WEBVTT 00:12.530 --> 00:16.130 Professor Paul Bloom: I'll begin the class officially 00:16.129 --> 00:18.629 with a different sort of demonstration. 00:18.630 --> 00:21.560 I want to just show you one of the change-blindness studies 00:21.559 --> 00:23.529 that has been done in the real world. 00:23.530 --> 00:29.530 00:29.530 --> 00:33.850 And these videotapes are not available publicly. 00:33.850 --> 00:37.500 We get them from the web and see them as little Java scripts. 00:37.500 --> 00:41.770 So, this is one of the first studies done by Dan Simons when 00:41.771 --> 00:43.221 he was at Cornell. 00:43.220 --> 00:46.630 And his adviser at the time was our Frank Keil, 00:46.633 --> 00:48.863 who's now in our department. 00:48.860 --> 00:49.850 So, here's the study. 00:49.850 --> 01:05.020 01:05.019 --> 01:10.729 [laughter] And you don't notice it. 01:10.730 --> 01:15.590 Change blindness is one of the more striking phenomena 01:15.589 --> 01:20.999 discovered by laboratory scientists and by psychologists. 01:21.000 --> 01:24.650 But it's important to realize, to get away from the sort of 01:24.653 --> 01:28.503 surprise of the gorilla and the fact that it's hard to see the 01:28.495 --> 01:31.325 flickering--the object that's flickering, 01:31.330 --> 01:33.980 and appreciate the big moral of this, because the big moral of 01:33.983 --> 01:36.593 this is actually, I think, striking and quite 01:36.593 --> 01:40.193 important. You think right now that you're 01:40.186 --> 01:42.216 perceiving the world. 01:42.220 --> 01:45.620 I look down on you and I think I have a whole sense of where 01:45.616 --> 01:48.056 everybody is. I can't see everybody perfectly 01:48.061 --> 01:50.531 in back. You're kind of far away and 01:50.528 --> 01:55.138 blurry but there's a sense in which I have a world around me. 01:55.140 --> 01:59.070 Similarly, if I'm to close my eyes for a second, 01:59.069 --> 02:03.999 everything just remains and I could sort of remember some of 02:04.001 --> 02:06.511 the things that are there. 02:06.510 --> 02:08.990 That's really good sound localization by me . 02:08.990 --> 02:13.410 So you're looking up and you think you have a sense of the 02:13.406 --> 02:16.346 world both in perception and memory. 02:16.349 --> 02:19.249 The change-blindness experiment suggested this isn't true. 02:19.250 --> 02:23.850 The change-blindness experiment suggests that if you look at me 02:23.846 --> 02:28.066 for a second and during that second all of your classmates 02:28.073 --> 02:31.383 change positions, including those next to you, 02:31.382 --> 02:33.762 you are extremely unlikely to notice. 02:33.759 --> 02:37.019 The change-blindness experiment suggests that if you turn your 02:37.020 --> 02:40.280 eyes away from me towards there for a second and turn back, 02:40.280 --> 02:44.280 and I'm dressed entirely differently, you wouldn't 02:44.284 --> 02:46.604 notice. The exceptions would be if you 02:46.600 --> 02:49.340 told yourself consciously, "Remember what this guy is 02:49.340 --> 02:50.670 wearing; he's wearing this, 02:50.672 --> 02:51.552 that and the other." 02:51.550 --> 02:56.480 But if you don't do it consciously you'll lose it, 02:56.482 --> 02:59.202 and usually this is okay. 02:59.199 --> 03:03.039 Usually, it's okay because your memory and your visual system 03:03.038 --> 03:05.788 exploits a basic fact about the universe, 03:05.789 --> 03:09.319 which is that most things stay the same most of the time. 03:09.319 --> 03:13.029 I don't have to explicitly remember that you're over there 03:13.031 --> 03:17.131 when I turn my head for a second because you'll be over there in 03:17.133 --> 03:19.773 any case. You don't need to hold precise 03:19.770 --> 03:21.560 representations of the world. 03:21.560 --> 03:26.430 And so you only notice it in certain clever circumstances. 03:26.430 --> 03:29.700 One sort of clever circumstance is when psychologists change 03:29.703 --> 03:32.203 reality as in the change-blindness studies. 03:32.199 --> 03:35.639 A second sort of circumstance is in movies. 03:35.639 --> 03:39.759 So, one of the big surprises when people started making 03:39.762 --> 03:44.112 movies involving cuts was it is extremely difficult to get 03:44.114 --> 03:46.714 everything continuously right. 03:46.710 --> 03:49.120 And you need to work very hard to notice. 