1 00:00:00,040 --> 00:00:02,460 The following content is provided under a Creative 2 00:00:02,460 --> 00:00:03,870 Commons license. 3 00:00:03,870 --> 00:00:06,910 Your support will help MIT OpenCourseWare continue to 4 00:00:06,910 --> 00:00:10,560 offer high-quality educational resources for free. 5 00:00:10,560 --> 00:00:13,460 To make a donation, or view additional materials from 6 00:00:13,460 --> 00:00:19,290 hundreds of MIT courses, visit MIT OpenCourseWare at 7 00:00:19,290 --> 00:00:20,540 ocw.mit.edu. 8 00:00:28,410 --> 00:00:30,930 PROFESSOR: So today we're going to talk about attention. 9 00:00:30,930 --> 00:00:36,870 And sometimes we talk about patients who are very unusual. 10 00:00:36,870 --> 00:00:38,350 In the next lecture, we will do that. 11 00:00:38,350 --> 00:00:42,210 We'll talk about spectacular disorders of attention. 12 00:00:42,210 --> 00:00:44,650 But today we're going to talk about attention as it 13 00:00:44,650 --> 00:00:47,910 typically occurs in you sitting here right now. 14 00:00:47,910 --> 00:00:50,440 I want to convince you that in addition to slips of attention 15 00:00:50,440 --> 00:00:55,520 that can occur as you get older, also how you attend to 16 00:00:55,520 --> 00:00:59,180 the world around you is weirder than you think. 17 00:00:59,180 --> 00:01:01,740 In a broad sense, I hope to convince you that it's 18 00:01:01,740 --> 00:01:02,760 happening all the time. 19 00:01:02,760 --> 00:01:05,269 Right now, as you go back to your room, as you walk to the 20 00:01:05,269 --> 00:01:06,500 next class, or your room. 21 00:01:06,500 --> 00:01:09,540 How you attend to what's around you in the world is 22 00:01:09,540 --> 00:01:11,730 weirder than you think. 23 00:01:11,730 --> 00:01:12,760 So what is attention? 24 00:01:12,760 --> 00:01:14,670 What do we mean by attending to things? 25 00:01:14,670 --> 00:01:17,340 So William James was a psychologist at Harvard who, 26 00:01:17,340 --> 00:01:21,160 in many ways, articulated concepts in psychology that, 27 00:01:21,160 --> 00:01:23,950 till this day, turned out to be beautiful ways to talk 28 00:01:23,950 --> 00:01:26,200 about them in an everyday sense. 29 00:01:26,200 --> 00:01:28,060 "Everybody knows what attention is. 30 00:01:28,060 --> 00:01:30,960 It's the taking possession by the mind in clear and vivid 31 00:01:30,960 --> 00:01:34,760 form of one out of what seems like several simultaneous 32 00:01:34,760 --> 00:01:36,530 objects or trains of thought. 33 00:01:36,530 --> 00:01:38,720 It applies withdrawal of some things in order to deal with 34 00:01:38,720 --> 00:01:38,940 others." 35 00:01:38,940 --> 00:01:42,090 So a big idea there is we can pay attention to some things 36 00:01:42,090 --> 00:01:43,820 and there's a lot of things around us. 37 00:01:43,820 --> 00:01:46,930 And so we have to pick where we focus our attention. 38 00:01:46,930 --> 00:01:48,650 A little bit like if you have a budget. 39 00:01:48,650 --> 00:01:50,820 If you have some money, what do you spend your money on, 40 00:01:50,820 --> 00:01:53,400 and what do not spend your money on? 41 00:01:53,400 --> 00:01:54,730 So we're going to talk about attention. 42 00:01:54,730 --> 00:01:56,540 It's kind of similar to the idea of awareness. 43 00:01:56,540 --> 00:01:57,130 What are you aware of? 44 00:01:57,130 --> 00:01:59,020 Or what are you conscious of? 45 00:01:59,020 --> 00:02:02,760 And two themes across the talk will be we often attend to 46 00:02:02,760 --> 00:02:04,050 more than we realize. 47 00:02:04,050 --> 00:02:07,310 And I'll show you very specific evidence for that. 48 00:02:07,310 --> 00:02:10,060 But we often attend to things less than we 49 00:02:10,060 --> 00:02:13,350 realize all the time. 50 00:02:13,350 --> 00:02:14,480 And here's roughly the outline. 51 00:02:14,480 --> 00:02:16,670 We're saying attention, what you pay attention to, is kind 52 00:02:16,670 --> 00:02:20,190 of a gatekeeper to what you perceive and what you know. 53 00:02:20,190 --> 00:02:21,380 Attention is very limited. 54 00:02:21,380 --> 00:02:23,610 We can only process one thing at a time. 55 00:02:23,610 --> 00:02:26,940 And because of that, we miss a lot around us all the time. 56 00:02:26,940 --> 00:02:28,970 There are unconscious processes that pick up 57 00:02:28,970 --> 00:02:29,560 information. 58 00:02:29,560 --> 00:02:32,870 So sometimes things that we don't notice do influence us. 59 00:02:32,870 --> 00:02:34,840 And attention can be viewed as kind of a 60 00:02:34,840 --> 00:02:36,970 dialogue between two things. 61 00:02:36,970 --> 00:02:39,390 Between things in the world that capture your attention. 62 00:02:39,390 --> 00:02:41,840 If somebody throws a ball at you, something dramatic 63 00:02:41,840 --> 00:02:44,540 happens-- you hear a scary sound that captures your 64 00:02:44,540 --> 00:02:45,190 attention-- 65 00:02:45,190 --> 00:02:47,220 we talk about that as "bottom-up" processing. 66 00:02:47,220 --> 00:02:48,610 It impinges on you. 67 00:02:48,610 --> 00:02:50,210 It forces you to attend to it. 68 00:02:50,210 --> 00:02:53,600 Because it's danger, it's moving, it's unusual. 69 00:02:53,600 --> 00:02:54,830 And then, "top-down" things. 70 00:02:54,830 --> 00:02:55,680 What's your goal? 71 00:02:55,680 --> 00:02:57,310 What are you paying attention to? 72 00:02:57,310 --> 00:02:57,980 What do you care about? 73 00:02:57,980 --> 00:02:58,860 What are you up to? 74 00:02:58,860 --> 00:03:00,815 People call that top-down processing. 75 00:03:00,815 --> 00:03:03,520 So things that the environment drives into your attention 76 00:03:03,520 --> 00:03:05,440 because they're so dramatic, bottom-up. 77 00:03:05,440 --> 00:03:07,460 Things that you're looking for, thinking about, you know 78 00:03:07,460 --> 00:03:09,210 are useful for your purposes, top-down. 79 00:03:11,880 --> 00:03:16,930 There's tremendous evidence that we encode a very small 80 00:03:16,930 --> 00:03:21,710 fraction of what's around us in a very specific way. 81 00:03:21,710 --> 00:03:24,760 Estimate with me how often a day, on average, since you've 82 00:03:24,760 --> 00:03:26,770 been about five or six or seven and you've been allowed 83 00:03:26,770 --> 00:03:29,160 to have any money in your possession at all. 84 00:03:29,160 --> 00:03:33,290 Do you think you've held or seen a penny? 85 00:03:33,290 --> 00:03:35,540 OK, so we all know the horror when you're 86 00:03:35,540 --> 00:03:36,910 handed $0.4 in change. 87 00:03:36,910 --> 00:03:39,100 That's 4 pennies right there. 88 00:03:39,100 --> 00:03:42,540 So we'll make an estimate. 89 00:03:42,540 --> 00:03:47,230 Pretend from age 5 to 20 you've been handling change. 90 00:03:47,230 --> 00:03:49,470 So maybe you've encountered 3 pennies a day. 91 00:03:49,470 --> 00:03:51,330 OK, that's reasonable? 92 00:03:51,330 --> 00:03:52,690 OK. 93 00:03:52,690 --> 00:03:54,780 We don't have to be just in the right ballpark. 94 00:03:54,780 --> 00:03:58,290 There's 365 days a year, skip 15 years. 95 00:03:58,290 --> 00:03:59,670 Just do the multiplication. 96 00:03:59,670 --> 00:04:05,770 You end up with about 16,380 penny experiences. 97 00:04:05,770 --> 00:04:08,710 All right, so we all know that if you head towards midterms, 98 00:04:08,710 --> 00:04:11,320 it's going to be hard to study a lot of stuff quickly. 99 00:04:11,320 --> 00:04:13,390 But if somebody said to you, if you study something 100 00:04:13,390 --> 00:04:18,589 approximately 17,000 times, you'd think, well, that one 101 00:04:18,589 --> 00:04:20,350 I'm going to get. 102 00:04:20,350 --> 00:04:22,930 So let's ask the question. 103 00:04:22,930 --> 00:04:25,590 Does the Lincoln penny face to the left or the right? 104 00:04:25,590 --> 00:04:28,360 Is there anything above his head or below his head? 105 00:04:28,360 --> 00:04:30,560 Is there anything on the penny to the left or to the right? 106 00:04:30,560 --> 00:04:32,600 And how about the other side? 107 00:04:32,600 --> 00:04:36,320 And how many of you with approximately 17,000, or 108 00:04:36,320 --> 00:04:39,800 certainly many thousands of visual experiences with a 109 00:04:39,800 --> 00:04:43,820 penny would bet something that means a lot to you on the 110 00:04:43,820 --> 00:04:46,000 correctness of your answer to these 5 fundamental questions? 111 00:04:50,480 --> 00:04:51,730 This is good teamwork. 112 00:04:55,260 --> 00:04:56,340 Let's take it multiple choice. 113 00:04:56,340 --> 00:04:57,590 That's better, right? 114 00:05:02,767 --> 00:05:04,380 AUDIENCE: It faces the right. 115 00:05:04,380 --> 00:05:05,860 PROFESSOR: You're making me feel better because 116 00:05:05,860 --> 00:05:06,810 of the fouls up. 117 00:05:06,810 --> 00:05:08,330 And I appreciate that. 118 00:05:08,330 --> 00:05:11,170 This is actually what it looks like. 119 00:05:11,170 --> 00:05:14,490 So what do you think it means that 17,000 exposures to a 120 00:05:14,490 --> 00:05:19,110 penny don't leave in you a reasonably accurate memory of 121 00:05:19,110 --> 00:05:23,190 the main things you see in a penny? 122 00:05:23,190 --> 00:05:26,390 How could that happen? 123 00:05:26,390 --> 00:05:27,670 And it happens to everybody. 124 00:05:27,670 --> 00:05:28,730 In fact, when they do experiments, 125 00:05:28,730 --> 00:05:30,540 here's people's drawings. 