WEBVTT - Ep10 rebroadcast "Why is it so hard to spot a counterfeit bill?" 0:00:00.080 --> 0:00:02.160 Hi, this is David Eagleman. I want to wish you 0:00:02.200 --> 0:00:05.080 a very happy holidays. We're going to take a break 0:00:05.080 --> 0:00:06.560 for a couple of weeks, and then we're back in 0:00:06.680 --> 0:00:12.840 January with new episodes on emotion intelligence, time perception, smell 0:00:12.880 --> 0:00:16.720 and taste, brain, computer interfaces, and much more. In the meantime, 0:00:16.720 --> 0:00:19.160 we're going to replay one of our favorite episodes from 0:00:19.160 --> 0:00:21.600 the past year, and I'll look forward to seeing you 0:00:21.640 --> 0:00:32.240 in January. What do counterfeit bills have to do with brains? 0:00:32.479 --> 0:00:35.000 And why is it so hard for you to recognize 0:00:35.040 --> 0:00:37.680 a real penny? And what does this have to do 0:00:37.800 --> 0:00:42.120 with the building on the fifty euro bill recently getting 0:00:42.159 --> 0:00:45.960 replaced with the face of the goddess Europa And what 0:00:46.040 --> 0:00:47.839 did I have to do with that? And why did 0:00:47.920 --> 0:00:51.400 I campaign to the European Central Bank that all bills 0:00:51.479 --> 0:00:54.720 should be blank with a single hologram in the middle. 0:00:58.720 --> 0:01:03.200 Welcome to intercouse with me, David Eagleman. I'm a neuroscientist, 0:01:03.520 --> 0:01:07.200 and you might wonder how a neuroscientist got involved in 0:01:07.280 --> 0:01:12.040 studying counterfeit money. Well, hangtight, because in this episode we're 0:01:12.040 --> 0:01:22.440 going to see how all these issues intersect. One of 0:01:22.520 --> 0:01:26.400 the most pervasive mistakes that we make is believing that 0:01:26.440 --> 0:01:30.320 our visual systems give us a faithful representation of what's 0:01:30.400 --> 0:01:33.440 out there. We think of our vision the way we 0:01:33.480 --> 0:01:36.480 think of a movie camera, like it's just capturing the 0:01:36.520 --> 0:01:39.920 information from the world. But there are a whole bunch 0:01:39.959 --> 0:01:45.039 of simple demonstrations that can quickly disabuse you of this notion. 0:01:45.720 --> 0:01:49.760 For example, imagine that I show you two pictures. I 0:01:50.120 --> 0:01:52.320 flash one, then it goes off, Then the second one 0:01:52.360 --> 0:01:54.400 comes on, then it goes off, then the first one 0:01:54.440 --> 0:01:57.279 comes back on, and so on. So it alternates between 0:01:57.320 --> 0:02:00.240 pictures A and B, and it looks like it's the 0:02:00.280 --> 0:02:03.800 same picture. But let's say I let you know that 0:02:03.840 --> 0:02:07.400 there is a difference between the pictures, and your job 0:02:07.600 --> 0:02:10.560 is simply to spot the difference. Well, if you've ever 0:02:10.639 --> 0:02:14.840 played these sorts of games before, you know how surprisingly 0:02:15.000 --> 0:02:18.360 difficult it can be to tell the difference between them. 0:02:18.720 --> 0:02:22.239 It turns out that we are really blind to changes 0:02:22.520 --> 0:02:25.800 in the scene, and these can be quite large changes. 0:02:26.160 --> 0:02:29.440 So you might have a big statue that's present in 0:02:29.480 --> 0:02:32.399 the background of one photo and not the other, or 0:02:32.560 --> 0:02:37.720 a jeep or an airplane, and the difference goes unseen. 0:02:38.280 --> 0:02:40.639 You can see some demos of this at Eagleman dot 0:02:40.639 --> 0:02:41.880 com Slash podcast. 0:02:42.120 --> 0:02:43.440 So what happens is. 0:02:43.360 --> 0:02:48.359 That your attention slowly crawls over the photograph and you 0:02:48.520 --> 0:02:54.200 analyze the interesting landmarks until you finally detect what is changing. 