03:49.120 --> 03:52.200 So, there's all of these continuity errors that creep up 03:52.198 --> 03:55.718 into movies and you have to be a film buff or writing it down to 03:55.724 --> 03:56.904 even notice this. 03:56.900 --> 04:01.470 And the overall moral here then is that your perception of 04:01.468 --> 04:05.318 reality is a lot more sparse, a lot more limited, 04:05.315 --> 04:07.795 than you might think it is. 04:07.800 --> 04:11.530 So, this is where we were at the end of last class. 04:11.530 --> 04:14.550 We were talking about the different sorts of memories: 04:14.546 --> 04:17.046 Sensory memory, which is the sort of fraction 04:17.050 --> 04:20.520 of a second of sensory residue of what you're hearing and what 04:20.521 --> 04:22.811 you're seeing, working memory, 04:22.805 --> 04:26.315 short-term memory, and then long-term memory. 04:26.319 --> 04:29.549 And we talked last class about how things get into sensory 04:29.553 --> 04:32.563 memory, into working memory, the role of attention. 04:32.560 --> 04:34.810 And in fact, the change-blindness studies 04:34.812 --> 04:38.362 are actually just studies of how something gets from your senses 04:38.360 --> 04:41.570 to your consciousness and what does and what doesn't. 04:41.569 --> 04:45.519 Now I want to move to the distinction between working 04:45.515 --> 04:49.305 memory, short-term memory, and long-term memory. 04:49.310 --> 04:55.200 Now, the obvious distinction is actually just in fact--is 04:55.197 --> 04:57.507 storage differences. 04:57.509 --> 05:04.739 So, long-term memory or "LTM" has a huge storage capacity. 05:04.740 --> 05:08.520 This is your memory like the hard drive of your computer. 05:08.519 --> 05:10.589 This is the memory you walk around with. 05:10.589 --> 05:14.459 It includes all the words in English, just for example, 05:14.463 --> 05:15.973 60 to 80,000 words. 05:15.970 --> 05:20.380 It includes everybody you've ever met, languages, 05:20.375 --> 05:24.135 faces, stories, locations, nursery rhymes, 05:24.138 --> 05:26.248 songs, TV programs. 05:26.250 --> 05:28.220 Nobody knows the storage. 05:28.220 --> 05:31.730 It is not true that you remember everything that has 05:31.734 --> 05:33.324 ever happened to you. 05:33.319 --> 05:36.119 There's no reason to believe that this is true. 05:36.120 --> 05:39.660 At the same time though, you have a huge amount stored 05:39.664 --> 05:43.484 in your brain in long-term storage and nobody actually--It 05:43.477 --> 05:46.417 has to be limited because it's a finite, 05:46.420 --> 05:50.480 limited brain. But nobody knows how big it is. 05:50.480 --> 05:55.180 Nobody knows how many terabytes you carry around in your brain 05:55.176 --> 05:56.866 and--but it's a lot. 05:56.870 --> 06:00.880 Compare this to working memory – the short-term memory, 06:00.878 --> 06:03.238 which is actually very limited. 06:03.240 --> 06:06.790 Your memory of what you could store on--in--where you could 06:06.794 --> 06:09.924 hold in consciousness right now is quite limited. 06:09.920 --> 06:12.430 Here is an exercise. 06:12.430 --> 06:14.100 Do not write these things down. 06:14.100 --> 06:15.130 I want you to remember them. 06:15.129 --> 06:28.259 I'm just going to give you a few numbers: 14,59, 06:28.256 --> 06:36.356 11,109, 43,58, 98,487, 25,389, 06:36.355 --> 06:40.345 54. Please write them down. 06:40.350 --> 06:58.940 06:58.940 --> 07:02.830 View this as an IQ test if that would relax you. 07:02.830 --> 07:10.540 07:10.540 --> 07:16.360 How many of you who decided to participate in this experiment 07:16.360 --> 07:18.300 got three or less? 07:18.300 --> 07:20.400 Good. Good. 07:20.399 --> 07:29.099 Four, five, six, seven, eight, 07:29.100 --> 07:36.380 nine or more? Anybody get all eleven? 07:36.379 --> 07:40.879 This is a particularly difficult memory task. 07:40.880 --> 07:42.630 The numbers are meaningless. 07:42.629 --> 07:45.209 And I told--and I forgot to tell you to get your pen and 07:45.210 --> 07:47.180 pencil ready, so some of you just glared at 07:47.181 --> 07:48.611 me. But [laughter] 07:48.608 --> 07:53.238 under normal circumstances the cognitive psychologist George 07:53.235 --> 07:57.625 Miller said that this sort of suggested that the standard 07:57.628 --> 08:01.548 memory storage of short-term memory is seven, 08:01.550 --> 08:02.990 plus or minus two. 08:02.990 --> 08:08.900 And what that means is anywhere from five to nine roughly. 