thumps his chest, and then he struts 0:13:34.559 --> 0:13:37.880 out stage left, and all the while the basketball players 0:13:37.880 --> 0:13:41.079 are just continuing to pass the balls between them. Now, 0:13:41.160 --> 0:13:44.719 almost everybody doesn't notice the gorilla the first time they 0:13:44.760 --> 0:13:48.520 watched the video. Why It's because their attention is so 0:13:48.760 --> 0:13:53.439 focused on counting the passes. So given that our visual 0:13:53.440 --> 0:13:57.360 attention is limited, we miss the things that were not 0:13:57.600 --> 0:14:00.600 specifically focused on. And this is why it's known as 0:14:00.800 --> 0:14:06.200 inattentional blindness. Now, this can all be seen as a downside, 0:14:06.200 --> 0:14:08.920 but it's actually just what a brain needs to do 0:14:09.400 --> 0:14:13.400 to navigate its complex environment. It focuses on certain signals 0:14:13.760 --> 0:14:17.640 while ignoring others. For example, think about when you're at 0:14:17.679 --> 0:14:22.400 a noisy cafe. You have to selectively focus on your 0:14:22.560 --> 0:14:27.360 companion's voice and not hear all the irrelevant conversations around you, 0:14:27.440 --> 0:14:30.880 even though all of this hits your ear drums like. 0:14:30.960 --> 0:14:44.680 This, But what you attend to is this, Hi, can 0:14:44.720 --> 0:14:45.480 I get you something? 0:14:46.680 --> 0:14:50.560 So it's obviously very useful to be able to selectively 0:14:50.680 --> 0:14:55.760 attend because the world is a cacophonous flood, and selective 0:14:55.800 --> 0:15:00.360 attention allows us to zoom in on the signals of interest. 0:15:00.600 --> 0:15:03.080 I'm just gonna take a one second tangent to give 0:15:03.120 --> 0:15:05.560 you a sense of something here. Some of you know 0:15:05.640 --> 0:15:07.760 that I spun a company out of my lab called 0:15:07.880 --> 0:15:10.840 Neo Sensory, And in this company, we make a wristband 0:15:11.120 --> 0:15:13.440 for people who are deaf, and the wristband has a 0:15:13.480 --> 0:15:16.200 microphone that picks up on the sounds around you, and 0:15:16.240 --> 0:15:19.440 then it turns those sounds into patterns vibration on the skin. 0:15:19.760 --> 0:15:23.000 So it's doing what your inner ear, your cochlea is doing. 0:15:23.240 --> 0:15:26.640 We're just transferring that job to the skin. Now, here's 0:15:26.640 --> 0:15:29.280 what I wanted to touch on. Typically, when someone puts 0:15:29.320 --> 0:15:32.600 this on for the first time, they say, WHOA, I'm 0:15:32.680 --> 0:15:36.560 feeling so much buzzing going on around here, and it 0:15:36.640 --> 0:15:39.560 takes them a little while to understand that all this 0:15:39.680 --> 0:15:44.040 buzzing is part of this soundscape around them. In other words, 0:15:44.080 --> 0:15:47.040 the next time you're at the cafe, pay attention not 0:15:47.120 --> 0:15:51.240 just to your companion's conversation, but really listen to all 0:15:51.400 --> 0:15:55.000 the sounds that are hitting your ears. The overhead music 0:15:55.120 --> 0:15:58.280 and the barista and the coffee machines, and the doors 0:15:58.360 --> 0:16:01.960 opening and closing, and the squeaking and every single one 0:16:02.000 --> 0:16:04.800 of the conversations going on around you. All of that 0:16:05.040 --> 0:16:09.040 is hitting your ear drums. We're so good at ignoring 0:16:09.160 --> 0:16:10.960 all of that that we only take in what is 0:16:11.000 --> 0:16:13.440 salient to us. And it's only when you push that 0:16:13.520 --> 0:16:16.400 information in through a new channel, like through your wrist, 0:16:16.720 --> 0:16:20.320 that you realize how much noise is actually out there. 0:16:20.760 --> 0:16:24.680 Your brain has to learn how to pay selective attention 0:16:25.080 --> 0:16:29.080 to the vibrations on your wrist. Okay, so that's selective attention. 0:16:29.400 --> 0:16:31.800 How does this kind of thing get studied in the 0:16:31.840 --> 0:16:36.240 neuroscience laboratory. Well, think of something like a card game 0:16:36.320 --> 0:16:39.080 like the one called Set or something called the Wisconsin 0:16:39.160 --> 0:16:42.520 Card Sorting Task. And in games like this, each card 0:16:43.080 --> 0:16:46.520 has some number of shapes of a certain color. So 0:16:46.640 --> 0:16:49.520 this card might have three blue diamonds and this one 0:16:49.640 --> 0:16:54.400 has four red squares, and that one has two green circles. 