0:02:54.480 --> 0:02:57.920 And once your brain has latched on to the appropriate object, 0:02:58.080 --> 0:03:01.720 then the change is trivial to see. But this happens 0:03:01.880 --> 0:03:06.880 only after a pretty exhaustive inspection. This is what's known 0:03:06.960 --> 0:03:13.320 as change blindness, and it highlights the importance of attention, 0:03:13.960 --> 0:03:17.560 which is to say, to see an object change, you 0:03:17.639 --> 0:03:20.560 have to attend to it. And what an experiment like 0:03:20.600 --> 0:03:24.160 this surface is is that you're not actually seeing the 0:03:24.160 --> 0:03:28.959 world with the rich detail that you implicitly believed you were. 0:03:29.720 --> 0:03:32.880 In fact, you're not aware of most of what's hitting 0:03:32.919 --> 0:03:36.440 your eyes. So imagine you're watching a short film with 0:03:36.600 --> 0:03:38.280 a single actor in it. 0:03:38.360 --> 0:03:40.440 He's cooking an omelet. 0:03:40.200 --> 0:03:43.960 And the camera cuts to a different angle as the 0:03:44.000 --> 0:03:48.040 actor continues his cooking. Surely you would notice if the 0:03:48.080 --> 0:03:52.720 actor changed into a different person, right, But in studies, 0:03:53.280 --> 0:03:58.280 two thirds of observers don't. In one really astonishing demonstration 0:03:58.360 --> 0:04:02.840 of change blindness, random pedestrians in a courtyard were stopped 0:04:02.920 --> 0:04:05.640 by an experimenter, and they were asked for directions, so 0:04:05.680 --> 0:04:09.200 they start giving directions, and at some point, as the 0:04:09.680 --> 0:04:13.400 unsuspecting person is in the middle of explaining this, workmen 0:04:13.760 --> 0:04:17.880 carrying a door walk right in between the two people, 0:04:18.400 --> 0:04:24.880 and unbeknownst to the pedestrian, the experimenter is stealthily replaced 0:04:25.320 --> 0:04:28.280 by a confederate who's been hiding behind the door. So 0:04:28.440 --> 0:04:31.960 the people are swapped, and after the door passes, a 0:04:32.120 --> 0:04:35.640 new person is standing there, and the majority of subjects 0:04:36.040 --> 0:04:40.479 continue giving directions without noticing that the person they were 0:04:40.480 --> 0:04:43.760 talking to is not the same one that they were 0:04:43.839 --> 0:04:48.400 talking to. In other words, they were only encoding small 0:04:48.440 --> 0:04:51.640 amounts of the information that was hitting their eyes. The 0:04:51.800 --> 0:04:57.400 rest was assumption. Now, neuroscientists weren't the first to discover 0:04:57.839 --> 0:05:01.000 that placing your eyes on something is no guarantee of 0:05:01.040 --> 0:05:04.880 seeing it. Magicians figured this out long ago, and they 0:05:04.960 --> 0:05:10.240 perfected ways of leveraging this knowledge. So by directing your attention, 0:05:11.240 --> 0:05:16.000 magicians perform sleight of hand in full view. Their actions. 0:05:16.000 --> 0:05:19.680 Their hand motions should give away the game, but they 0:05:19.720 --> 0:05:24.960 can rest assured that your brain processes only small bits 0:05:25.040 --> 0:05:28.479 of the visual scene. It's not everything that's hitting your retinas. 0:05:28.839 --> 0:05:32.520 And by the way, this limitation of our vision helps 0:05:32.560 --> 0:05:36.520 to explain the colossal number of traffic accidents in which 0:05:36.640 --> 0:05:40.680 drivers will hit pedestrians who are in plane view, or 0:05:40.720 --> 0:05:43.400 they'll collide with cars that are right in front of them. 0:05:43.520 --> 0:05:47.680 Sometimes you'll even see stories where a car intersects with 0:05:47.760 --> 0:05:50.920 a train on the track. In some of these cases, 0:05:51.520 --> 0:05:54.360 the eyes are in the right place, but the brain 0:05:54.640 --> 0:06:01.000 isn't registering the stimulus. Vision is more than looking now. 0:06:01.160 --> 0:06:03.960 This is all an example of a larger point, which 0:06:04.040 --> 0:06:07.440 is that we're not generally aware of what's going on 0:06:07.720 --> 0:06:11.919 until we ask the question. So, for example, what is 0:06:11.960 --> 0:06:14.919 the position of your tongue in your mouth right now? 0:06:15.360 --> 0:06:18.080 Once you're asked the question, you can answer it, but. 0:06:18.040 --> 0:06:22.560 Presumably you weren't aware of the answer until I asked 0:06:22.640 --> 0:06:26.400 the question. Why does the brain work this way because 0:06:26.839 --> 0:06:31.000 your brain generally doesn't need to know most things. It 0:06:31.279 --> 0:06:34.400 simply needs to know how to go out and retrieve 0:06:34.480 --> 0:06:37.400 the data when it needs it. It perceives things on 0:06:37.480 --> 0:06:42.240 a need to know basis. You don't continuously track the 0:06:42.279 --> 0:06:45.719 position of your tongue in your consciousness because that knowledge 0:06:46.120 --> 0:06:50.400 is only useful in rare circumstances. So again, you're not 0:06:51.000 --> 0:06:54.680 aware of much of anything until you ask yourself about it. 0:06:55.200 --> 0:06:58.520 What does your left shoe feel like on your foot 0:06:58.560 --> 0:06:59.920 right now? Pit? 0:07:00.000 --> 0:07:02.919 Which is the hum of the air conditioner in the background. 0:07:03.600 --> 0:07:07.679 Just like with change blindness, we are unaware of most 0:07:07.800 --> 0:07:11.120 of what should be obvious to our senses. It's only 0:07:11.200 --> 0:07:15.080 after we deploy our attention onto small bits of the 0:07:15.120 --> 0:07:18.200 scene that we become aware of what we were missing, 0:07:18.680 --> 0:07:23.840 And before we engage our attention, we're typically not aware 0:07:24.520 --> 0:07:27.360 that we're not aware of those details. So not only 0:07:27.480 --> 0:07:30.640 is our perception of the world a construction that doesn't 0:07:30.680 --> 0:07:35.000 accurately represent the outside, it's weirder than that because we 0:07:35.120 --> 0:07:40.040 have the false impression of a full, rich picture, when 0:07:40.080 --> 0:07:43.680 in fact, we only see what we need to know 0:07:43.800 --> 0:07:46.280 and what we ask questions about, and we don't see 0:07:46.280 --> 0:07:48.640 anything more than that. Now, this is a topic I'm 0:07:48.680 --> 0:07:52.040 going to return to in many episodes, but for now 0:07:52.720 --> 0:07:55.400 I want to ask what this has to do with 0:07:55.560 --> 0:08:05.200 the economic currencies on this planet. So the scene is Lima, Peru, 0:08:05.480 --> 0:08:09.120 a little before dawn. There's a bunch of houses and buildings, 0:08:09.520 --> 0:08:15.000 and suddenly fifteen hundred Peruvian police officers and US Secret 0:08:15.040 --> 0:08:19.840 Service agents stormed the doors. This was called Operation Sunset, 0:08:19.920 --> 0:08:23.200 and it was the finale of a long running investigation 0:08:23.360 --> 0:08:28.200 aimed at reducing the flow of counterfeit US dollars from Peru. 0:08:28.560 --> 0:08:32.680 They brought down six plants that had presses and plates 0:08:32.720 --> 0:08:36.520 and negatives and piles of fake one hundred dollars bills. 0:08:36.960 --> 0:08:39.880 And in just the past few years, counterfeit notes worth 0:08:39.920 --> 0:08:44.720 more than seventy five million dollars have been confiscated in Peru. 0:08:44.800 --> 0:08:47.680 And this is just one of lots of countries where 0:08:47.720 --> 0:08:50.840 you have fraudsters who are cooking up fake dollars. And 0:08:50.880 --> 0:08:55.720 it's essentially impossible to estimate the volume of counterfeit banknotes 0:08:56.080 --> 0:08:58.199 passing through the US economy. 0:08:58.559 --> 0:09:01.359 What you find is that the alreadies claim they've. 0:09:01.160 --> 0:09:04.760 Captured most of the fakes, and the counterfeitters claim exactly 0:09:04.760 --> 0:09:09.960 the opposite. Anyhow, counterfeiting is a problem that faces every 0:09:09.960 --> 0:09:15.000 major player in the currency game. Russia seizes counterfeits in 0:09:15.000 --> 0:09:18.720 the range of two hundred million rubles per year. China 0:09:18.840 --> 0:09:23.280 captures five hundred million fake yuan. A few years ago, 0:09:23.760 --> 0:09:27.880 India terminated their five hundred and one thousand rupee notes, 0:09:28.200 --> 0:09:30.320 and this was at least partially because of the rising 0:09:30.400 --> 0:09:35.959 concern about forgeries and their inability to stem that. In Venezuela, 0:09:36.000 --> 0:09:40.080 the inflation was so bad that it became worthwhile for 0:09:40.280 --> 0:09:43.920 fraudsters to bleach the ten bole of our bill and 0:09:44.080 --> 0:10:05.959 use the paper to forge notes in other currencies. So 0:10:06.320 --> 