08:08.899 --> 08:11.239 Some of you, I bet, can beat that. 08:11.240 --> 08:16.190 Some of you on a not-so-good day maybe won't make it that 08:16.191 --> 08:18.851 much. Now "seven plus or minus two" 08:18.846 --> 08:21.656 is what you--;so, that's what you hold in 08:21.656 --> 08:25.356 consciousness. I can tell you 14,21. 08:25.360 --> 08:27.880 You walk around, "Oh, yeah, 14,21." 08:27.879 --> 08:29.739 You hold that in consciousness with no problem. 08:29.740 --> 08:32.340 But I throw eleven numbers at you, you can't. 08:32.340 --> 08:33.160 Some dribble out. 08:33.159 --> 08:36.279 You can't hold that in your conscious window in your 08:36.277 --> 08:37.497 short-term memory. 08:37.500 --> 08:43.470 Now, this raises the question "seven plus or minus two" what? 08:43.470 --> 08:48.770 And the answer seems to be what George Miller calls "chunks." 08:48.769 --> 08:54.419 And a chunk is a basic memory unit, something you think of as 08:54.423 --> 08:57.253 a single, individual entity. 08:57.250 --> 09:01.640 So, suppose you see the string of letters "L, 09:01.643 --> 09:03.943 A, M, A, I, S, O, N." 09:03.940 --> 09:08.190 If you don't know--If you can't form these into words and you 09:08.190 --> 09:11.590 have to remember them, these are eight chunks. 09:11.590 --> 09:13.830 You have to just pick them up separately. 09:13.830 --> 09:17.850 On the other hand, if you break them up into four 09:17.846 --> 09:22.026 words you could just remember it as four chunks. 09:22.029 --> 09:26.249 And if you break it up into two words in French, 09:26.246 --> 09:31.266 "la maison," "the house," it could just be one or two. 09:31.269 --> 09:34.879 How much you know depends--affects how much you 09:34.879 --> 09:39.119 memorize--how much you could store in memory because it 09:39.116 --> 09:43.036 affects what counts as a basic unit of memory. 09:43.039 --> 09:45.539 And there's all sorts of examples of this. 09:45.539 --> 09:48.779 If I tell you "1,1, 0,1, 1,0, 0,1, 09:48.783 --> 09:54.483 0,1, 1,0," those of you who don't know binary numbers might 09:54.484 --> 10:00.594 have to remember that as "1, 1,0, 0," whatever I said. 10:00.590 --> 10:04.360 Those of you who are computer scientists or mathematicians or, 10:04.356 --> 10:07.126 for whatever reason, know binary numbers could 10:07.134 --> 10:09.794 convert it into a single binary number. 10:09.790 --> 10:14.130 Anybody know what the number is? 10:14.130 --> 10:17.100 No, I cannot say it again. 10:17.100 --> 10:18.510 [laughter] Some number, 10:18.506 --> 10:22.596 24, or not 24--to some number, 24, and then you remember "24." 10:22.600 --> 10:25.850 It's easier. Suppose you see a chessboard 10:25.845 --> 10:30.345 and the chessboard is set up and you don't know how to play 10:30.350 --> 10:32.410 chess. It is murderously hard to 10:32.406 --> 10:34.186 remember that. They've done the experiments. 10:34.190 --> 10:36.600 They've taken people in a lab who don't know how to play 10:36.599 --> 10:38.179 chess. They set up a chessboard and 10:38.182 --> 10:39.152 then they say, "Okay. 10:39.150 --> 10:41.270 Look at this for five minutes." 10:41.269 --> 10:44.559 Then they take it away, set it up again, 10:44.564 --> 10:46.934 and it's murderously hard. 10:46.929 --> 10:50.719 "There is a horse-y thing on the side there and everything." 10:50.720 --> 10:54.880 But if these chess pieces are set up in some way that's 10:54.877 --> 10:59.237 logical for a chess player, then a chess master could look 10:59.240 --> 11:02.080 at it and remember it in a glance, "Oh. 11:02.080 --> 11:06.430 It's the Fibonacci defense" or something like that [laughs], 11:06.432 --> 11:08.942 and then immediately recover it. 11:08.940 --> 11:12.430 Similarly, football coaches have been tested on their 11:12.425 --> 11:14.565 memories of football diagrams. 11:14.570 --> 11:17.050 And they have a photographic memory for football diagrams 11:17.045 --> 11:19.295 because it corresponds to things that make sense. 11:19.299 --> 11:22.449 Architects could have a photographic memory, 11:22.448 --> 11:26.618 a perfect memory for floor plans because it makes sense to 11:26.621 --> 11:30.771 them. They understand it. 11:30.769 --> 11:34.269 And so the way you store things in memory, and this is a theme 11:34.272 --> 11:37.662 we're going to return to when we get to long-term memory, 11:37.659 --> 11:42.579 depends in a large extent on how much you understand it. 11:42.580 --> 11:46.380 And this shows up in expertise effects. 