126 00:05:30,540 --> 00:05:34,190 Don't feel bad if you didn't know it. 127 00:05:34,190 --> 00:05:36,590 Half the people don't even think that the version that 128 00:05:36,590 --> 00:05:38,970 you show them, if you ask them is plausibly correct, you show 129 00:05:38,970 --> 00:05:41,830 the correct answer, they go no. 130 00:05:41,830 --> 00:05:43,910 That can't be right, OK? 131 00:05:43,910 --> 00:05:45,130 So here's one thing. 132 00:05:45,130 --> 00:05:48,450 If we don't attend to something very specifically, 133 00:05:48,450 --> 00:05:52,700 we can see it a tremendous number of times and it leaves 134 00:05:52,700 --> 00:05:57,320 no impression on our mind and no memory of it. 135 00:05:57,320 --> 00:06:01,160 Now, why is that in an everyday sense? 136 00:06:01,160 --> 00:06:02,400 We don't know on a deep science sense. 137 00:06:02,400 --> 00:06:03,540 But what do you think the point of it is? 138 00:06:03,540 --> 00:06:06,870 Why don't you know what a penny looks like? 139 00:06:06,870 --> 00:06:08,000 Because it's not important somebody said. 140 00:06:08,000 --> 00:06:08,540 Yeah. 141 00:06:08,540 --> 00:06:13,870 Because it's close enough to see that brown smallish thing, 142 00:06:13,870 --> 00:06:16,010 and you don't need the rest of any information to hand the 143 00:06:16,010 --> 00:06:17,680 money back and forth. 144 00:06:17,680 --> 00:06:21,060 So what you attend to, what you decide is important, makes 145 00:06:21,060 --> 00:06:22,610 incredible difference. 146 00:06:22,610 --> 00:06:24,350 And nothing else makes up for it. 147 00:06:24,350 --> 00:06:31,400 17,000 exposures don't make up for what a penny looks like. 148 00:06:42,400 --> 00:06:46,830 So, it was research at MIT that used this method of 149 00:06:46,830 --> 00:06:49,390 simultaneously processing two messages in the two ears, the 150 00:06:49,390 --> 00:06:52,490 shadowed message, the one you're copying, and the second 151 00:06:52,490 --> 00:06:54,920 message, the one that you try to pick up, but 152 00:06:54,920 --> 00:06:55,960 it's hard to do. 153 00:06:55,960 --> 00:06:58,760 So they ask in the ear that you're not paying attention 154 00:06:58,760 --> 00:07:01,080 to, what can you pick up? 155 00:07:01,080 --> 00:07:05,570 Here is what you can pick up if a voice is present. 156 00:07:05,570 --> 00:07:08,540 Now you just imagine silence versus somebody talking. 157 00:07:08,540 --> 00:07:09,940 That you can notice. 158 00:07:09,940 --> 00:07:12,420 If the voice changes from a man to a woman-- 159 00:07:12,420 --> 00:07:15,300 if it changes from a man to woman, you can pick that up in 160 00:07:15,300 --> 00:07:16,860 the unattended ear, the one you're not 161 00:07:16,860 --> 00:07:17,780 paying attention to. 162 00:07:17,780 --> 00:07:19,585 Or if the voice goes-- a person's talking and, all of a 163 00:07:19,585 --> 00:07:20,880 sudden, it goes er. 164 00:07:20,880 --> 00:07:22,490 You notice that. 165 00:07:22,490 --> 00:07:24,660 Here's what you don't notice. 166 00:07:24,660 --> 00:07:26,010 Anything about the content. 167 00:07:26,010 --> 00:07:28,190 I mean, that's what you experience. 168 00:07:28,190 --> 00:07:30,745 You can switch the language of the message. 169 00:07:30,745 --> 00:07:33,480 You can go from English to Spanish, or Italian. 170 00:07:33,480 --> 00:07:35,215 A person won't notice that it's changed. 171 00:07:39,730 --> 00:07:42,860 You can also change whether it's typical speech or 172 00:07:42,860 --> 00:07:45,710 backward nonsense speech, which sounds very weird. 173 00:07:45,710 --> 00:07:47,220 You don't notice that either. 174 00:07:47,220 --> 00:07:49,125 So you notice a few things. 175 00:07:49,125 --> 00:07:50,380 But there's a lot of other things you 176 00:07:50,380 --> 00:07:52,370 don't notice at all. 177 00:07:52,370 --> 00:07:56,350 Now, you may have all had the experience of what's called 178 00:07:56,350 --> 00:07:57,950 the cocktail party phenomenon. 179 00:07:57,950 --> 00:07:59,400 Do you know what that is? 180 00:07:59,400 --> 00:08:00,760 Have you ever been talking with 181 00:08:00,760 --> 00:08:01,710 people, you're at a party. 182 00:08:01,710 --> 00:08:03,650 You're at some events, or something like that. 183 00:08:03,650 --> 00:08:06,910 And somebody mentions your name. 184 00:08:06,910 --> 00:08:09,240 Especially your first name, but your name. 185 00:08:09,240 --> 00:08:10,450 Maybe your last name works, too. 186 00:08:10,450 --> 00:08:11,530 And then all of a sudden, you notice a 187 00:08:11,530 --> 00:08:12,710 conversation around you. 188 00:08:12,710 --> 00:08:14,650 It's almost hard to pay attention to the person you're 189 00:08:14,650 --> 00:08:16,880 talking to because the person over there has just said your 190 00:08:16,880 --> 00:08:18,880 name and you wonder what they're saying about you. 191 00:08:18,880 --> 00:08:21,170 Now, here's a really interesting paradox. 192 00:08:21,170 --> 00:08:23,100 How did you not notice that other 193 00:08:23,100 --> 00:08:24,920 conversation, which you didn't? 194 00:08:24,920 --> 00:08:26,270 That's like the shadowing task. 195 00:08:26,270 --> 00:08:28,890 I'm talking to one person, there's some hubbub over here. 196 00:08:28,890 --> 00:08:31,020 I don't know what they're talking about. 197 00:08:31,020 --> 00:08:34,400 All a sudden, they say my name and I notice what they're 198 00:08:34,400 --> 00:08:34,820 talking about. 199 00:08:34,820 --> 00:08:35,690 Well, how did that happen? 200 00:08:35,690 --> 00:08:36,919 I wasn't paying attention to it. 201 00:08:36,919 --> 00:08:38,390 I don't know anything about it. 202 00:08:38,390 --> 00:08:41,799 But my name pounced out in my mind and drew my 203 00:08:41,799 --> 00:08:42,630 attention to it. 204 00:08:42,630 --> 00:08:44,230 So it's kind of a paradox. 205 00:08:44,230 --> 00:08:47,220 I'm not paying attention, but somehow I pick up my name. 206 00:08:47,220 --> 00:08:49,750 And so we'll talk a little bit more about this. 207 00:08:49,750 --> 00:08:52,640 But people have looked at this very experimentally. 208 00:08:52,640 --> 00:08:54,400 So here's how they approached it. 209 00:08:54,400 --> 00:08:56,960 They would, again, hear a sentence in one ear, "The man 210 00:08:56,960 --> 00:08:59,590 approached the bank." That's the one they're repeating. 211 00:08:59,590 --> 00:09:02,330 And in the other ear, they would hear one of two 212 00:09:02,330 --> 00:09:03,010 different words. 213 00:09:03,010 --> 00:09:05,830 In the unattended ear, they're not getting much out of that. 214 00:09:05,830 --> 00:09:08,590 They would hear the word "money" or "river." So now the 215 00:09:08,590 --> 00:09:11,190 bank could be interpreted as you went to the bank to 216 00:09:11,190 --> 00:09:13,580 deposit your paycheck, or you want to the bank to have a 217 00:09:13,580 --> 00:09:14,720 picnic by the river. 218 00:09:14,720 --> 00:09:18,980 Two different meanings of the word "bank" are suggested by 219 00:09:18,980 --> 00:09:21,900 the word you hear in the ear that you're not processing 220 00:09:21,900 --> 00:09:22,620 information in. 221 00:09:22,620 --> 00:09:24,420 And then the people tell you, well, tell me what the 222 00:09:24,420 --> 00:09:25,740 sentence is you just heard, kind of. 223 00:09:25,740 --> 00:09:29,390 And people are primed to give the interpretation of the 224 00:09:29,390 --> 00:09:32,480 sentence that goes with the word that's suggested in the 225 00:09:32,480 --> 00:09:33,940 ear they weren't listening to. 226 00:09:33,940 --> 00:09:37,630 So something in their mind and brain heard enough of that to 227 00:09:37,630 --> 00:09:39,700 influence what they're interpreting in 228 00:09:39,700 --> 00:09:41,220 the attended ear. 229 00:09:41,220 --> 00:09:42,580 Here's another example. 230 00:09:42,580 --> 00:09:44,650 They're shadowing, they're paying attention and repeating 231 00:09:44,650 --> 00:09:47,630 aloud an essay in one ear. 232 00:09:47,630 --> 00:09:51,810 And in the other ear, they hear something that's like a 233 00:09:51,810 --> 00:09:55,090 word "taxi fare." And they pick words that are spelled 234 00:09:55,090 --> 00:09:57,590 two different ways, homonyms. 235 00:09:57,590 --> 00:09:59,750 So the word F-A-R-E or F-A-I-R. 236 00:09:59,750 --> 00:10:01,240 Now, if you were typically-- 237 00:10:01,240 --> 00:10:04,160 if I ask you to write down the word "fair" and spell it 238 00:10:04,160 --> 00:10:05,620 correctly, you might say, well, which one do you want or 239 00:10:05,620 --> 00:10:06,460 something like that? 240 00:10:06,460 --> 00:10:08,140 But if I didn't tell you anything, most people would 241 00:10:08,140 --> 00:10:10,090 put down the word F-A-I-R most of the time. 242 00:10:10,090 --> 00:10:12,570 It's a more common word. 243 00:10:12,570 --> 00:10:14,120 But what they found is this. 244 00:10:14,120 --> 00:10:15,045 They finished the experiment. 245 00:10:15,045 --> 00:10:16,510 They shadowed one thing. 246 00:10:16,510 --> 00:10:19,430 They had a word like "taxi fare" in the unattended ear. 247 00:10:19,430 --> 00:10:21,800 And they said, well, did you notice you got that word? 248 00:10:21,800 --> 00:10:25,230 Well, if you listen to it, you remembered you heard "taxi 249 00:10:25,230 --> 00:10:28,840 fare." But if you were shadowed, you pretty much-- 250 00:10:28,840 --> 00:10:30,650 you very rarely remembered that you heard it. 251 00:10:30,650 --> 00:10:33,050 So it's not consciously available to you that you 252 00:10:33,050 --> 00:10:35,500 heard it because you were so busy in the other ear. 253 00:10:35,500 --> 00:10:37,170 And in your mind, listening to that ear. 254 00:10:37,170 --> 00:10:40,230 But if you're asked to spell the word, how often do you 255 00:10:40,230 --> 00:10:43,520 come up with the rare spelling because of what you 256 00:10:43,520 --> 00:10:44,600 heard in your ear? 257 00:10:44,600 --> 00:10:46,740 It's exactly the same. 258 00:10:46,740 --> 00:10:51,140 So part of your mind heard that word, thought about that 259 00:10:51,140 --> 00:10:53,900 phrase "taxi fare." And when somebody says to you write 260 00:10:53,900 --> 00:10:58,300 down the word "fair," your mind got moved to think about 261 00:10:58,300 --> 00:10:59,095 the alternative spellings. 