0:10:10.640 it turns out there are counterfeiters everywhere, and they're busy 0:10:10.760 --> 0:10:16.200 pushing fake banknotes into streets and squares in every major city. 0:10:16.720 --> 0:10:19.640 You may wonder how I came to care about counterfeits, 0:10:19.679 --> 0:10:23.440 given that I'm a neuroscientist. Well, what happened is some 0:10:23.640 --> 0:10:26.480 years ago I was giving a presentation at a brain 0:10:26.559 --> 0:10:29.880 science conference, and afterwards I was approached by this guy 0:10:30.160 --> 0:10:33.400 and we started talking, and I was surprised to learn 0:10:33.440 --> 0:10:36.360 that he was working for the European Central Bank. And 0:10:36.400 --> 0:10:38.880 I was surprised because it wasn't clear to me why 0:10:38.920 --> 0:10:43.040 someone like that would be at a highly academic neuroscience conference. 0:10:43.360 --> 0:10:46.480 As it turns out, his job was to find a 0:10:46.600 --> 0:10:52.280 solution to Europe's counterfeiting problem. Their problem was that the 0:10:52.559 --> 0:10:57.560 EU comes across hundreds of thousands of fake euro banknotes 0:10:57.640 --> 0:11:01.600 every year, mostly in the form of twenties and fifties. 0:11:02.600 --> 0:11:05.400 At the heart of the issue is the fact that 0:11:05.520 --> 0:11:10.920 governments spend millions creating security features. So while we're talking, 0:11:10.960 --> 0:11:14.000 take out a twenty dollars bill, or whatever country you're in, 0:11:14.040 --> 0:11:15.240 take out a high value bill. 0:11:15.280 --> 0:11:17.640 They all have similar security features. 0:11:17.880 --> 0:11:20.200 Now, if you look at it, you'll see that as 0:11:20.240 --> 0:11:24.880 you tilt the note slightly back and forth, there's color 0:11:25.080 --> 0:11:28.320 changing ink. So the color changes here of let's say 0:11:28.320 --> 0:11:30.480 the twenty and now if you hold this up to 0:11:30.559 --> 0:11:34.679 the light, what you'll see is a water mark that's 0:11:34.760 --> 0:11:38.040 otherwise invisible to you. You can't see it normally, and 0:11:38.040 --> 0:11:40.040 it's not printed on the front or the back, but 0:11:40.120 --> 0:11:43.960 it's printed on an intermediate layer and on euros, and 0:11:44.040 --> 0:11:46.760 actually on the one hundred dollars bill, what you see 0:11:46.800 --> 0:11:50.960 are holograms. So when you angle this and change this, 0:11:51.520 --> 0:11:54.160 it moves around slightly and appears to be three D. 0:11:54.840 --> 0:11:59.920 And you'll also see colored strips that display the value. 0:11:59.520 --> 0:12:00.280 Of the note. 0:12:00.360 --> 0:12:03.440 And if you happen to have an ultraviolet light, turn 0:12:03.520 --> 0:12:06.040 that on and hold the bill under it, and you'll 0:12:06.040 --> 0:12:09.080 see that the bill turns very colorful. You've got all 0:12:09.120 --> 0:12:14.360 these invisible fibers in it that fluoresce under ultraviolet. So 0:12:14.520 --> 0:12:19.520 a piece of currency is a bustling collection of security devices. 0:12:20.000 --> 0:12:23.679 The problem is that no one pays attention to these 0:12:23.720 --> 0:12:28.320 security devices. People who pass notes around in daily transactions 0:12:28.880 --> 0:12:34.240 don't typically pause to examine them carefully. The special ink, 0:12:34.400 --> 0:12:39.080 the watermark, the strip, the fibers, the hologram. These are 0:12:39.240 --> 0:12:43.880 very expensive security features and they are a totally wasted 0:12:44.200 --> 0:12:48.280 effort because counterfeit bills get passed from person to person 0:12:48.640 --> 0:12:52.800 and no one ever notices these or notices if they 0:12:52.840 --> 0:12:57.640 are badly rendered. Do you know how counterfeit bills almost 0:12:57.679 --> 0:13:01.520 always are found at the bank And that's because real 0:13:01.600 --> 0:13:05.480 notes have one more feature. They have machine readable code 0:13:05.640 --> 0:13:08.520 in the bill. Counterfeitters can do a good job making 0:13:08.559 --> 0:13:11.320 a counterfeit, but they can't make the machine readable code, 0:13:11.559 --> 0:13:14.640 and that's why the bank catches