11:46.379 --> 11:51.049 Now, this is what's happening so far in short-term memory, 11:51.052 --> 11:53.432 how much you hold in there. 11:53.429 --> 11:56.469 The question is how do you get it into long-term memory? 11:56.470 --> 12:00.700 So, you have long-term memory, your major storage system. 12:00.700 --> 12:06.300 How does information get from your consciousness to long-term 12:06.297 --> 12:08.357 storage? Well, there's one 12:08.355 --> 12:12.255 thing--there's one way which sort of works sometimes but not 12:12.259 --> 12:14.969 very well. And it's called "maintenance 12:14.967 --> 12:17.587 rehearsal." Suppose I said you have to 12:17.587 --> 12:20.817 remember this number, this string of numbers. 12:20.820 --> 12:23.870 And if you remember it in twenty minutes you will get one 12:23.866 --> 12:24.896 thousand dollars. 12:24.899 --> 12:29.139 And the string is my phone number when I was a kid. 12:29.139 --> 12:34.459 I'll include the area code: 514-688-9057. 12:34.460 --> 12:36.570 Now, if you tell that to a four-year-old, 12:36.569 --> 12:39.469 well, the four-year-old will say, "I'll remember it." 12:39.470 --> 12:41.970 And then you ask them, "What did I just say?" 12:41.970 --> 12:43.200 "Well, I don't know." 12:43.200 --> 12:46.380 If you tell it to an--because you know something--If a lot 12:46.383 --> 12:49.123 depended on it, you would know to do something. 12:49.120 --> 12:53.020 What you would do is you'd say to yourself, "514-688-9057,51 12:53.018 --> 12:56.778 4-688-9057,514-688 --" You'd rehearse it in your head over 12:56.784 --> 13:00.014 and over again. The problem is you could hold 13:00.007 --> 13:02.107 it as long as you can do that. 13:02.110 --> 13:04.670 It's like these movies. 13:04.669 --> 13:07.749 You see this all the time, like an episode of 24: 13:07.754 --> 13:11.124 "Jack, call CTU and tell them Agent 11 is trapped in a--" And 13:11.120 --> 13:14.430 I can't even remember this but the way to remember it is you 13:14.429 --> 13:17.849 hold--you've just got to repeat it over and over again in your 13:17.850 --> 13:20.760 head. But this will not typically get 13:20.755 --> 13:22.855 things into long-term memory. 13:22.860 --> 13:28.070 To get things into long-term memory, rehearsal is usually not 13:28.068 --> 13:30.858 enough. You need to do other things. 13:30.860 --> 13:38.160 Typically, what you need is structure and organization. 13:38.159 --> 13:41.259 And one way to demonstrate this was in a classic "depth of 13:41.261 --> 13:44.531 processing" experiment which nicely illustrates the fact that 13:44.526 --> 13:46.536 the more you structure something, 13:46.539 --> 13:50.449 the deeper you think about it, the better it gets entrenched 13:50.447 --> 13:52.167 in the long-term memory. 13:52.169 --> 13:55.759 So, in this study what they did was they asked people--they told 13:55.760 --> 13:59.180 people that there's going to be words flashed on a screen. 13:59.179 --> 14:02.669 And all of the subjects saw the same strings of words. 14:02.670 --> 14:04.740 There were forty-eight words. 14:04.740 --> 14:07.440 They were not told to memorize the words. 14:07.440 --> 14:10.970 One third of the subjects was told, "Look. 14:10.970 --> 14:13.880 Some of these words are going to come out in capital letters, 14:13.881 --> 14:15.581 some of them not capital letters. 14:15.580 --> 14:18.160 Press a button for capitals, non-capitals." 14:18.160 --> 14:20.890 "Sure." The other group was told, 14:20.893 --> 14:24.913 "Some of these words will rhyme with 'train,' Others won't. 14:24.909 --> 14:27.019 Press a button if it rhymes in 'train'." 14:27.019 --> 14:30.419 The third group was told, "Does it fit into the sentence 14:30.424 --> 14:33.214 ‘The girl placed the blank on the table'? 14:33.210 --> 14:34.330 Press a button if it does. 14:34.330 --> 14:36.090 Press a button if it doesn't." 14:36.090 --> 14:40.720 Then they were asked as a surprise, "What words did you 14:40.719 --> 14:42.839 see?" And the findings looked like 14:42.843 --> 14:45.413 this. When they were asked to focus 14:45.408 --> 14:49.648 on just what the word looked like, memory was very poor, 14:49.652 --> 14:52.742 the sound better, the meaning better. 14:52.740 --> 14:56.710 If you want to remember something, the best way to 14:56.712 --> 15:01.092 remember it is to give it meaning, to give it sense. 