262 00:10:59,095 --> 00:11:02,580 Your unconscious processing of that content has influenced 263 00:11:02,580 --> 00:11:04,870 you in experimentally verifiable way. 264 00:11:04,870 --> 00:11:06,880 So what this shows, if you go back to the cocktail party 265 00:11:06,880 --> 00:11:09,500 effect, which is the everyday kind of version of this, is 266 00:11:09,500 --> 00:11:13,040 that weirdly enough some kinds of things that we don't pay 267 00:11:13,040 --> 00:11:15,450 attention to have such a property that in a 268 00:11:15,450 --> 00:11:18,910 machine-like reflexive way our mind brain process it to some 269 00:11:18,910 --> 00:11:20,650 extent anyway. 270 00:11:20,650 --> 00:11:22,250 And then there's something curious that we don't 271 00:11:22,250 --> 00:11:25,880 understand deeply, which is somehow if it's your name, 272 00:11:25,880 --> 00:11:27,260 it's going like, I'm hearing my name. 273 00:11:27,260 --> 00:11:27,990 I'm hearing my name. 274 00:11:27,990 --> 00:11:29,580 Pay attention over here. 275 00:11:29,580 --> 00:11:32,640 That's a deep psychological, computational theory. 276 00:11:32,640 --> 00:11:35,020 And somehow you pull it out because it's triggered your 277 00:11:35,020 --> 00:11:35,620 mind enough. 278 00:11:35,620 --> 00:11:38,200 Because our names are so important to us. 279 00:11:38,200 --> 00:11:38,420 Yeah? 280 00:11:38,420 --> 00:11:40,392 AUDIENCE: Is this similar to [INAUDIBLE]? 281 00:11:49,266 --> 00:11:50,530 PROFESSOR: OK, tell me again. 282 00:11:50,530 --> 00:11:52,790 So the phenomenon where if you notice a word-- go ahead. 283 00:11:52,790 --> 00:11:53,190 So tell me again. 284 00:11:53,190 --> 00:11:53,450 Sorry. 285 00:11:53,450 --> 00:11:54,842 AUDIENCE: Like you're just looking around and suddenly a 286 00:11:54,842 --> 00:11:58,285 specific word [INAUDIBLE]. 287 00:11:58,285 --> 00:11:59,220 PROFESSOR: Yeah. 288 00:11:59,220 --> 00:12:00,880 It's very much aligned to the same thing. 289 00:12:00,880 --> 00:12:03,900 The question was, something pops in your head-- a word-- 290 00:12:03,900 --> 00:12:05,190 and then you look back and you go, wait a minute. 291 00:12:05,190 --> 00:12:06,190 I just read that, but I wasn't even 292 00:12:06,190 --> 00:12:07,710 thinking about it or something. 293 00:12:07,710 --> 00:12:09,040 Somebody said it or it's on a sign. 294 00:12:09,040 --> 00:12:11,400 But yes. 295 00:12:11,400 --> 00:12:14,820 Some experience you're not paying much attention to tips 296 00:12:14,820 --> 00:12:16,080 your mind to think a certain way. 297 00:12:16,080 --> 00:12:17,580 And all of a sudden you're thinking that way and you 298 00:12:17,580 --> 00:12:19,090 don't know what tipped your mind that way. 299 00:12:19,090 --> 00:12:20,340 That's very similar, yeah. 300 00:12:30,590 --> 00:12:35,285 So again, the message here is, as you're walking around in 301 00:12:35,285 --> 00:12:37,850 the world, there's very few things that you're very 302 00:12:37,850 --> 00:12:39,510 specifically really aware of. 303 00:12:39,510 --> 00:12:44,700 And many things can change on you in these examples. 304 00:12:44,700 --> 00:12:48,380 So one of the reasons-- this is just a thought that by 305 00:12:48,380 --> 00:12:50,700 being selective in attention, you can focus on what you want 306 00:12:50,700 --> 00:12:51,310 to focus on. 307 00:12:51,310 --> 00:12:53,720 It's not only a weakness, it's a strength. 308 00:12:53,720 --> 00:12:55,620 So you could pick out the message in red or you could 309 00:12:55,620 --> 00:12:56,330 pick out the message in blue. 310 00:12:56,330 --> 00:12:58,590 Where you attend helps you select 311 00:12:58,590 --> 00:12:59,920 information that's relevant. 312 00:12:59,920 --> 00:13:01,390 But here's two more works. 313 00:13:01,390 --> 00:13:05,135 Almost all the examples I just showed you were from Dan 314 00:13:05,135 --> 00:13:07,420 Simons, who's done really beautiful work in this way. 315 00:13:07,420 --> 00:13:09,040 Made it very available to-- 316 00:13:09,040 --> 00:13:11,310 so here's two YouTube ones. 317 00:13:25,200 --> 00:13:29,880 Sometimes people have said, OK, these are cute laboratory 318 00:13:29,880 --> 00:13:31,960 things where you're having people counting. 319 00:13:31,960 --> 00:13:33,850 But the real world is different. 320 00:13:33,850 --> 00:13:36,100 When I'm out of the classroom and I'm out of the laboratory, 321 00:13:36,100 --> 00:13:38,490 the real world you kind of notice a lot more. 322 00:13:38,490 --> 00:13:40,660 Because it's real people, real situations. 323 00:13:40,660 --> 00:13:42,800 So this is one more example that's a 324 00:13:42,800 --> 00:13:44,050 beautiful example of this. 325 00:13:54,180 --> 00:13:56,810 So do you think it would have worked if the replacement 326 00:13:56,810 --> 00:13:58,440 person was very different? 327 00:13:58,440 --> 00:14:01,780 If we was a man or a woman, of if a very taller person versus 328 00:14:01,780 --> 00:14:02,900 a person not-- 329 00:14:02,900 --> 00:14:03,780 no. 330 00:14:03,780 --> 00:14:05,290 But what does that mean? 331 00:14:05,290 --> 00:14:08,260 Why does it work for this one? 332 00:14:08,260 --> 00:14:12,740 And why would it-- it means that in the person's head, 333 00:14:12,740 --> 00:14:17,330 they have a very rough picture of who they're talking to. 334 00:14:17,330 --> 00:14:18,210 Does that make sense? 335 00:14:18,210 --> 00:14:19,610 So there's a lot of room for replacement. 336 00:14:19,610 --> 00:14:22,140 Not total room, but a very rough picture. 337 00:14:22,140 --> 00:14:24,010 But it's right in front of them. 338 00:14:24,010 --> 00:14:26,590 OK, now, you've seen this thing, so I won't do it as a 339 00:14:26,590 --> 00:14:30,080 demonstration now, where people simply read the words, 340 00:14:30,080 --> 00:14:31,570 the red, orange, brown. 341 00:14:31,570 --> 00:14:34,930 And that's easy because you're a very automatic reader. 342 00:14:34,930 --> 00:14:38,670 If you're a typical adult, you read about 250 words a minute. 343 00:14:38,670 --> 00:14:40,810 Zoom, zoom, zoom. 344 00:14:40,810 --> 00:14:43,090 But if you have to name the color of the print when the 345 00:14:43,090 --> 00:14:46,430 print is contradicted by the word, so you're reading the 346 00:14:46,430 --> 00:14:52,160 word red printed in green, but your job is to say green. 347 00:14:52,160 --> 00:14:55,040 That naming of the print when the word is not matching the 348 00:14:55,040 --> 00:14:57,040 color of the print slows people down. 349 00:14:57,040 --> 00:14:57,900 They make mistakes. 350 00:14:57,900 --> 00:14:58,780 They're slower. 351 00:14:58,780 --> 00:15:01,240 And that's because there's a big lesson in this. 352 00:15:01,240 --> 00:15:02,980 It's great to get automatic at things. 353 00:15:02,980 --> 00:15:05,690 That's how we're skilled at things-- speaking, driving, 354 00:15:05,690 --> 00:15:06,910 everything physical. 355 00:15:06,910 --> 00:15:09,530 We don't even think about how we walk. 356 00:15:09,530 --> 00:15:12,900 The price of automaticity being really skilled and 357 00:15:12,900 --> 00:15:15,170 reflexive is you lose control. 358 00:15:15,170 --> 00:15:16,230 You lose control. 359 00:15:16,230 --> 00:15:18,935 So you don't want to think about what that word says, but 360 00:15:18,935 --> 00:15:19,560 you can't help but. 361 00:15:19,560 --> 00:15:22,300 Now, let me touch one more thing. 362 00:15:22,300 --> 00:15:25,270 How many people think-- and just be honest in your seat-- 363 00:15:25,270 --> 00:15:28,882 that hypnosis is kind of bologna? 364 00:15:28,882 --> 00:15:29,990 OK. 365 00:15:29,990 --> 00:15:30,410 All right, a lot of you. 366 00:15:30,410 --> 00:15:31,420 This is MIT, right? 367 00:15:31,420 --> 00:15:32,040 OK. 368 00:15:32,040 --> 00:15:35,250 So I'm going to tell you that I think a lot of hypnosis, the 369 00:15:35,250 --> 00:15:38,030 kinds of things where if I learn how to hypnotize people, 370 00:15:38,030 --> 00:15:42,740 then they can all vote for me, or something like that, that 371 00:15:42,740 --> 00:15:43,900 doesn't work. 372 00:15:43,900 --> 00:15:46,770 And people who are serious about hypnosis will tell you 373 00:15:46,770 --> 00:15:49,710 that hypnosis is in the person, it's not in the 374 00:15:49,710 --> 00:15:50,490 hypnotizer. 375 00:15:50,490 --> 00:15:52,840 It's in the person, it's not in the hypnotizer. 376 00:15:52,840 --> 00:15:56,370 And it turns out that, in ways that are not well understood, 377 00:15:56,370 --> 00:15:58,970 there are some people who are highly hypnotizeable and some 378 00:15:58,970 --> 00:16:00,590 people who are not. 379 00:16:00,590 --> 00:16:02,290 So this is why I think we're suspicious. 380 00:16:02,290 --> 00:16:04,450 Because if we're not, we're going, well, they must be 381 00:16:04,450 --> 00:16:06,160 faking it, or pretending, or whatever. 382 00:16:06,160 --> 00:16:08,170 But let me tell you what they did and let me see if this 383 00:16:08,170 --> 00:16:10,560 convinces you at all as an experiment. 