0:16:54.680 --> 0:16:58.520 And let's say you're trying to match a certain property 0:16:58.600 --> 0:17:02.440 between cards, like the blue color and the shapes and 0:17:02.480 --> 0:17:05.160 the number of shapes. That doesn't matter. So because you're 0:17:05.200 --> 0:17:09.159 holding this rule in mind match the color, the neural 0:17:09.240 --> 0:17:13.800 networks in your visual cortex are mostly responding just to 0:17:14.040 --> 0:17:17.600 the color you stare at the cards, but really it's 0:17:17.640 --> 0:17:21.560 only your neurons involved in color that are shouting off. 0:17:22.320 --> 0:17:24.960 Now the rule changes and you have to pay attention 0:17:25.440 --> 0:17:28.720 to the number of objects. So you match a card 0:17:28.760 --> 0:17:31.520 with three shapes over here to a card with three 0:17:31.560 --> 0:17:35.000 shapes over there. Now, what hits your eyes is exactly 0:17:35.040 --> 0:17:37.159 the same, but the cells in your brain that are 0:17:37.200 --> 0:17:42.360 responding are those involved in numeracy and same when you're 0:17:42.440 --> 0:17:46.160 concentrating on matching shapes, like you want to match squares 0:17:46.200 --> 0:17:49.200 here with squares there. Your brain now plays a very 0:17:49.240 --> 0:17:52.919 different game. Everything hits the eyes, but your brain only 0:17:53.080 --> 0:17:57.560 really runs the algorithms involved in shape. So we can 0:17:57.640 --> 0:18:01.439 see selective attention directly in the brain. So back to 0:18:01.480 --> 0:18:05.879 the beta Miinhoff effect. Our brains encode almost nothing around 0:18:05.960 --> 0:18:10.560 us unless it somehow becomes relevant to us. Once something 0:18:10.920 --> 0:18:13.879 has been flagged as important or interesting to your brain, 0:18:14.280 --> 0:18:17.760 you will notice it more, creating the illusion that it 0:18:17.800 --> 0:18:20.480 is suddenly everywhere, And that leads us to the second 0:18:20.920 --> 0:18:25.000 parent of the better Minhoft love child, which is another 0:18:25.160 --> 0:18:29.919 psychological process running under the hood. The first was selective attention. 0:18:30.040 --> 0:18:34.560 The second is called confirmation bias. Confirmation bias is our 0:18:34.760 --> 0:18:39.199 tendency to notice everything that confirms what we already believe. 0:18:39.720 --> 0:18:42.760 We notice that we focus on it, we give greater 0:18:43.000 --> 0:18:47.919 credence to it. So generally we talk about confirmation bias 0:18:48.000 --> 0:18:53.119 as favoring information that confirms our pre existing beliefs and 0:18:53.320 --> 0:18:57.399 disregarding evidence that contradicts our beliefs. Or we seek out 0:18:57.480 --> 0:19:02.600 information that supports our existing views, or we interpret ambiguous 0:19:02.640 --> 0:19:05.760 evidence in a way that supports our existing opinions. But 0:19:05.840 --> 0:19:09.040 in this case of the beta Minhoff phenomenon, there's a 0:19:09.119 --> 0:19:12.760 simpler aspect of this going on. We are simply more 0:19:12.960 --> 0:19:16.480 likely to take the second appearance of the new word 0:19:16.600 --> 0:19:20.199 or car or shoe as something meaningful. Our brains are 0:19:20.280 --> 0:19:22.720 now on the lookout for it, and every time we 0:19:23.160 --> 0:19:26.280 see it or hear that word or song or fact again, 0:19:26.680 --> 0:19:32.600 that strengthens our belief our conviction that the event is 0:19:32.760 --> 0:19:36.840 happening with increased frequency, and lots of research shows that 0:19:36.880 --> 0:19:41.920 we remember information that supports our beliefs more readily than 0:19:42.040 --> 0:19:45.879 information that contradicts them. So once a new piece of 0:19:45.960 --> 0:19:49.879 information is introduced to our minds, like the name of 0:19:49.920 --> 0:19:53.000 a band or a new gadget, confirmation bias makes us 0:19:53.080 --> 0:19:57.520 believe that the thing is now suddenly ubiquitous. Every time 0:19:57.560 --> 0:20:02.520 we see it, we further sum meant this belief into place. 