those. But the counterfeitters 0:13:14.679 --> 0:13:17.319 don't need to do that because their bills get passed 0:13:17.360 --> 0:13:21.680 around and no one ever notices. And this problem of 0:13:21.840 --> 0:13:26.920 inattention led the European Central Bank to explore a simple question, 0:13:27.559 --> 0:13:31.120 could banknotes be better designed so that the person on 0:13:31.160 --> 0:13:34.480 the street would be more likely to notice when something 0:13:34.679 --> 0:13:37.720 was a miss? And for that they finally started to 0:13:37.760 --> 0:13:41.560 think about neuroscience, and that's how I got involved because 0:13:41.600 --> 0:13:45.200 as our understanding of the brain has developed over recent decades, 0:13:45.280 --> 0:13:49.520 neuroscience has forged insights into a range of fields, from 0:13:49.920 --> 0:13:54.440 early education to consumer behavior to government policy, and modern 0:13:54.480 --> 0:13:59.520 brain science has wide ranging applications, particularly when it comes 0:13:59.520 --> 0:14:02.760 to under standing the ways in which our perceptions of. 0:14:02.679 --> 0:14:04.599 The world are inaccurate. 0:14:05.040 --> 0:14:09.120 In the case of banknotes, the questions were straightforward, what 0:14:09.360 --> 0:14:13.040 details do we notice or not notice? And why? What 0:14:13.200 --> 0:14:18.080 features could be better designed to be brain compatible. Of 0:14:18.160 --> 0:14:22.960 all the anti counterfeiting measures that the government takes, which ones. 0:14:22.680 --> 0:14:25.640 Work and how could their efforts be better spent. 0:14:26.320 --> 0:14:29.040 So the man at the conference eventually contracted with me 0:14:29.160 --> 0:14:32.560 to help them figure out the question how should Europe 0:14:32.640 --> 0:14:35.960 redesign their bills? So we signed a bunch of security 0:14:35.960 --> 0:14:39.080 documents and then I flew to the headquarters of the 0:14:39.120 --> 0:14:42.880 European Central Bank in Frankfurt, Germany. Now this was very 0:14:42.920 --> 0:14:47.520 cool because although the building was just a boring government building, I. 0:14:47.520 --> 0:14:50.560 Was beeped through a badge protected. 0:14:50.080 --> 0:14:54.320 Door, and then another, and then another, and we finally 0:14:54.440 --> 0:14:58.040 arrived at this small inner room at the heart of 0:14:58.080 --> 0:15:01.000 the building, and there were all these piles of money 0:15:01.040 --> 0:15:04.000 of euros, mostly twenties and fifties, and. 0:15:03.960 --> 0:15:05.359 They were all counterfeit. 0:15:06.200 --> 0:15:10.160 Each pile had some feature that distinguished it, whether that 0:15:10.360 --> 0:15:13.360 was the way they had done the shiny foil or 0:15:13.360 --> 0:15:15.960 the way they had done the watermarks or whatever. And 0:15:16.000 --> 0:15:19.360 this is how the bank arranged them. Now, many of 0:15:19.400 --> 0:15:23.480 these forgeries were quite good. If you weren't paying super 0:15:23.480 --> 0:15:26.680 close attention, the kind of attention no one ever really 0:15:26.760 --> 0:15:29.800 pays to money, you couldn't tell there was anything amiss 0:15:30.480 --> 0:15:34.280 with most of these forgeries. You could tell under ultraviolet light. 0:15:34.400 --> 0:15:37.000 But if you didn't have that, there was really nothing 0:15:37.040 --> 0:15:40.160 that would even make you think to pay extra attention. 0:15:40.680 --> 0:15:42.920 And here's a weird fact that I learned there in 0:15:42.960 --> 0:15:48.120 the vaults, these really good forgeries, they always have a 0:15:48.320 --> 0:15:50.160 slight error in them. 0:15:50.560 --> 0:15:52.080 And think about why this is weird. 