384 00:16:10,560 --> 00:16:15,180 So they told people sometimes to read those words where it's 385 00:16:15,180 --> 00:16:18,800 confusing because you're reading the color of the print 386 00:16:18,800 --> 00:16:21,180 and the word is contradictory, it slows you down. 387 00:16:21,180 --> 00:16:25,090 They told them, imagine those characters are characters in a 388 00:16:25,090 --> 00:16:28,840 foreign language that you do not know. 389 00:16:28,840 --> 00:16:30,850 Would the content of the word mess you up for the 390 00:16:30,850 --> 00:16:32,380 color of the print? 391 00:16:32,380 --> 00:16:33,670 No. 392 00:16:33,670 --> 00:16:36,260 Because if it's a language you don't know, it's just stuff 393 00:16:36,260 --> 00:16:37,120 and you can just name-- 394 00:16:37,120 --> 00:16:38,130 OK? 395 00:16:38,130 --> 00:16:39,280 And they asked them to imagine that. 396 00:16:39,280 --> 00:16:40,020 Now, here's what happened. 397 00:16:40,020 --> 00:16:42,060 For people who are low hypnotizability, you can give 398 00:16:42,060 --> 00:16:44,270 people questionnaires and they're pretty good at these 399 00:16:44,270 --> 00:16:45,520 kinds of measures. 400 00:16:47,460 --> 00:16:52,240 They're slowed down by about a tenth of a second per word 401 00:16:52,240 --> 00:16:54,130 when the information is contradictory. 402 00:16:54,130 --> 00:16:55,740 And when they're asked to imagine it's a foreign 403 00:16:55,740 --> 00:16:57,410 language, it's about the same. 404 00:16:57,410 --> 00:17:00,000 The low hypnotizables, you can tell them, imagine it's a 405 00:17:00,000 --> 00:17:01,750 print you don't know. 406 00:17:01,750 --> 00:17:03,710 It doesn't matter. 407 00:17:03,710 --> 00:17:04,890 Here's the high hypnotizable. 408 00:17:04,890 --> 00:17:06,750 Also slowed down. 409 00:17:06,750 --> 00:17:09,050 But when they imagine it's a print in a foreign language 410 00:17:09,050 --> 00:17:11,750 they don't know, not slowed down at all. 411 00:17:14,380 --> 00:17:15,800 So that's behavior. 412 00:17:15,800 --> 00:17:18,119 That's evidence. 413 00:17:18,119 --> 00:17:20,630 And we don't think you can fake it because how 414 00:17:20,630 --> 00:17:21,220 would you fake it? 415 00:17:21,220 --> 00:17:24,170 But here, in case you were still worried about that, 416 00:17:24,170 --> 00:17:27,109 here's the brain imaging evidence from just the people 417 00:17:27,109 --> 00:17:30,880 with high hypnosis, the high hypnotizable people. 418 00:17:30,880 --> 00:17:33,330 What you see here turned on is the part-- 419 00:17:33,330 --> 00:17:35,290 our brain areas are turned on when you have to do the 420 00:17:35,290 --> 00:17:37,530 difficult task where it's contradictory 421 00:17:37,530 --> 00:17:39,510 versus the easy task. 422 00:17:39,510 --> 00:17:41,855 All this area you see turned on in the anterior cingulate, 423 00:17:41,855 --> 00:17:43,250 and especially-- 424 00:17:43,250 --> 00:17:45,300 which is an area that we know is involved when things are 425 00:17:45,300 --> 00:17:48,750 contradictory or paradoxical and people are sorting it out. 426 00:17:48,750 --> 00:17:50,970 Look what happens to that when these individuals are 427 00:17:50,970 --> 00:17:53,050 imagining it's a foreign text. 428 00:17:53,050 --> 00:17:55,740 It's pretty much gone. 429 00:17:55,740 --> 00:17:57,900 Is that OK? 430 00:17:57,900 --> 00:17:59,730 I'm as skeptical as anybody about behavioral 431 00:17:59,730 --> 00:18:00,870 claims, I really am. 432 00:18:00,870 --> 00:18:04,400 I think hypnosis is real for some people and very powerful 433 00:18:04,400 --> 00:18:06,390 for some people. 434 00:18:06,390 --> 00:18:07,850 And for other people it doesn't work at all. 435 00:18:07,850 --> 00:18:11,920 I happen to be in the low hypnotizable category. 436 00:18:11,920 --> 00:18:14,170 I'm as skeptical as anybody, but I think there are people 437 00:18:14,170 --> 00:18:17,800 who can do remarkable things by hypnotic suggestion. 438 00:18:17,800 --> 00:18:20,900 But it's in them, it's not in the-- a person who hypnotizes 439 00:18:20,900 --> 00:18:24,160 you, so to speak, can help you a little bit. 440 00:18:24,160 --> 00:18:28,020 But it's in you whether you have that approach to control 441 00:18:28,020 --> 00:18:29,270 of your thoughts or not. 442 00:18:32,220 --> 00:18:33,760 So let's go back to a test. 443 00:18:33,760 --> 00:18:36,520 So we're talking, again, about two ideas. 444 00:18:36,520 --> 00:18:37,270 That there's things-- 445 00:18:37,270 --> 00:18:40,470 top- down and bottom-up processes, and different ways 446 00:18:40,470 --> 00:18:42,010 that you can attend to the world. 447 00:18:42,010 --> 00:18:44,980 So I'm going to ask you to consult your own intuitions 448 00:18:44,980 --> 00:18:46,570 for the following thing. 449 00:18:46,570 --> 00:18:50,770 I'm going to show you things displayed on the monitor. 450 00:18:50,770 --> 00:18:52,650 And your job is simply to answer in your own mind as 451 00:18:52,650 --> 00:18:54,470 quickly as you possibly can. 452 00:18:54,470 --> 00:18:56,080 Maybe just, if you're willing to do it out loud, let's just 453 00:18:56,080 --> 00:18:58,370 say yes or no aloud, if you're willing to do it. 454 00:18:58,370 --> 00:19:00,790 As quickly as you, whether you see a red X. Sometimes there's 455 00:19:00,790 --> 00:19:01,760 a red X, sometimes not. 456 00:19:01,760 --> 00:19:04,250 So ready? 457 00:19:04,250 --> 00:19:07,610 Sorry, that's my get ready. 458 00:19:07,610 --> 00:19:07,885 AUDIENCE: No. 459 00:19:07,885 --> 00:19:09,370 PROFESSOR: No, great. 460 00:19:09,370 --> 00:19:10,860 See, MIT is easy. 461 00:19:10,860 --> 00:19:12,051 Sorry. 462 00:19:12,051 --> 00:19:12,920 OK, ready? 463 00:19:12,920 --> 00:19:14,046 Red X or not? 464 00:19:14,046 --> 00:19:14,720 AUDIENCE: Yes. 465 00:19:14,720 --> 00:19:15,510 PROFESSOR: Great. 466 00:19:15,510 --> 00:19:16,950 Ready? 467 00:19:16,950 --> 00:19:18,380 OK. 468 00:19:18,380 --> 00:19:19,990 Thank you. 469 00:19:19,990 --> 00:19:20,350 Ready? 470 00:19:20,350 --> 00:19:21,483 Red X or not? 471 00:19:21,483 --> 00:19:22,745 AUDIENCE: Yes. 472 00:19:22,745 --> 00:19:23,200 PROFESSOR: Whoa. 473 00:19:23,200 --> 00:19:24,050 You guys are amazing. 474 00:19:24,050 --> 00:19:26,660 Now, look how many places you had to look to find it. 475 00:19:26,660 --> 00:19:28,170 But did it feel that way, or did it seem to, what they 476 00:19:28,170 --> 00:19:28,900 call, pop out? 477 00:19:28,900 --> 00:19:29,530 It's just obvious. 478 00:19:29,530 --> 00:19:30,840 It's just there. 479 00:19:30,840 --> 00:19:33,430 So imagine I gave you this task. 480 00:19:33,430 --> 00:19:37,580 I said, go into this room and please get a particular book. 481 00:19:37,580 --> 00:19:39,610 And you go into the room and there's one book there. 482 00:19:39,610 --> 00:19:40,430 And you've picked it up. 483 00:19:40,430 --> 00:19:41,596 Would that be easy? 484 00:19:41,596 --> 00:19:42,450 Yeah. 485 00:19:42,450 --> 00:19:45,290 You go into a room and you go, my gosh, there's hundreds of 486 00:19:45,290 --> 00:19:45,940 books in this room. 487 00:19:45,940 --> 00:19:47,230 Would that be easy? 488 00:19:47,230 --> 00:19:50,520 Now, here are many, many more things to find the red X. But 489 00:19:50,520 --> 00:19:55,390 your mind instantly, easily, trivially finds it, right? 490 00:19:55,390 --> 00:19:58,371 So now, let's try another version of this. 491 00:19:58,371 --> 00:19:58,760 AUDIENCE: Yes. 492 00:19:58,760 --> 00:19:59,740 PROFESSOR: Yes? 493 00:19:59,740 --> 00:20:00,990 OK, get ready. 494 00:20:03,298 --> 00:20:04,680 AUDIENCE: Yes. 495 00:20:04,680 --> 00:20:05,370 PROFESSOR: All right. 496 00:20:05,370 --> 00:20:05,870 Slower? 497 00:20:05,870 --> 00:20:06,450 Harder? 498 00:20:06,450 --> 00:20:07,460 OK, some disagreement. 499 00:20:07,460 --> 00:20:08,160 OK. 500 00:20:08,160 --> 00:20:10,070 All right, so let me show you data under well controlled 501 00:20:10,070 --> 00:20:10,760 conditions. 502 00:20:10,760 --> 00:20:13,320 And here's the concept. 503 00:20:13,320 --> 00:20:15,250 People think that in many ways when we look out in the world, 504 00:20:15,250 --> 00:20:17,320 there's features, things like simple things-- 505 00:20:17,320 --> 00:20:18,490 shape or color. 506 00:20:18,490 --> 00:20:19,970 Those are the two we're looking at here. 507 00:20:19,970 --> 00:20:23,140 But in order to combine those, to know the same thing out 508 00:20:23,140 --> 00:20:26,420 there has this shape and this color, to glue together 509 00:20:26,420 --> 00:20:29,680 features, we have to have attention. 510 00:20:29,680 --> 00:20:35,100 And so when the display looked like this, a single feature-- 511 00:20:35,100 --> 00:20:35,800 redness-- 512 00:20:35,800 --> 00:20:37,380 helps you. 513 00:20:37,380 --> 00:20:39,180 A single feature-- redness-- 514 00:20:39,180 --> 00:20:39,610 helps you. 515 00:20:39,610 --> 00:20:41,750 You don't have to worry about the shape. 516 00:20:41,750 --> 00:20:45,560 Now, you have to worry about the shape and the color to 517 00:20:45,560 --> 00:20:49,630 decide the red X. Worse yet, now it's a 518 00:20:49,630 --> 00:20:50,920 slower, harder process. 519 00:20:50,920 --> 00:20:53,590 And when people do this, here's exactly what they find. 520 00:20:53,590 --> 00:20:56,760 If it's just a feature by shape or by color and that 521 00:20:56,760 --> 00:20:58,940 alone would tell you the answer, you can have one 522 00:20:58,940 --> 00:21:01,100 thing, a few things, and many things. 