0:20:03.080 --> 0:20:07.000 So researchers typically talk about the beta Minhoff phenomenon in 0:20:07.119 --> 0:20:11.199 terms of selective attention and confirmation bias. But because I 0:20:11.240 --> 0:20:14.359 talk about internal models all the time, I want to 0:20:14.400 --> 0:20:16.560 cast what we're talking about today in terms of that. 0:20:17.400 --> 0:20:19.879 So as a reminder, your brain works to make a 0:20:19.960 --> 0:20:23.200 model of what is going on in the world out there. 0:20:23.240 --> 0:20:25.720 That's the internal model. Now, the part I want to 0:20:25.720 --> 0:20:29.840 emphasize is just how that actually limits your perception of 0:20:29.880 --> 0:20:32.760 the world. So let me give you an example. Imagine 0:20:32.760 --> 0:20:36.560 that you're on a hike with friends, and as you're 0:20:36.600 --> 0:20:40.600 tramping up the trails, one of your friends is a 0:20:40.800 --> 0:20:46.199 mycologist who studies mushrooms, and he notices the colors and 0:20:46.520 --> 0:20:50.800 unusually broad variety of the mushrooms around you. Now these 0:20:50.880 --> 0:20:53.199 fung guy at the base of trees, these have been 0:20:53.280 --> 0:20:58.119 hitting everyone's retinas, but this detail was invisible to the 0:20:58.119 --> 0:21:00.679 rest of you. Let's say one of your other friends 0:21:00.680 --> 0:21:04.360 as a pediatrist, and she notices something about the subtle 0:21:04.480 --> 0:21:07.440 turn of your feet as you hike, but no one 0:21:07.440 --> 0:21:11.240 else saw that. And your other friend is a climate scientist, 0:21:11.280 --> 0:21:15.520 and he notices the erosion around the tree line, and 0:21:15.680 --> 0:21:18.240 that's invisible to all the rest of you, even though 0:21:18.280 --> 0:21:21.480 all your eyes are pointing in that direction. The point 0:21:21.520 --> 0:21:24.879 is the same, data is laid out before all of you, 0:21:25.520 --> 0:21:29.600 but each of you knows how to look differently. Your 0:21:29.640 --> 0:21:33.600 internal model is built from the raw materials of your 0:21:33.720 --> 0:21:37.479 experience with the world, and each of us vacuums up 0:21:37.520 --> 0:21:41.919 the data from our peculiar thin trajectory in the world, 0:21:42.000 --> 0:21:46.240 which builds up our models and therefore our interpretations and 0:21:46.320 --> 0:21:51.480 our biases based on our experiences. And that's why when 0:21:51.520 --> 0:21:56.240 you're hiking, everyone is seeing a different world. So that's 0:21:56.320 --> 0:22:00.919 why we notice things more once they have some relevance 0:22:00.960 --> 0:22:05.480 to us, because the borders of your internal model have stretched, 0:22:05.840 --> 0:22:09.880 so now your brain catches words or cars or shoes 0:22:10.119 --> 0:22:14.840 that were previously just noise. These items had no filter 0:22:15.080 --> 0:22:18.600 shaped like them, and therefore they're mentioned was as good 0:22:18.640 --> 0:22:22.760 as invisible. Now that the model is stretched, you catch 0:22:22.800 --> 0:22:26.560 that new thing all over the place. So let's summarize 0:22:26.600 --> 0:22:29.399 where we are so far. Although we might swear that 0:22:29.440 --> 0:22:34.560 we've never heard a term mentioned before, the shoddiness of 0:22:34.680 --> 0:22:39.199 human memories suggests that's generally not true. Instead, what the 0:22:39.240 --> 0:22:44.240 beter Minhof phenomenon illustrates is how deeply we filter the world. 0:22:44.440 --> 0:22:48.520 We don't notice those things that we have no interpretation for. 0:22:49.040 --> 0:22:53.000 So think of walking in a foreign country and hearing 0:22:53.080 --> 0:22:57.040 people's speech simply as background noise, something you can't interpret. 