0:15:52.520 --> 0:15:57.280 You've got a otherwise amazingly well done counterfeit bill, but 0:15:57.400 --> 0:16:01.680 it's got an error. For example, there are essentially perfect 0:16:01.720 --> 0:16:05.680 counterfeits of the American twenty dollars bill. But if you 0:16:05.760 --> 0:16:09.960 look at the number two, the way that the bottom 0:16:10.000 --> 0:16:13.080 of the two spreads out what's called the trumpeting in 0:16:13.240 --> 0:16:17.680 font language. It's just slightly wider than it is on 0:16:17.720 --> 0:16:20.000 a real bill, So what's going on with that? How 0:16:20.000 --> 0:16:23.240 can you get the whole bill right but not get 0:16:23.240 --> 0:16:26.240 that little detail right? And there's actually one more part 0:16:26.240 --> 0:16:29.320 to this mystery. There's only one mistake on a bill 0:16:29.360 --> 0:16:32.120 and no more than that. And the answer is it's 0:16:32.160 --> 0:16:35.800 not a mistake. It's on purpose, and it's because counterfeitters 0:16:36.320 --> 0:16:40.080 need a way to distinguish their own bills from. 0:16:40.000 --> 0:16:40.800 The real ones. 0:16:41.200 --> 0:16:44.520 This is so they don't spend their real money accidentally. Okay, 0:16:44.600 --> 0:16:47.320 so some of these bills were almost perfect with this 0:16:47.480 --> 0:16:50.880 single tiny change, but that's not the part that's interesting. 0:16:51.280 --> 0:16:52.880 What's interesting is. 0:16:52.800 --> 0:16:57.200 That most of the counterfeits were not as expertly done 0:16:57.400 --> 0:17:00.680 as you might expect. On almost all of them, the 0:17:00.800 --> 0:17:04.760 part where the color changing ink normally is didn't actually 0:17:04.840 --> 0:17:08.400 change color, and on others the water mark had been 0:17:08.800 --> 0:17:13.119 hand drawn really poorly, and in one pile of notes, 0:17:13.200 --> 0:17:18.320 the silver hologram had been impersonated with a glued piece 0:17:18.400 --> 0:17:22.560 of shiny material from the seal of a compact disc case. 0:17:23.240 --> 0:17:23.840 Many of the. 0:17:23.720 --> 0:17:27.840 Notes had essentially none of the actual security features. But 0:17:27.920 --> 0:17:31.840 the surprising thing is this, To get a counterfeit into circulation, 0:17:32.880 --> 0:17:36.719 most counterfeiters don't bother to go through the trouble of 0:17:36.840 --> 0:17:40.639 making an almost perfect one. Why because it makes almost 0:17:40.720 --> 0:17:45.439 no difference. Almost no one in the general public notices 0:17:45.760 --> 0:17:48.840 a bad forgery. People will stuff the notes into their 0:17:48.840 --> 0:17:52.080 wallets and they'll pass them on, and the notes spread 0:17:52.240 --> 0:17:56.080 like a virus through the population. Because as banknote users, 0:17:56.520 --> 0:18:00.400 we are surprisingly unobservant. We just don't pay it tension 0:18:00.920 --> 0:18:04.800 to the bills in our hand. The government spends an 0:18:04.920 --> 0:18:08.240 incredible amount on security features, and we. 0:18:08.280 --> 0:18:11.160 Don't even look at it now. 0:18:11.320 --> 0:18:15.280 The ECB tried to tackle this problem by hosting public 0:18:15.320 --> 0:18:19.360 awareness campaigns where they encouraged people to stop and look 0:18:19.480 --> 0:18:23.800 carefully at what they were holding. But the approach failed. 0:18:23.840 --> 0:18:28.359 People wouldn't put in the effort. Governments carry enormously about 0:18:28.440 --> 0:18:32.600 security features and the population doesn't. So the Central Bank 0:18:33.040 --> 0:18:35.919 now wanted to try a different tack. They wanted to 0:18:35.960 --> 0:18:40.320 work out what the human visual system actually notices and 0:18:40.400 --> 0:18:44.359 why there was no point spending a massive amount on 0:18:44.440 --> 0:18:48.959 security features that are only noticed by security experts. And 0:18:49.000 --> 0:18:51.640 that was the beginning of a long relationship I had 0:18:51.680 --> 0:18:56.600 with the ECB. For legal reasons, they couldn't manufacture or 0:18:56.760 --> 0:18:59.680 mail to me counterfeit bills and so I had to 0:18:59.680 --> 0:19:05.000 figure out how to make pseudo bills myself. First I 0:19:05.000 --> 0:19:08.600 had to get approval from my university for this very 0:19:08.680 --> 0:19:12.080 unusual study, and then my students that I set to work. 0:19:12.480 --> 0:19:14.840 As it turns out, it's not that difficult to make 0:19:15.000 --> 0:19:18.520 a rough counterfeit. Now, it happens that I was recently 0:19:18.520 --> 0:19:22.280 talking with my friend Adam Savage of