523 00:21:01,100 --> 00:21:03,820 And boom, your mind instantly discovers it. 524 00:21:03,820 --> 00:21:05,180 Instantly. 525 00:21:05,180 --> 00:21:07,270 This is how long it takes you to do it. 526 00:21:07,270 --> 00:21:10,460 But if you have to combine two things together, the shape and 527 00:21:10,460 --> 00:21:13,310 the color on the very same thing, combine those things, 528 00:21:13,310 --> 00:21:15,180 you need the glue of attention. 529 00:21:15,180 --> 00:21:17,840 And the more things that are out there, the more your mind 530 00:21:17,840 --> 00:21:20,470 has to search, search, search, search, until it 531 00:21:20,470 --> 00:21:21,850 finally finds it. 532 00:21:21,850 --> 00:21:24,690 It has to look around, and look around, and look around 533 00:21:24,690 --> 00:21:26,980 to find the answer. 534 00:21:26,980 --> 00:21:30,970 So some kinds of things in the world we notice without 535 00:21:30,970 --> 00:21:32,550 attention features. 536 00:21:32,550 --> 00:21:35,030 We can look everywhere at once in the display. 537 00:21:35,030 --> 00:21:36,700 That's how we're so fast. 538 00:21:36,700 --> 00:21:38,230 Things just pop out. 539 00:21:38,230 --> 00:21:40,580 And the flat slope is-- mathematically. 540 00:21:40,580 --> 00:21:42,540 Doesn't matter how many things are out there-- boom, you get 541 00:21:42,540 --> 00:21:43,990 it right away. 542 00:21:43,990 --> 00:21:47,310 Other things where we have to glue together two features 543 00:21:47,310 --> 00:21:50,130 require our attention to do the gluing 544 00:21:50,130 --> 00:21:50,940 things that are together. 545 00:21:50,940 --> 00:21:54,200 We have to search spot, spot, spot, serially. 546 00:21:54,200 --> 00:21:56,280 We don't feel the intuition of pop-out. 547 00:21:56,280 --> 00:22:00,180 And we have a steep slope in search. 548 00:22:00,180 --> 00:22:03,270 Now I need somebody from their chair who's pretty confident 549 00:22:03,270 --> 00:22:04,520 in their math. 550 00:22:07,260 --> 00:22:09,350 It's a little bit harder than before, but I'll tell you 551 00:22:09,350 --> 00:22:10,920 don't have to go into triple digits. 552 00:22:14,010 --> 00:22:16,050 Someone help me. 553 00:22:16,050 --> 00:22:17,850 I owe you one. 554 00:22:17,850 --> 00:22:18,760 I'll come back to you if there's one more. 555 00:22:18,760 --> 00:22:20,540 I can't remember, sorry. 556 00:22:20,540 --> 00:22:21,900 OK, ready? 557 00:22:21,900 --> 00:22:23,570 So you're going to see some numbers and I want you to add 558 00:22:23,570 --> 00:22:24,850 them and tell me what they are. 559 00:22:24,850 --> 00:22:25,950 And I'm going to be pretty fast. 560 00:22:25,950 --> 00:22:26,622 Is that OK? 561 00:22:26,622 --> 00:22:26,960 AUDIENCE: [INAUDIBLE]. 562 00:22:26,960 --> 00:22:27,700 PROFESSOR: [INAUDIBLE]. 563 00:22:27,700 --> 00:22:28,950 AUDIENCE: OK. 564 00:22:31,475 --> 00:22:33,095 PROFESSOR: Yes. 565 00:22:33,095 --> 00:22:34,025 OK? 566 00:22:34,025 --> 00:22:35,885 Ready? 567 00:22:35,885 --> 00:22:36,350 AUDIENCE: 14. 568 00:22:36,350 --> 00:22:37,190 PROFESSOR: Yes. 569 00:22:37,190 --> 00:22:38,474 Ready? 570 00:22:38,474 --> 00:22:39,560 AUDIENCE: 5. 571 00:22:39,560 --> 00:22:39,830 PROFESSOR: Yes. 572 00:22:39,830 --> 00:22:40,816 Ready? 573 00:22:40,816 --> 00:22:41,670 AUDIENCE: 10. 574 00:22:41,670 --> 00:22:42,350 PROFESSOR: Yes. 575 00:22:42,350 --> 00:22:42,730 Ready? 576 00:22:42,730 --> 00:22:44,030 AUDIENCE: 13. 577 00:22:44,030 --> 00:22:44,655 PROFESSOR: Yes. 578 00:22:44,655 --> 00:22:48,570 Now what were the letters between them? 579 00:22:48,570 --> 00:22:50,000 Sorry, it's a trick. 580 00:22:50,000 --> 00:22:51,005 It would have been mean. 581 00:22:51,005 --> 00:22:51,300 AUDIENCE: [INAUDIBLE]. 582 00:22:51,300 --> 00:22:52,196 PROFESSOR: Sorry? 583 00:22:52,196 --> 00:22:53,988 AUDIENCE: [INAUDIBLE] the last one so far. 584 00:22:53,988 --> 00:22:54,436 PROFESSOR: Interesting. 585 00:22:54,436 --> 00:22:54,890 OK. 586 00:22:54,890 --> 00:22:56,750 Now, you can go, oh, that was a trick. 587 00:22:56,750 --> 00:22:57,750 How reasonable is that? 588 00:22:57,750 --> 00:23:02,220 Well, let me make the argument that in order to add the 5 and 589 00:23:02,220 --> 00:23:03,570 the 8, you had to look at the letters. 590 00:23:03,570 --> 00:23:04,650 You had to look at the letters. 591 00:23:04,650 --> 00:23:06,050 They're in the middle. 592 00:23:06,050 --> 00:23:10,650 But because your mind is not attuned to that, it's as if 593 00:23:10,650 --> 00:23:11,970 they weren't there almost. 594 00:23:11,970 --> 00:23:14,005 You know some thing's there, right? 595 00:23:14,005 --> 00:23:15,570 It's right in front of you. 596 00:23:15,570 --> 00:23:16,810 It doesn't seem that hard. 597 00:23:16,810 --> 00:23:19,040 But again, if your mind is tuned to something, the other 598 00:23:19,040 --> 00:23:20,960 stuff gets fuzzy. 599 00:23:20,960 --> 00:23:23,610 And you got the R somewhere. 600 00:23:23,610 --> 00:23:25,380 Now, let me ask one more question. 601 00:23:25,380 --> 00:23:28,171 Do you the colors the letters were in? 602 00:23:28,171 --> 00:23:29,600 No. 603 00:23:29,600 --> 00:23:31,370 And in fact, you don't. 604 00:23:31,370 --> 00:23:34,200 So thanks very much for willing to do that. 605 00:23:34,200 --> 00:23:39,540 So again, we just said to bind shape and color, we need the 606 00:23:39,540 --> 00:23:40,420 glue of attention. 607 00:23:40,420 --> 00:23:41,880 We're not attending to the middle letters. 608 00:23:41,880 --> 00:23:43,340 They seem irrelevant even though they're 609 00:23:43,340 --> 00:23:44,190 right in front of us. 610 00:23:44,190 --> 00:23:46,670 Even though we have to walk through them in our mind's eye 611 00:23:46,670 --> 00:23:48,090 to move from one number to the other. 612 00:23:48,090 --> 00:23:50,270 They're like, I don't care. 613 00:23:50,270 --> 00:23:51,700 Boom, you move to the other side. 614 00:23:51,700 --> 00:23:53,680 So it's not just seeing something, it's attending to 615 00:23:53,680 --> 00:23:54,760 something that really matters. 616 00:23:54,760 --> 00:23:57,150 And on top of that, you need a lot of attention if you do 617 00:23:57,150 --> 00:24:00,010 glue together the shape and the color, that 618 00:24:00,010 --> 00:24:01,180 requires a lot of focus. 619 00:24:01,180 --> 00:24:04,480 What does this mean in everyday life? 620 00:24:04,480 --> 00:24:06,750 In everyday life, we're probably walking around 621 00:24:06,750 --> 00:24:11,225 constantly having free floating features around us 622 00:24:11,225 --> 00:24:13,340 and things we're not paying attention to. 623 00:24:13,340 --> 00:24:17,170 Right now I'm probably seeing a yellow computer over there. 624 00:24:17,170 --> 00:24:21,820 And I'm seeing some blue fire extinguisher sign over there. 625 00:24:21,820 --> 00:24:25,670 Now, why am I not freaked out by that? 626 00:24:25,670 --> 00:24:27,250 And why are you not freaked out by that? 627 00:24:27,250 --> 00:24:28,310 And why is it almost certain? 628 00:24:28,310 --> 00:24:29,630 Because if you're not attending to things, you're 629 00:24:29,630 --> 00:24:31,340 constantly having free floating 630 00:24:31,340 --> 00:24:33,320 features that are incorrect. 631 00:24:33,320 --> 00:24:37,220 Colors and shapes are just floating around you. 632 00:24:37,220 --> 00:24:39,370 Why are we not disturbed by that? 633 00:24:39,370 --> 00:24:42,030 Because we're not attending it to start with. 634 00:24:42,030 --> 00:24:43,800 Because it's like you didn't notice the letters were there. 635 00:24:43,800 --> 00:24:44,530 Well, then you're not bothered. 636 00:24:44,530 --> 00:24:45,970 They're just free floating information. 637 00:24:45,970 --> 00:24:47,390 You're not paying attention to it to start with. 638 00:24:47,390 --> 00:24:49,410 But that's what's happening in your mind all the time. 639 00:24:49,410 --> 00:24:51,330 And these experiments demonstrate that empirically 640 00:24:51,330 --> 00:24:54,180 and directly. 641 00:24:54,180 --> 00:24:57,370 There's also interesting temporal constraints. 642 00:24:57,370 --> 00:24:59,040 I'll describe this experiment because it's too 643 00:24:59,040 --> 00:25:00,000 hard to pull off here. 644 00:25:00,000 --> 00:25:02,470 So there's a phenomenon called the attentional blink. 645 00:25:02,470 --> 00:25:05,020 What's weird about these things, right? 646 00:25:05,020 --> 00:25:07,280 Why is our minds like this? 647 00:25:07,280 --> 00:25:09,630 You can speculate on that, but it's really well documented. 648 00:25:09,630 --> 00:25:11,340 We attend to a small percentage of 649 00:25:11,340 --> 00:25:12,540 what's around us. 650 00:25:12,540 --> 00:25:15,275 And we need a lot of resources to know something well or 651 00:25:15,275 --> 00:25:16,170 learn something well. 652 00:25:16,170 --> 00:25:17,720 So here's another example. 653 00:25:17,720 --> 00:25:20,800 What participants see is letters presented very quickly 654 00:25:20,800 --> 00:25:21,600 one after the other-- 655 00:25:21,600 --> 00:25:26,260 E-L-H. Their job is to report any numbers that they see. 656 00:25:26,260 --> 00:25:27,920 So you go, boom, boom, boom, boom. 657 00:25:27,920 --> 00:25:30,300 What were the numbers? 658 00:25:30,300 --> 00:25:32,480 Pretty fast, but humanly possible. 659 00:25:32,480 --> 00:25:35,520 The interesting question is, what's the time gap between 660 00:25:35,520 --> 00:25:36,630 the two numbers that you see? 661 00:25:36,630 --> 00:25:38,360 There's usually two numbers. 