0:22:57.680 --> 0:23:02.600 If you then learn like the word for good or 0:23:02.640 --> 0:23:06.760 for sorry or for thanks, then that constantly jumps out 0:23:06.760 --> 0:23:09.600 of the conversation to your ears. Why because you can 0:23:09.640 --> 0:23:12.520 now interpret it. It's not that people are saying it 0:23:12.560 --> 0:23:16.080 more often, It's that you finally figured out how to 0:23:16.440 --> 0:23:20.679 listen for it. As you better model the world, the 0:23:20.760 --> 0:23:23.639 more you catch. So even though you may think you 0:23:23.760 --> 0:23:28.520 never heard of this nineteen seventies German domestic terrorist group before, 0:23:29.119 --> 0:23:32.560 you presumably have, but you were busy with the other 0:23:32.800 --> 0:23:36.920 things of salience in your life, and it just didn't register. 0:23:37.760 --> 0:23:42.440 You were counting basketball passes and missed this gorilla. If 0:23:42.480 --> 0:23:45.919 you don't have a model for things, you just don't 0:23:46.160 --> 0:24:06.360 catch them. Now, the fascinating thing about the beter Miinhoff 0:24:06.359 --> 0:24:08.680 phenomenon is that it shows up in lots of ways. 0:24:08.800 --> 0:24:13.000 Once your attention is drawn to something, some words, some car, 0:24:13.119 --> 0:24:16.480 or some shoes, it seems to you to be everywhere. 0:24:17.040 --> 0:24:20.560 And the linguist that I mentioned before, Arnold's Wicki points 0:24:20.600 --> 0:24:23.560 out that there are several special cases of this. For example, 0:24:23.920 --> 0:24:27.560 one is called the outgroup illusion, and this is where 0:24:27.600 --> 0:24:31.000 there's some characteristic of a group who you don't like, 0:24:31.520 --> 0:24:36.040 and you start seeing that characteristic more often. So, for example, 0:24:36.280 --> 0:24:40.160 imagine that there are two groups who live on opposite 0:24:40.200 --> 0:24:44.639 sides of the track, the Augustinians and the Justinians. So 0:24:44.720 --> 0:24:48.920 let's say you're an Augustinian and you think the Justinians 0:24:49.240 --> 0:24:54.080 are very selfish. So then anytime some Justinian happens to 0:24:54.359 --> 0:24:58.800 act selfishly, you're noticing that, just like you would notice 0:24:58.840 --> 0:25:01.640 the new word or the shoe chooser the car. Your 0:25:01.760 --> 0:25:06.240 confirmation bias says, yes, I knew that was true about 0:25:06.240 --> 0:25:10.919 those guys, but presumably you don't notice as diligently the 0:25:11.000 --> 0:25:14.840 equivalent behaviors in your own group. And on the other 0:25:14.880 --> 0:25:18.520 side of the tracks, the Justinians look at the Augustinians 0:25:18.560 --> 0:25:22.879 and think they are very immoral. So whenever an Augustinian 0:25:22.960 --> 0:25:27.720 behaves poorly, a version of the better Meinhoff phenomenon takes 0:25:27.760 --> 0:25:31.480 place here, and it seems like this is a characteristic 0:25:31.560 --> 0:25:35.080 that is seen a whole lot in those Augustinians. And 0:25:35.200 --> 0:25:38.640 of course the Justinians will tend to notice less when 0:25:38.760 --> 0:25:42.920 immoral behavior happens in their own group, because their internal 0:25:42.960 --> 0:25:47.960 models suggest, like everybody's internal models, that their in group 0:25:48.240 --> 0:25:53.040 generally behaves well, behaves within certain moral boundaries, but that 0:25:53.280 --> 0:25:57.520 outgroup they are capable of anything. Now, relatedly, there was 0:25:57.560 --> 0:25:59.640 a paper a couple of years ago out of Duke 0:25:59.720 --> 0:26:03.520 Universe where they showed that if there's some group that 0:26:03.600 --> 0:26:08.240 you feel is threatening to you, they seem more populous, 0:26:08.359 --> 0:26:11.800 as in, there seem to be more of them, presumably 0:26:11.840 --> 0:26:15.960 because your internal model is noticing them more. Look, there's 0:26:16.000 --> 0:26:18.600 one on the sidewalk, there's one in the clothing store, 0:26:18.640 --> 0:26:21.960 there's one in that restaurant. So just like the beta 0:26:22.040 --> 0:26:26.960 Minhof phenomenon. Your estimates are off based on the details 0:26:27.040 --> 0:26:30.160 of your internal model. Why does this happen Because you're 0:26:30.200 --> 0:26:33.280 not noticing the groups that you are not threatened by. 0:26:33.359 --> 0:26:36.879 You're noticing those individuals who belong to the group that 0:26:36.960 --> 0:26:40.240