15:01.090 --> 15:04.900 This is illustrated through a very ancient technique, 15:04.903 --> 15:08.423 which is that the way to remember things that are 15:08.423 --> 15:12.823 otherwise arbitrary is to give them some organization through 15:12.823 --> 15:19.793 memory tricks, through vivid imagery or songs 15:19.785 --> 15:24.035 or poetry. And there's a lot of examples 15:24.036 --> 15:26.006 of this. Do you know how to remember 15:26.007 --> 15:28.797 that the hippocampus--There's a part of the brain called the 15:28.800 --> 15:31.510 hippocampus. This is the worst memory trick 15:31.512 --> 15:34.852 ever but it will stick with you for twenty years. 15:34.850 --> 15:37.920 The hippocampus is involved in spatial memory. 15:37.919 --> 15:41.379 It's involved in finding your way around. 15:41.379 --> 15:46.039 Think to yourself, "The way I find my way around 15:46.040 --> 15:49.710 campus is through the hippocampus." 15:49.710 --> 15:51.410 And you think, "Well, that's stupid," but 15:51.408 --> 15:53.908 you'll never forget now that the hippocampus is in charge of 15:53.913 --> 15:55.993 spatial memory. It's going to be all you retain 15:55.985 --> 15:56.715 from this course. 15:56.720 --> 16:03.060 Memory books on how to remember people's names usually try to 16:03.062 --> 16:08.882 exploit this sort of thing when you try to get poetry or 16:08.875 --> 16:12.575 dramatic images. So, the memory books always 16:12.576 --> 16:16.036 typically involve somebody--like you meet somebody with very 16:16.042 --> 16:18.332 spiky hair and they say, "My name is Mr. 16:18.333 --> 16:21.273 Fish" and then you remember--you think of their--of 16:21.270 --> 16:23.620 a big fish impaled on their hair. 16:23.620 --> 16:26.710 And then whenever you see them you remember their name. 16:26.710 --> 16:30.260 It only really works for names like "Fish" but [laughter] 16:30.258 --> 16:33.298 the idea is you try to generate vivid imagery. 16:33.299 --> 16:37.689 When stuck with a situation where you have to remember ten 16:37.694 --> 16:42.014 letters, turn it into a song where--or a dirty poem where 16:42.012 --> 16:45.792 each of the letters is the first words of it. 16:45.789 --> 16:49.439 When having to remember something that seems totally 16:49.437 --> 16:53.587 arbitrary, try to figure out a grand and obscene image that 16:53.585 --> 16:55.655 will come to mind easily. 16:55.659 --> 16:59.079 And this is how--these are one way to get things into memory. 16:59.080 --> 17:02.700 At a deep level though, the way to get things into 17:02.701 --> 17:07.131 memory, and this applies to this course no less than anything 17:07.134 --> 17:09.194 else, is by understanding 17:09.193 --> 17:10.883 the--understanding it. 17:10.880 --> 17:14.910 I'm going to read you something and I want you to try your best 17:14.910 --> 17:16.860 to remember what I tell you. 17:16.859 --> 17:18.999 These are not going to be strings of numbers. 17:19.000 --> 17:22.700 These are going to be--This is going to be a series of 17:22.699 --> 17:26.329 sentences: "A newspaper is better than a magazine. 17:26.329 --> 17:28.769 A seashore is a better place than the street. 17:28.769 --> 17:31.179 At first it is better to run than to walk. 17:31.180 --> 17:32.730 You may have to try several times. 17:32.730 --> 17:34.570 It takes some skill but it's easy to learn. 17:34.569 --> 17:36.429 Even young children can enjoy it. 17:36.430 --> 17:39.520 Once successful, complications are minimal. 17:39.520 --> 17:41.950 Birds seldom get too close. 17:41.950 --> 17:45.060 Rain, however, soaks in very fast. 17:45.059 --> 17:48.019 Too many people doing the same things can also cause problems. 17:48.020 --> 17:49.740 One needs lots of room. 17:49.740 --> 17:52.910 If there are no complications, it can be very peaceful. 17:52.910 --> 17:55.100 Finally, a rock will serve as an anchor. 17:55.099 --> 17:59.969 If things break loose from it, however, you will not have a 17:59.971 --> 18:05.131 second chance." And here is what I said . 18:05.130 --> 18:10.030 This is murderously hard to remember. 18:10.030 --> 18:17.770 Now try it. Knowing what this is about, 18:17.769 --> 18:25.759 being able to put a context to it helps the memory and helps it 18:25.758 --> 18:29.968 come to mind. [laughter] Okay. 18:29.970 --> 18:32.820 So, this is about how to get memory--how to get information 18:32.824 --> 18:33.764 into your memory. 