662 00:25:38,360 --> 00:25:40,490 We'll call this one the first one, time 1. 663 00:25:40,490 --> 00:25:41,970 The second one, time 2. 664 00:25:41,970 --> 00:25:44,230 What's the gap between them, and how does that influence 665 00:25:44,230 --> 00:25:46,000 whether you're likely to report them? 666 00:25:46,000 --> 00:25:50,450 So here's what's called the attentional blink and a 667 00:25:50,450 --> 00:25:52,760 slightly scary graph that I'll tell you. 668 00:25:52,760 --> 00:25:56,830 And let me tell you what it's like, the metaphor, and then 669 00:25:56,830 --> 00:25:58,060 let me show you the result. 670 00:25:58,060 --> 00:25:58,860 The metaphor is this. 671 00:25:58,860 --> 00:26:03,930 Your mind, moment to moment to moment, is kind of like if you 672 00:26:03,930 --> 00:26:07,820 could imagine you're next to a stream, a big stream with a 673 00:26:07,820 --> 00:26:09,120 lot of fish zooming by. 674 00:26:09,120 --> 00:26:10,290 You have a fishnet. 675 00:26:10,290 --> 00:26:13,110 You dip into the stream. 676 00:26:13,110 --> 00:26:15,260 You catch something at that moment. 677 00:26:15,260 --> 00:26:18,740 But if a fish comes by just as you pull your fishnet out, 678 00:26:18,740 --> 00:26:19,925 will you catch that one? 679 00:26:19,925 --> 00:26:22,530 No. 680 00:26:22,530 --> 00:26:24,320 So here's what happens. 681 00:26:24,320 --> 00:26:27,320 Here's the time between the presentation. 682 00:26:29,830 --> 00:26:34,570 If there's approximately three numbers between them, a couple 683 00:26:34,570 --> 00:26:37,820 hundred milliseconds, you're almost a chance for the 684 00:26:37,820 --> 00:26:39,970 possibility of reporting the second digit. 685 00:26:39,970 --> 00:26:43,010 It's sort of like the fish you can't catch because you just 686 00:26:43,010 --> 00:26:43,860 took your net out. 687 00:26:43,860 --> 00:26:47,110 Except it's your mind, it's not a physical limitation. 688 00:26:47,110 --> 00:26:48,160 So here's what it's like. 689 00:26:48,160 --> 00:26:49,100 So this is perfect performance. 690 00:26:49,100 --> 00:26:51,530 This is chance performance. 691 00:26:51,530 --> 00:26:54,710 If it immediately follows, one number immediately follows the 692 00:26:54,710 --> 00:26:57,480 next number, you're going to get both. 693 00:26:57,480 --> 00:26:59,850 And the way to think about that is, it's kind of like 694 00:26:59,850 --> 00:27:00,740 fishnet again. 695 00:27:00,740 --> 00:27:02,590 You just put in your fishnet. 696 00:27:02,590 --> 00:27:03,460 You got one fish. 697 00:27:03,460 --> 00:27:05,170 And if you're lucky, while you're kind of pulling it out, 698 00:27:05,170 --> 00:27:06,410 the next fish just swims in there. 699 00:27:06,410 --> 00:27:08,180 You got two. 700 00:27:08,180 --> 00:27:10,720 Now your fishnet's out. 701 00:27:10,720 --> 00:27:12,650 And if something else comes, and something else comes, and 702 00:27:12,650 --> 00:27:14,910 something else comes, you're pretty bad at recognizing it 703 00:27:14,910 --> 00:27:17,180 was there at all. 704 00:27:17,180 --> 00:27:19,440 A little bit of time comes back and you put your fishnet 705 00:27:19,440 --> 00:27:22,520 back in and you're catching everything again. 706 00:27:22,520 --> 00:27:24,640 About 1/2 second. 707 00:27:24,640 --> 00:27:26,900 But how weird is that that your mind works like that for 708 00:27:26,900 --> 00:27:28,370 spotting the digits? 709 00:27:28,370 --> 00:27:30,690 That it has to take a break? 710 00:27:30,690 --> 00:27:33,030 There's a part of your mind that's getting the digits. 711 00:27:33,030 --> 00:27:37,180 And once it's gotten a digit, for the next couple of hundred 712 00:27:37,180 --> 00:27:41,090 milliseconds, it's not going to have room to pay attention 713 00:27:41,090 --> 00:27:42,170 to the following digits. 714 00:27:42,170 --> 00:27:45,090 How weird is that? 715 00:27:45,090 --> 00:27:47,620 It's because our mind has all these kind of limitations on 716 00:27:47,620 --> 00:27:49,860 what we perceive in the environment. 717 00:27:49,860 --> 00:27:52,830 Here's another sort of fun example of that. 718 00:27:52,830 --> 00:27:54,080 So here's a task. 719 00:27:54,080 --> 00:27:57,030 One of these dots are going to-- 720 00:27:57,030 --> 00:27:58,520 One dot is going to blink. 721 00:27:58,520 --> 00:28:01,250 And your job is to keep your eye on that dot. 722 00:28:01,250 --> 00:28:03,090 You know when they say, keep your eye on the ball. 723 00:28:03,090 --> 00:28:04,260 Keep your eye on the dot. 724 00:28:04,260 --> 00:28:07,055 And when it stops moving, notice which is the one you've 725 00:28:07,055 --> 00:28:07,800 been keeping your eye on. 726 00:28:07,800 --> 00:28:09,275 So the first one's going to be easy, but it 727 00:28:09,275 --> 00:28:09,970 will give you a feeling. 728 00:28:09,970 --> 00:28:11,200 And then you'll see the answer given to you. 729 00:28:11,200 --> 00:28:11,890 Ready? 730 00:28:11,890 --> 00:28:14,260 People call this object tracking. 731 00:28:14,260 --> 00:28:15,976 OK, here we go. 732 00:28:15,976 --> 00:28:17,226 See that one blinking? 733 00:28:32,500 --> 00:28:35,185 Now, let's try it one thing moving, but even faster. 734 00:29:14,810 --> 00:29:16,650 So let's try tracking four things. 735 00:29:57,620 --> 00:29:58,870 Last example. 736 00:30:16,160 --> 00:30:20,590 So here's a graph with a high level of 737 00:30:20,590 --> 00:30:21,970 performance how people do. 738 00:30:21,970 --> 00:30:24,390 So two things matter, and you might have felt that. 739 00:30:24,390 --> 00:30:27,070 How many are viewing the number of targets? 740 00:30:27,070 --> 00:30:29,800 And under many circumstances, people seem to be able to do 741 00:30:29,800 --> 00:30:31,780 about three to four. 742 00:30:31,780 --> 00:30:33,120 But also, the speed matters, too. 743 00:30:33,120 --> 00:30:33,990 You might have felt that, too. 744 00:30:33,990 --> 00:30:35,480 Both things contribute. 745 00:30:35,480 --> 00:30:38,560 There's a lot of interest in this three to four limit that 746 00:30:38,560 --> 00:30:40,410 people seem to have for things that they track, 747 00:30:40,410 --> 00:30:42,420 three to four items. 748 00:30:42,420 --> 00:30:45,000 And there's a lot of work, at Harvard actually, in 749 00:30:45,000 --> 00:30:46,680 developmental psychology and related. 750 00:30:46,680 --> 00:30:48,670 Let me just say a word about that to you. 751 00:30:48,670 --> 00:30:51,270 There's a thought that pretty young when you're a kid, you 752 00:30:51,270 --> 00:30:52,920 get to about this three to four things. 753 00:30:52,920 --> 00:30:55,510 That our minds naturally contract about, or notice, or 754 00:30:55,510 --> 00:30:58,490 keep a count of about three to four things in the world. 755 00:30:58,490 --> 00:31:03,130 And so people who study, for example, math and the way that 756 00:31:03,130 --> 00:31:05,960 humans perform it, think that we have in us, in our brains, 757 00:31:05,960 --> 00:31:08,180 approximately two systems that are physically different and 758 00:31:08,180 --> 00:31:10,090 culturally different. 759 00:31:10,090 --> 00:31:11,455 And there's lots of debates about this, I 760 00:31:11,455 --> 00:31:12,760 should tell you this. 761 00:31:12,760 --> 00:31:16,430 But one of them being something like accurate 762 00:31:16,430 --> 00:31:19,020 counting up to about three or four. 763 00:31:19,020 --> 00:31:22,580 And then, cultures that don't have a organized system of 764 00:31:22,580 --> 00:31:24,780 math, the next answer is many. 765 00:31:24,780 --> 00:31:26,240 You've heard this, right? 766 00:31:26,240 --> 00:31:30,580 It's like one, two, three, maybe four, and then a lot. 767 00:31:30,580 --> 00:31:35,220 Now, we don't stop counting at four, but we might if we were 768 00:31:35,220 --> 00:31:37,600 in a true state of nature and we didn't have an organized 769 00:31:37,600 --> 00:31:38,460 system of math. 770 00:31:38,460 --> 00:31:40,200 And weren't taught that in how it relates to 771 00:31:40,200 --> 00:31:41,610 our language abilities. 772 00:31:41,610 --> 00:31:43,780 So there's a lot of debate about the exactness of this. 773 00:31:43,780 --> 00:31:45,690 But there's something fascinating about this idea 774 00:31:45,690 --> 00:31:48,910 that our minds can have about four units of information kept 775 00:31:48,910 --> 00:31:51,195 in mind, kept active, that we can track. 776 00:31:51,195 --> 00:31:55,750 And that lots of things in the world reflect that limitation. 777 00:31:55,750 --> 00:31:58,360 Including, for example, exact counting. 778 00:31:58,360 --> 00:32:01,730 And without the cultural invention of mathematics, we 779 00:32:01,730 --> 00:32:04,540 would count one, two, three, four, lots. 780 00:32:04,540 --> 00:32:06,690 But there's debates about this because exactly how you 781 00:32:06,690 --> 00:32:09,080 measure it can influence how you think about it. 782 00:32:09,080 --> 00:32:10,420 Now, can attention be trained? 783 00:32:10,420 --> 00:32:14,550 Could you be trained to have much more attention? 784 00:32:14,550 --> 00:32:16,730 So in many ways, we don't think you can. 785 00:32:16,730 --> 00:32:18,290 But in some ways, you can. 786 00:32:18,290 --> 00:32:21,140 And part of this has come out of this ironic line of 787 00:32:21,140 --> 00:32:23,640 research about things that are supposed to be bad for us. 788 00:32:23,640 --> 00:32:25,340 Some of them might be OK for us. 789 00:32:25,340 --> 00:32:27,620 Video games, OK? 790 00:32:27,620 --> 00:32:31,310 If you're in my generation, people are going, oh, boy, 791 00:32:31,310 --> 00:32:35,060 those video games are messing up all these kids. 792 00:32:35,060 --> 00:32:36,950 Isn't that what you guys think? 