Mythbuster's fame, and 0:19:22.320 --> 0:19:25.199 it turned out that he also has done lots of 0:19:25.640 --> 0:19:29.800 forgeries of things, not illegally but legal versions of making 0:19:29.960 --> 0:19:33.159 replicas of all kinds of wonderful things. So before I 0:19:33.200 --> 0:19:35.719 tell you about my experiences, I wanted to ring up 0:19:35.760 --> 0:19:43.600 Adam to hear about his. So, Adam, you've done some counterfeiting, 0:19:43.880 --> 0:19:44.960 tell me about that. 0:19:47.400 --> 0:19:50.160 I don't want to brag that I've done some counterfeiting, 0:19:50.600 --> 0:19:54.639 but I'm obsessed with the idea of replication. That's one 0:19:54.680 --> 0:19:57.840 of my main practices here in the cave, and so like. 0:19:58.119 --> 0:20:02.120 Just for instance, recently I obtained a really lovely replica 0:20:02.160 --> 0:20:05.720 of one of Leonardo da Vinci's coticies. This is known 0:20:05.760 --> 0:20:09.280 as the Paris manuscript A, and I've been slowly spending 0:20:09.280 --> 0:20:12.639 a couple of weeks making a completely accurate cover for 0:20:12.760 --> 0:20:16.120 it based on the information from the collection that it's 0:20:16.119 --> 0:20:18.520 since So I've been printing up the cover full size 0:20:18.520 --> 0:20:21.800 of and having the collection stamps made, and I'm gonna 0:20:22.080 --> 0:20:25.240 slowly make this thing indistinguishable from the original. So that's 0:20:25.280 --> 0:20:29.080 the kind of counterfeiting I really dig is the experiential counterfeit. 0:20:29.200 --> 0:20:31.359 Oh that's amazing. Now let me ask you this. 0:20:31.920 --> 0:20:36.040 Do you find that if there are changes, things you 0:20:36.119 --> 0:20:37.120 don't get perfect? 0:20:37.119 --> 0:20:39.680 Do you find that people don't notice those. 0:20:40.760 --> 0:20:46.800 People's threshold for what they notice is it's an interesting 0:20:46.880 --> 0:20:51.200 mixture because you can kind of bypass some of their 0:20:51.320 --> 0:20:55.160 filters with the right amount of aging, or even sometimes 0:20:55.200 --> 0:20:58.560 a smell will just have someone bypass all their filters 0:20:58.560 --> 0:21:01.720 for what they can see. So it's really different object 0:21:01.720 --> 0:21:05.399 by object about what the threshold is for what lends 0:21:05.440 --> 0:21:09.119 what feels like an experience of veracity. 0:21:09.720 --> 0:21:12.120 Have you ever tried counterfeiting money? Is that something you've 0:21:12.160 --> 0:21:14.440 ever tried? I have made. 0:21:14.680 --> 0:21:16.920 So behind me here on my desk is one of 0:21:16.960 --> 0:21:19.160 my favorite pieces in my collection, and it's the bank 0:21:19.200 --> 0:21:22.960 box from the Bourne identity. And this is the actual 0:21:23.000 --> 0:21:25.680 bank box from the Borne identity. I bought it empty, 0:21:26.440 --> 0:21:29.080 but one of my long term plans is to actually 0:21:29.119 --> 0:21:33.679 fill it full of original bills or correct bills. And 0:21:33.920 --> 0:21:36.200 this is made easier by the fact that a lot 0:21:36.240 --> 0:21:38.160 of the bills that were in the Borne Identity bank 0:21:38.200 --> 0:21:42.760 box are now euros, and so those original currencies, the 0:21:42.800 --> 0:21:46.760 real it's, the frank, et cetera, are no longer extant, 0:21:46.920 --> 0:21:51.760 and so I get suitably and appropriately nervous even doing 0:21:51.800 --> 0:21:55.879 Google searches about replicating money. But to me, your job 0:21:55.960 --> 0:21:59.320 of actually trying some counterfeits to see what works sounds 0:21:59.320 --> 0:22:00.840 like an absolute dream come true. 0:22:00.840 --> 0:22:02.359 For me. I was like I was born to be 0:22:02.400 --> 0:22:03.040 a pen tester. 