18:33.760 --> 18:36.530 How do you get information out? 18:36.530 --> 18:39.470 So, it's exam period. 18:39.470 --> 18:41.970 You got the stuff presumably into your head. 18:41.970 --> 18:43.080 You have to get it out. 18:43.080 --> 18:44.370 You have to retrieve it. 18:44.370 --> 18:46.750 There is a court case. 18:46.750 --> 18:50.470 You have to figure out--You have to recount the crime that 18:50.466 --> 18:53.746 you witnessed. You see somebody and you want 18:53.754 --> 18:55.694 to know his or her name. 18:55.690 --> 18:58.620 And you heard it; you just have to get it out. 18:58.620 --> 19:00.410 Well, how do you do that? 19:00.410 --> 19:02.600 Well, there's "retrieval cues." 19:02.600 --> 19:04.300 Retrieval cues make sense. 19:04.299 --> 19:10.319 Retrieval cues are just things that have been associated with 19:10.323 --> 19:14.643 what you--what you're trying to remember. 19:14.640 --> 19:16.980 If I have to remember to replace the windows, 19:16.979 --> 19:19.959 when I walk in to my living room and see that a window is 19:19.956 --> 19:22.876 cracked that will remind me to replace the windows. 19:22.880 --> 19:26.530 If I had a lunch date with you and forgot about it, 19:26.527 --> 19:28.567 when I see you, "Oh, yeah. 19:28.569 --> 19:31.109 We were supposed to get together to have lunch." 19:31.109 --> 19:36.549 Retrieval cues bring things back but it's a little bit more 19:36.549 --> 19:38.799 complicated than that. 19:38.799 --> 19:41.919 There's a more general relationship between encoding 19:41.918 --> 19:45.158 and retrieval called the "compatibility principle." 19:45.160 --> 19:49.280 And what this means is you're much better to remember 19:49.281 --> 19:53.721 something in the context in which you have learned it. 19:53.720 --> 19:57.160 And this is also known as "context-dependent memory" and 19:57.160 --> 19:58.850 "state-dependent memory." 19:58.849 --> 20:02.659 It's illustrated by one of the strangest experiments in the 20:02.658 --> 20:06.658 history of psychology where they had people on a boat and then 20:06.663 --> 20:09.293 they had them scuba dive underwater. 20:09.289 --> 20:12.729 And they taught them things either on the boat or underwater 20:12.726 --> 20:14.586 with things that they held up. 20:14.590 --> 20:17.040 And then they tested them later. 20:17.039 --> 20:20.629 And it turns out that you'll remember it better if you're 20:20.629 --> 20:24.089 tested on it in the context in which you learned it. 20:24.089 --> 20:27.239 And it might be because then the retrieval cues help bring it 20:27.242 --> 20:30.322 back. But it's more general than that. 20:30.319 --> 20:33.519 If you have to remember something you learned in this 20:33.516 --> 20:37.326 class, you will do better if you try to think about the room in 20:37.328 --> 20:39.048 which you learned it in. 20:39.049 --> 20:42.979 You will do better on your final exam if you were to take 20:42.976 --> 20:47.036 it in this room than if you were to take it in another room 20:47.043 --> 20:50.973 because being in this room will bring back the cues. 20:50.970 --> 20:53.280 It's not just the environment. 20:53.279 --> 20:57.079 People who learn things when they're stoned remember them 20:57.081 --> 21:01.091 better--keeping stoned at a sort of a low-level that doesn't 21:01.085 --> 21:04.615 disrupt other mental activities--remember them better 21:04.615 --> 21:07.525 when they're sort of stoned again [laughter] 21:07.534 --> 21:09.914 than if they're non-stoned. 21:09.910 --> 21:13.510 Similarly--So, if you study while you drink 21:13.510 --> 21:18.570 you should tipple a little bit before coming in to the Final 21:18.568 --> 21:19.838 exam. [laughter] 21:19.839 --> 21:22.789 It's sort of like the "what happens in Vegas stays in Vegas" 21:22.789 --> 21:24.919 sort of result. And so, similarly, 21:24.916 --> 21:28.936 it even applies to moods in that if you learn something when 21:28.937 --> 21:32.687 depressed you have a slightly better recovery of it when 21:32.685 --> 21:36.365 you're in that same mood of depression than when you're 21:36.365 --> 21:39.645 elated. And the idea is that part of 21:39.651 --> 21:44.561 what memory is--part of what recovering memory is is getting 21:44.563 --> 21:49.063 back your original context in which you learned it. 21:49.059 --> 21:54.279 "Elaborative rehearsal" and retrieval involves the 21:54.282 --> 21:58.442 connections between different things. 