793 00:32:36,950 --> 00:32:38,690 No. 794 00:32:38,690 --> 00:32:39,160 OK. 795 00:32:39,160 --> 00:32:43,030 So Daphne Bavelier at Rochester has done a series of 796 00:32:43,030 --> 00:32:45,310 studies showing that if you play certain kinds of video 797 00:32:45,310 --> 00:32:48,700 games, you actually do expand your attentional capacity to 798 00:32:48,700 --> 00:32:49,600 practice on those games. 799 00:32:49,600 --> 00:32:51,490 Because they make you practice a lot. 800 00:32:51,490 --> 00:32:52,700 So here's what she did. 801 00:32:52,700 --> 00:32:55,120 She took people who didn't have experience 802 00:32:55,120 --> 00:32:56,080 playing video games. 803 00:32:56,080 --> 00:32:58,170 And now we're specifically talking about first-person 804 00:32:58,170 --> 00:33:00,940 shooter video games. 805 00:33:00,940 --> 00:33:05,340 And had them do 30 hours of practice. 806 00:33:05,340 --> 00:33:09,210 And she looked at their ability to do multiple object 807 00:33:09,210 --> 00:33:11,880 tracking of the kind you just saw, those dots whizzing 808 00:33:11,880 --> 00:33:13,950 around, before and after. 809 00:33:13,950 --> 00:33:16,390 So the games, as you know, are very different. 810 00:33:16,390 --> 00:33:18,340 But they involve keeping track of lots of things 811 00:33:18,340 --> 00:33:19,520 happening at once. 812 00:33:19,520 --> 00:33:20,920 And here's what she found. 813 00:33:20,920 --> 00:33:24,830 That these people who played 30 hours, these were people 814 00:33:24,830 --> 00:33:28,210 who were not gamers beforehand, got better at 815 00:33:28,210 --> 00:33:30,480 performing these tasks of keeping 816 00:33:30,480 --> 00:33:32,410 track of multiple objects. 817 00:33:32,410 --> 00:33:35,130 Of all things, a group that's super interested in that is 818 00:33:35,130 --> 00:33:36,410 the military. 819 00:33:36,410 --> 00:33:39,078 And why do you think that is? 820 00:33:39,078 --> 00:33:40,860 AUDIENCE: [INAUDIBLE]. 821 00:33:40,860 --> 00:33:43,400 PROFESSOR: OK. 822 00:33:43,400 --> 00:33:45,800 There was just a paper published recently that if 823 00:33:45,800 --> 00:33:48,320 they take a brain image of your brain, they can predict 824 00:33:48,320 --> 00:33:50,520 how good a gamer you'll become if you haven't 825 00:33:50,520 --> 00:33:52,360 been a gamer before. 826 00:33:52,360 --> 00:33:53,570 These are adults. 827 00:33:53,570 --> 00:33:55,050 If you have bigger basal ganglia, you'll become a 828 00:33:55,050 --> 00:33:59,420 better gamer in the next week starting from scratch. 829 00:33:59,420 --> 00:34:01,540 The military's interested because you may know this. 830 00:34:01,540 --> 00:34:05,560 An increasing amount of warfare is video game-like. 831 00:34:05,560 --> 00:34:07,320 Which is kind of the haunting thing. 832 00:34:07,320 --> 00:34:10,360 Because this is real people really being killed. 833 00:34:10,360 --> 00:34:12,550 But, do you know the whole thing with the drones? 834 00:34:12,550 --> 00:34:13,960 Have you followed this, some of you? 835 00:34:13,960 --> 00:34:14,560 OK. 836 00:34:14,560 --> 00:34:17,600 So the US has had a lot of success, and there's a lot of 837 00:34:17,600 --> 00:34:20,670 debates about the ethics of this, in sending in unmanned 838 00:34:20,670 --> 00:34:22,949 small airplanes into the Middle East. 839 00:34:22,949 --> 00:34:28,110 And using that to kill opponents. 840 00:34:28,110 --> 00:34:30,310 The person sitting in Washington, DC, or somewhere 841 00:34:30,310 --> 00:34:34,020 else, is watching kind of like a video game when they make 842 00:34:34,020 --> 00:34:35,270 that decision about whether that's 843 00:34:35,270 --> 00:34:37,530 somebody to be attacked. 844 00:34:37,530 --> 00:34:40,400 So a lot of operations now are computer-based and they're 845 00:34:40,400 --> 00:34:42,909 kind of like a video game at one level. 846 00:34:42,909 --> 00:34:45,830 And they're real-life hard decisions at another level. 847 00:34:45,830 --> 00:34:48,620 So video games, they're becoming more and more, as you 848 00:34:48,620 --> 00:34:52,340 can sense, a part of real world decisions and controls. 849 00:34:52,340 --> 00:34:56,350 If you play Tetris because it doesn't require that kind of 850 00:34:56,350 --> 00:34:58,360 fast thought stuff, you don't get any benefit. 851 00:34:58,360 --> 00:35:00,670 There could be other benefits, but not this one. 852 00:35:00,670 --> 00:35:04,040 So it's not just video games, it's the mental processes that 853 00:35:04,040 --> 00:35:07,150 are practiced in the video games that give you certain 854 00:35:07,150 --> 00:35:11,140 talents or not, or abilities you didn't have before. 855 00:35:11,140 --> 00:35:13,170 OK, the last experiment I want to tell you is this. 856 00:35:13,170 --> 00:35:16,060 So everything we've talked about now, except a little bit 857 00:35:16,060 --> 00:35:18,670 about the stuff in the second year when you're shadowing, 858 00:35:18,670 --> 00:35:25,670 has been about this idea that we don't pay much attention to 859 00:35:25,670 --> 00:35:26,180 what's around us. 860 00:35:26,180 --> 00:35:28,320 We don't get much in our mind compared to what we think we 861 00:35:28,320 --> 00:35:30,270 would, even if it's right in front of us, even if it's all 862 00:35:30,270 --> 00:35:30,910 over the place. 863 00:35:30,910 --> 00:35:33,520 Even if we've seen a penny 17,000 times. 864 00:35:33,520 --> 00:35:35,870 But there are some studies that have shown something 865 00:35:35,870 --> 00:35:39,030 pretty interesting also, which reveals something very 866 00:35:39,030 --> 00:35:41,650 paradoxical about the human mind. 867 00:35:41,650 --> 00:35:44,155 That we can sometimes know things without seeing. 868 00:35:44,155 --> 00:35:46,290 And people call that subliminal perception. 869 00:35:46,290 --> 00:35:48,120 So how this experiment works is this. 870 00:35:48,120 --> 00:35:49,545 And let me show you one example behaviorally. 871 00:35:49,545 --> 00:35:51,690 I'm going to show you one brain example. 872 00:35:51,690 --> 00:35:53,070 So here's what happens. 873 00:35:53,070 --> 00:35:55,260 Let's pretend you have to read words aloud. 874 00:35:55,260 --> 00:35:57,450 And we measure, to the millisecond, how fast 875 00:35:57,450 --> 00:35:58,780 you read the word. 876 00:35:58,780 --> 00:36:01,820 If I show you this word and I show you this word, and I 877 00:36:01,820 --> 00:36:04,080 measure how quickly you read the word "doctor," you're 878 00:36:04,080 --> 00:36:07,380 slower than if you read the word "nurse" and then you read 879 00:36:07,380 --> 00:36:10,610 the word "doctor." Because these two are related. 880 00:36:10,610 --> 00:36:12,270 And something in your mind gets warmed up 881 00:36:12,270 --> 00:36:14,430 when you see this. 882 00:36:14,430 --> 00:36:16,730 It doesn't get warmed up in a relevant way when 883 00:36:16,730 --> 00:36:17,260 you see this word. 884 00:36:17,260 --> 00:36:18,240 Does that make sense? 885 00:36:18,240 --> 00:36:20,090 You see one word and then you read the next one. 886 00:36:20,090 --> 00:36:22,340 If they're related semantically or conceptually 887 00:36:22,340 --> 00:36:24,400 by meaning, you're faster for the second one. 888 00:36:24,400 --> 00:36:25,330 OK, that's easy. 889 00:36:25,330 --> 00:36:27,000 And it's not too surprising. 890 00:36:27,000 --> 00:36:29,390 The clever discovery was this. 891 00:36:29,390 --> 00:36:31,600 What they do is they do something call-- they present 892 00:36:31,600 --> 00:36:34,450 these words under what they call masked conditions. 893 00:36:34,450 --> 00:36:36,510 On the computer monitor you get something busy, like a 894 00:36:36,510 --> 00:36:37,620 roll of X's. 895 00:36:37,620 --> 00:36:41,700 For 10 milliseconds only, ten-thousandths of a second, 896 00:36:41,700 --> 00:36:44,770 the word appears, and then there's a bunch of X's again. 897 00:36:44,770 --> 00:36:47,740 So it's like a flash of X's. 898 00:36:47,740 --> 00:36:49,930 And people, they create situations-- 899 00:36:49,930 --> 00:36:51,440 and I'll show you this again in a moment. 900 00:36:51,440 --> 00:36:54,490 They create situations where people say, I'm not sure 901 00:36:54,490 --> 00:36:55,410 anything was there. 902 00:36:55,410 --> 00:36:56,780 I certainly don't know what it is. 903 00:36:56,780 --> 00:36:59,150 You have to be there and you have to convince yourself, as 904 00:36:59,150 --> 00:37:01,810 a scientist, that you've done that. 905 00:37:01,810 --> 00:37:02,780 And then they have you read the word 906 00:37:02,780 --> 00:37:04,480 "doctor" in full view. 907 00:37:04,480 --> 00:37:07,250 And in these circumstances where people cannot tell you 908 00:37:07,250 --> 00:37:10,310 that they saw a word, and they certainly can't tell you what 909 00:37:10,310 --> 00:37:14,820 the word was, you still get faster. 910 00:37:14,820 --> 00:37:19,910 So they don't know that they saw "church" or "nurse," but 911 00:37:19,910 --> 00:37:22,110 something in their mind picks it up. 912 00:37:22,110 --> 00:37:24,170 And what's picked up in the mind under the subliminal 913 00:37:24,170 --> 00:37:28,250 presentation changes the behavior when you have an 914 00:37:28,250 --> 00:37:30,830 overt act of behavior reading the next word aloud. 