0:22:17.240 --> 0:22:21.080 Okay, so back to my counterfeiting contract. So the notes 0:22:21.080 --> 0:22:24.120 that we made in the lab had no security features 0:22:24.160 --> 0:22:27.440 like holograms or color changing ink, because that wasn't precisely 0:22:27.480 --> 0:22:30.840 what we were studying. We were studying how much somebody 0:22:30.880 --> 0:22:34.639 would notice if we presented them with a bill with 0:22:34.680 --> 0:22:38.239 a particular design, and we let them examine it and 0:22:38.359 --> 0:22:40.960 turn it around in their hands, and then we took 0:22:41.000 --> 0:22:44.520 it back and we handed them another note, and that 0:22:44.680 --> 0:22:47.919 other note might have some change that was made to it, 0:22:48.200 --> 0:22:50.960 and we asked them, is this the same bill or 0:22:51.000 --> 0:22:53.840 a different one? In other words, how much do they 0:22:53.880 --> 0:22:57.840 remember about the note that they just examined. Now, you 0:22:57.920 --> 0:23:01.440 might expect that if you get to examine a note carefully, 0:23:01.480 --> 0:23:03.400 it's not going to be any problem to see when 0:23:03.440 --> 0:23:08.280 something changes. But you'd be shocked at the limits of 0:23:08.359 --> 0:23:12.840 our observation and our memory. Take a simple experiment run 0:23:12.880 --> 0:23:17.040 by the psychologists Raymond Nickerson and Marilyn Adams in nineteen 0:23:17.119 --> 0:23:20.280 seventy nine. What they did is they showed people fifteen 0:23:20.520 --> 0:23:24.840 different drawings of the US penny, and one was the 0:23:25.200 --> 0:23:29.640 actual penny, while the other drawings were manipulated versions where 0:23:29.640 --> 0:23:32.199 you had the date in a different place, or you 0:23:32.240 --> 0:23:37.280 had the slogan changed, for example, United States of America 0:23:37.359 --> 0:23:41.359 instead of in God we Trust. And fewer than half 0:23:41.440 --> 0:23:46.080 the participants could identify which one was the correct penny. 0:23:46.520 --> 0:23:50.920 Although the participants handled pennies almost every day of their lives, 0:23:51.400 --> 0:23:55.040 they hadn't paid attention in the way they thought they had. 0:23:55.800 --> 0:23:59.960 Experience doesn't translate into expertise. 0:24:01.080 --> 0:24:01.439 Why not. 0:24:01.720 --> 0:24:05.359 It's because we only see what we closely attend to. 0:24:05.960 --> 0:24:09.000 So imagine you catch a glimpse of a man and 0:24:09.040 --> 0:24:11.600 a woman having a picnic in a park, all you 0:24:11.800 --> 0:24:16.000 actually see is something like man woman food. With time, 0:24:16.720 --> 0:24:20.160 as you crawl the scene with your attentional systems, you'll 0:24:20.200 --> 0:24:21.639 incorporate more details. 0:24:21.680 --> 0:24:22.440 Oh they've got. 0:24:22.240 --> 0:24:25.679 Bowls of soup. It's a checkered blanket. That's a little unusual. 0:24:26.160 --> 0:24:29.280 There are hills in the background. And with more time 0:24:29.359 --> 0:24:33.040 you'll notice, Oh, his spoon is slightly bent, or she's 0:24:33.040 --> 0:24:38.080 wearing a silver necklace, or the encircling trees are buckthorn. 0:24:38.640 --> 0:24:42.760 But we incorporate more detail into a scene only if 0:24:42.760 --> 0:24:46.399 we pay more attention and we ask more questions. And 0:24:46.440 --> 0:24:49.080 as I said before, it's like this with all of 0:24:49.080 --> 0:24:52.879 our senses. What is the feeling of your pants on 0:24:52.920 --> 0:24:54.520 your left knee cap right now? 0:24:55.080 --> 0:24:57.600 Or the feeling of your shirt on your shoulders. 0:24:58.160 --> 0:25:01.400 As soon as you ask these questions, you can become 0:25:01.480 --> 0:25:04.920 aware of the answer, but you didn't know until you queried. 0:25:05.400 --> 0:25:08.760 So our visual sense is not like a camera that's 0:25:08.880 --> 0:25:12.520 taking in the complete scene. Instead, we see only the 0:25:12.560 --> 0:25:14.679 details that we go out to seek. And this is 0:25:14.720 --> 0:25:18.960 the issue with the banknotes. We may glance at one 0:25:19.520 --> 0:25:23.520 to verify that it matches the general template we're expecting, 0:25:24.040 --> 0:25:28.639 but we don't scrutinize the details. And that's why the 0:25:28.680 --> 0:25:34.119 European Central banks campaign to stop and pay attention to money. 0:25:34.600 --> 0:25:35.600 It just didn't work. 0:25:36.359 --> 0:25:41.120 We don't attend to things that sufficiently fit our assumptions. 0:25:41.560 --> 0:25:44.960