21:58.440 --> 22:03.100 Elaborative rehearsal is that the more you think about 22:03.102 --> 22:06.712 something the easier it is to remember. 22:06.710 --> 22:09.540 If you have to think about--If you have to remember something, 22:09.540 --> 22:11.860 try to connect it to as many things as possible. 22:11.860 --> 22:13.310 Think of an image. 22:13.310 --> 22:14.660 Make a joke out of it. 22:14.660 --> 22:18.270 Imagine how you would explain it to somebody else. 22:18.269 --> 22:22.399 Imagine how the world--what the world would be like if it wasn't 22:22.402 --> 22:24.522 so. And the idea is that this sort 22:24.521 --> 22:27.921 of thinking about it makes connections in your memory from 22:27.924 --> 22:30.974 that thing you have to learn, to other memories. 22:30.970 --> 22:33.360 And so it makes it easier to recover. 22:33.359 --> 22:39.769 "Elaborative retrieval" refers to a finding that when you want 22:39.771 --> 22:45.761 to get something back out of memory people tend to give up 22:45.763 --> 22:49.483 too soon. It turns out that there's a lot 22:49.475 --> 22:54.085 of stuff that's in your memory but it needs work to extract; 22:54.089 --> 22:58.649 it needs various sort of searching strategies. 22:58.650 --> 23:03.530 One study asked people who were considerably older than you to 23:03.529 --> 23:06.729 remember their high school classmates. 23:06.730 --> 23:09.710 And in the first pass people were terrible. 23:09.710 --> 23:11.540 Maybe they had a couple of friends they kept in touch with. 23:11.540 --> 23:14.370 Otherwise, pretty bad. 23:14.369 --> 23:16.999 And this is a good experiment because you could use high 23:16.996 --> 23:19.426 school yearbooks to judge whether or not they get it 23:19.432 --> 23:21.412 right. But then what you do is you 23:21.413 --> 23:22.753 tell the person, "Look. 23:22.750 --> 23:26.870 Keep trying. Were you--What sort of--Who was 23:26.867 --> 23:28.947 your teacher? What sort of clubs did you 23:28.946 --> 23:30.646 belong to? What sort of sports you--did 23:30.654 --> 23:31.604 you participate in? 23:31.600 --> 23:33.460 How did you get to school? 23:33.460 --> 23:35.450 How did you get from school? 23:35.450 --> 23:37.770 What did you do during lunch? 23:37.770 --> 23:39.120 What did you do during break?" 23:39.119 --> 23:41.679 And you keep ask--"Do you know--have any friends whose 23:41.678 --> 23:44.428 letter--whose last name began with ‘B,' with ‘C,' with 23:44.429 --> 23:46.229 ‘D'?" And you keep pushing and 23:46.234 --> 23:47.354 pushing and pushing. 23:47.349 --> 23:50.689 And over the span of time things come back. 23:50.690 --> 23:54.440 Again, it's not true that you never forget. 23:54.440 --> 23:56.910 There is honest to God forgetting but sometimes you 23:56.910 --> 23:59.680 think you forget and it's because you haven't looked long 23:59.676 --> 24:02.206 enough. There's a real physical notion 24:02.213 --> 24:04.543 of searching for the right answer. 24:04.540 --> 24:07.740 We've talked about retrieval. 24:07.740 --> 24:11.530 Oh. Every class I've given somebody 24:11.532 --> 24:16.742 asks either in class or by e-mail what about déjà vu? 24:16.740 --> 24:22.440 And déjà vu is a feeling that an event has happened before. 24:22.440 --> 24:25.190 So, you're looking at me and I'm lecturing and you say, 24:25.193 --> 24:26.523 "I've heard this before. 24:26.520 --> 24:28.660 I know this before." 24:28.660 --> 24:31.280 You see somebody and say, "I've been in this situation 24:31.276 --> 24:33.716 before." This is not evidence for 24:33.719 --> 24:35.749 psychic powers, [laughter] 24:35.750 --> 24:40.220 which many people say it is, but nobody really knows why 24:40.218 --> 24:43.338 this exists. We know, and this is a clue, 24:43.340 --> 24:45.640 it's worse with frontal lobe damage. 24:45.640 --> 24:48.120 If you get damage to this part of the brain, 24:48.115 --> 24:50.585 you get a lot more déjà vu experiences. 24:50.589 --> 24:54.829 I asked some experts in memory, including Marcia Johnson who is 24:54.826 --> 24:58.446 chair of our department, what the best explanation for 24:58.448 --> 25:01.608 déjà vu is. And the answer she gave, 25:01.607 --> 25:05.017 the--say one big theory, goes like this. 25:05.019 --> 25:08.379 Déjà vu is a feeling that it's happened before. 25:08.380 --> 25:11.060 The answer is it has happened before. 