915 00:37:30,830 --> 00:37:34,380 So this is hardcore empirical evidence that your mind is 916 00:37:34,380 --> 00:37:36,990 able to pick up stuff at the level of meaning. 917 00:37:36,990 --> 00:37:38,400 Think about the paradox. 918 00:37:38,400 --> 00:37:41,350 You think when I read a word, I see it and then I understand 919 00:37:41,350 --> 00:37:42,530 its meaning. 920 00:37:42,530 --> 00:37:46,170 Here it's flashed in such a way that you don't feel like 921 00:37:46,170 --> 00:37:49,380 you see it, but you extract its meaning anyway. 922 00:37:49,380 --> 00:37:51,290 And we know that because it makes you read the next word 923 00:37:51,290 --> 00:37:53,410 faster only if they're related in meaning. 924 00:37:53,410 --> 00:37:55,630 So your mind is subliminally primed. 925 00:37:55,630 --> 00:38:00,050 Now, some years ago people used to worry that advertisers 926 00:38:00,050 --> 00:38:02,720 could do this really cleverly and make you go do stuff. 927 00:38:02,720 --> 00:38:04,960 The famous example is historically, can they make 928 00:38:04,960 --> 00:38:05,890 you buy popcorn? 929 00:38:05,890 --> 00:38:08,480 Which is now, since it's $48, that's harder to do. 930 00:38:08,480 --> 00:38:12,420 But can they make you buy popcorn at the movie theater? 931 00:38:12,420 --> 00:38:15,770 Or can they make you buy a Ford, or a Chevrolet, or 932 00:38:15,770 --> 00:38:17,570 something like that, with the right subliminal message and 933 00:38:17,570 --> 00:38:18,730 that kind of stuff? 934 00:38:18,730 --> 00:38:21,900 Every experiment that's been done to ask whether subliminal 935 00:38:21,900 --> 00:38:25,680 messages of this kind would make you get up and go do 936 00:38:25,680 --> 00:38:29,320 something, like buy a car, or change what country you live 937 00:38:29,320 --> 00:38:31,500 in, or something, it never works. 938 00:38:31,500 --> 00:38:32,740 It never works. 939 00:38:32,740 --> 00:38:35,110 Nobody's ever been able to show that in any 940 00:38:35,110 --> 00:38:36,380 experimental condition. 941 00:38:36,380 --> 00:38:38,090 Which doesn't prevent some companies from selling these 942 00:38:38,090 --> 00:38:41,310 products to store owners that are thinking they're getting 943 00:38:41,310 --> 00:38:43,130 their sales up. 944 00:38:43,130 --> 00:38:45,390 But there's been never any evidence that these kinds of 945 00:38:45,390 --> 00:38:48,270 subliminal presentations make people do something big and 946 00:38:48,270 --> 00:38:51,140 complicated, like get up and go buy popcorn, or choose one 947 00:38:51,140 --> 00:38:52,410 product over another. 948 00:38:52,410 --> 00:38:54,390 People have never been able to show that under controlled 949 00:38:54,390 --> 00:38:54,970 experiment. 950 00:38:54,970 --> 00:38:57,490 But they have been able to show these kinds of changes in 951 00:38:57,490 --> 00:39:01,060 behavior that last a moment. 952 00:39:01,060 --> 00:39:03,270 So what's happening in your brain? 953 00:39:03,270 --> 00:39:04,720 So this is the last experiment. 954 00:39:04,720 --> 00:39:07,240 Again, they're presenting these things that in the same 955 00:39:07,240 --> 00:39:09,260 place in the computer monitor all these things 956 00:39:09,260 --> 00:39:10,870 to fill your eyes. 957 00:39:10,870 --> 00:39:16,080 And then they present a word for either 29 milliseconds- 958 00:39:16,080 --> 00:39:19,840 they present it either very briefly or in full view. 959 00:39:19,840 --> 00:39:24,410 OK, so here it's visible because they remove the 960 00:39:24,410 --> 00:39:27,060 confusing visual information just before and after it. 961 00:39:27,060 --> 00:39:29,120 Here, people don't report seeing it. 962 00:39:29,120 --> 00:39:30,810 And scientifically, it's really important because you 963 00:39:30,810 --> 00:39:33,300 just say, oh, people didn't see it. 964 00:39:33,300 --> 00:39:34,630 But it's really important in our field to 965 00:39:34,630 --> 00:39:36,090 say, we're so convinced. 966 00:39:36,090 --> 00:39:37,520 And why are we so convinced? 967 00:39:37,520 --> 00:39:39,590 Because we say, was there any word present? 968 00:39:39,590 --> 00:39:41,290 Because sometimes there's a word present and sometimes 969 00:39:41,290 --> 00:39:41,790 there isn't. 970 00:39:41,790 --> 00:39:43,470 Just is there a word present? 971 00:39:43,470 --> 00:39:45,780 If you could see the word, because these things were a 972 00:39:45,780 --> 00:39:48,350 little bit away in time and you could see it, you almost 973 00:39:48,350 --> 00:39:49,980 always got it if it was visible. 974 00:39:49,980 --> 00:39:52,130 You almost never got it if it was here. 975 00:39:52,130 --> 00:39:54,770 You almost never said there was a word present. 976 00:39:54,770 --> 00:39:56,660 And could you name it? 977 00:39:56,660 --> 00:39:58,060 Yes, if you saw it like this. 978 00:39:58,060 --> 00:40:00,300 You couldn't name it if it was presented brief-- if it was 979 00:40:00,300 --> 00:40:01,590 surrounded by this visual stuff that 980 00:40:01,590 --> 00:40:03,030 makes it hard to see. 981 00:40:03,030 --> 00:40:06,170 If they give you a memory test for it, if you saw it fully in 982 00:40:06,170 --> 00:40:07,320 full view, you got it. 983 00:40:07,320 --> 00:40:09,400 If it was hidden like this, you didn't. 984 00:40:09,400 --> 00:40:10,730 And finally, they give you a fourth choice. 985 00:40:10,730 --> 00:40:13,820 They say, did you see the word "note" or did you see the word 986 00:40:13,820 --> 00:40:15,900 "coat?" Choose between the two. 987 00:40:15,900 --> 00:40:19,510 People are still at chance 50-50. 988 00:40:19,510 --> 00:40:20,790 They're really pushing people to say, 989 00:40:20,790 --> 00:40:22,260 really, did you see anything? 990 00:40:22,260 --> 00:40:22,950 Do you know what it was? 991 00:40:22,950 --> 00:40:25,610 There are completely a chance they don't remember anything. 992 00:40:25,610 --> 00:40:27,980 They can't even tell if a word was there. 993 00:40:27,980 --> 00:40:29,800 But we know it influences their mind. 994 00:40:29,800 --> 00:40:31,890 And now let me show you what we know of how it influences 995 00:40:31,890 --> 00:40:33,090 your brain. 996 00:40:33,090 --> 00:40:35,610 So what's shown here in green are parts of your brain that 997 00:40:35,610 --> 00:40:39,230 respond when a word is shown that you can see. 998 00:40:39,230 --> 00:40:41,680 And what's shown here in red are the parts of your brain 999 00:40:41,680 --> 00:40:45,320 that respond for that subliminally presented word, 1000 00:40:45,320 --> 00:40:48,860 the word that's presented but you can't tell that any word 1001 00:40:48,860 --> 00:40:50,870 was there, never mind what word it was. 1002 00:40:50,870 --> 00:40:54,290 So at first you can see there's a lot more brain 1003 00:40:54,290 --> 00:40:57,750 response for the visible than the masked or subliminal word. 1004 00:40:57,750 --> 00:40:59,190 But that makes a lot of sense. 1005 00:40:59,190 --> 00:41:01,240 You see something, you think about, you know it happened. 1006 00:41:01,240 --> 00:41:03,480 But you can measure what's happening in the brain. 1007 00:41:03,480 --> 00:41:04,860 It's very small. 1008 00:41:04,860 --> 00:41:06,510 Take this spot. 1009 00:41:06,510 --> 00:41:08,400 Here's the response if it's in full view. 1010 00:41:08,400 --> 00:41:11,350 Here's a tiny, tiny, tiny, tiny response when it's 1011 00:41:11,350 --> 00:41:13,080 presented in that minimalist way. 1012 00:41:13,080 --> 00:41:14,910 But it's measurable and it's present. 1013 00:41:14,910 --> 00:41:18,020 And we know that it's enough for you to figure out 1014 00:41:18,020 --> 00:41:20,960 unconsciously what the meaning of that word was. 1015 00:41:20,960 --> 00:41:23,540 So then, we ended with this paradox, which is on the one 1016 00:41:23,540 --> 00:41:27,480 hand, we live in a world where we only notice a small 1017 00:41:27,480 --> 00:41:29,050 fraction of what's around us. 1018 00:41:29,050 --> 00:41:30,960 What we focus our attention to because we have limited 1019 00:41:30,960 --> 00:41:32,780 attentional capacity. 1020 00:41:32,780 --> 00:41:34,860 On the other hand, there are channels of information that 1021 00:41:34,860 --> 00:41:36,580 sneak into your mind. 1022 00:41:36,580 --> 00:41:39,860 That probably influence you only for a few moments in the 1023 00:41:39,860 --> 00:41:42,180 unattended ear, in the cocktail parity. 1024 00:41:42,180 --> 00:41:44,240 You're actually picking up that conversation because if 1025 00:41:44,240 --> 00:41:46,330 they say your name, you'll get it. 1026 00:41:46,330 --> 00:41:48,620 Here's an example where your brain is responding to a word 1027 00:41:48,620 --> 00:41:51,310 you cannot identify or know that it was presented, and 1028 00:41:51,310 --> 00:41:54,840 those things can slightly move your behavior around. 1029 00:41:54,840 --> 00:41:58,800 But not in a science fiction or marketing dream that they 1030 00:41:58,800 --> 00:42:00,770 could make you go do something really big. 1031 00:42:00,770 --> 00:42:02,260 OK, that's never been shown and people have 1032 00:42:02,260 --> 00:42:03,370 tried to show that. 1033 00:42:03,370 --> 00:42:05,770 So attention is really weirder than you think. 1034 00:42:05,770 --> 00:42:08,120 You don't notice a lot of stuff, but a few things you 1035 00:42:08,120 --> 00:42:10,680 don't think are influencing you are influencing you here 1036 00:42:10,680 --> 00:42:11,360 and there all the time. 1037 00:42:11,360 --> 00:42:12,610 Thanks very much.