25:11.060 --> 25:14.090 It's happened half a second ago. 25:14.089 --> 25:19.289 And so what happens is sometimes there is a glitch, 25:19.292 --> 25:22.312 a disturbance in the force. 25:22.310 --> 25:25.080 I don't know. There's a glitch [laughter] 25:25.084 --> 25:28.584 and you are talking and then something happens to you and you 25:28.582 --> 25:30.042 put it in your memory. 25:30.039 --> 25:33.619 But it's as if you don't put the stamp on it of what time and 25:33.615 --> 25:36.225 what date. So, you're talking to me and 25:36.230 --> 25:40.440 then you store it in memory but you don't store it in memory as 25:40.438 --> 25:41.998 happening right now. 25:42.000 --> 25:44.880 Then half a second goes by and you're talking to me and you 25:44.881 --> 25:46.671 say, "This is strangely familiar." 25:46.670 --> 25:51.650 And that's one theory of what goes on in déjà vu. 25:51.650 --> 25:54.000 Okay. So far, there's the sort of 25:54.003 --> 25:57.713 good news – remembering – but then there's bad news – 25:57.706 --> 26:00.106 forgetting. How many people can remember, 26:00.111 --> 26:02.931 without looking down at your notes, at least two of the 26:02.926 --> 26:04.486 numbers I gave you earlier? 26:04.490 --> 26:08.730 26:08.730 --> 26:12.050 How many people can remember at least four? 26:12.050 --> 26:16.110 Oh, impressive. If I asked you in an hour, 26:16.109 --> 26:18.429 the number would go down. 26:18.430 --> 26:21.040 These are sort of statistics in a similar experiment. 26:21.039 --> 26:24.799 And this graph illustrates that people forget. 26:24.800 --> 26:27.930 Over time, you'll forget. 26:27.930 --> 26:29.480 Why do you forget? 26:29.480 --> 26:31.520 Why is there forgetting at all? 26:31.519 --> 26:33.519 Well, there's different explanations for this. 26:33.519 --> 26:38.049 One explanation is your brain's a physical thing, 26:38.050 --> 26:43.430 it's a physical piece of meat, and it kind of goes bad. 26:43.430 --> 26:45.170 Physical things decay. 26:45.170 --> 26:49.210 And so, the memory traces that are laid onto your brain will 26:49.206 --> 26:50.776 just decay over time. 26:50.780 --> 26:54.200 A second answer is interference. 26:54.200 --> 26:56.050 So, remember those numbers? 26:56.049 --> 27:01.079 Here's a few more: 114,81, 66,42. 27:01.079 --> 27:04.949 Well, the more information that comes in that's similar to the 27:04.947 --> 27:08.617 stuff you're trying to remember, it blocks your recovery of 27:08.624 --> 27:10.214 original information. 27:10.210 --> 27:14.510 So your ability to remember something can be impaired by 27:14.510 --> 27:19.200 learning more things which are related to it because they get 27:19.201 --> 27:21.001 confused in memory. 27:21.000 --> 27:24.660 Finally, and maybe this is most interesting, there are changes 27:24.660 --> 27:25.860 in retrieval cues. 27:25.859 --> 27:29.189 So, the more time goes by the more the world changes. 27:29.190 --> 27:32.400 And if your memory is to some extent dependent on cues 27:32.396 --> 27:37.256 bringing it back to life, then the change in retrieval 27:37.259 --> 27:43.949 cues can make it more difficult to recall certain things. 27:43.950 --> 27:48.820 This leads to a puzzle where there's considerable scientific 27:48.824 --> 27:52.464 debate over the case of childhood amnesia. 27:52.460 --> 27:56.290 And the case of childhood amnesia is--doesn't refer to 27:56.293 --> 27:59.913 when a child gets brain damage and gets amnesia. 27:59.910 --> 28:04.280 What it refers to is people have a difficult time recovering 28:04.280 --> 28:05.910 very early memories. 28:05.910 --> 28:10.530 I want people to just take a second and try to think back on 28:10.528 --> 28:14.988 what your first memory is and roughly how old you were. 28:14.990 --> 28:20.660 28:20.660 --> 28:25.750 How many people don't think you have a first memory until you 28:25.750 --> 28:28.890 were about five years old or older? 28:28.890 --> 28:33.220 Okay. How many people think you have 28:33.220 --> 28:37.510 the first memory of around age four or younger? 28:37.509 --> 28:41.649 How many people think you have the first memory of around age 28:41.649 --> 28:42.959 three or younger? 28:42.960 --> 28:47.150 Two or younger? How many of you think you have 28:47.146 --> 28:50.516 the first memory when you were about one years old or younger? 28:50.519 --> 28:54.729 And I'm not asking about past lives but that [laughs] 28:54.729 --> 28:56.429 happened last year. 28:56.430 --> 29:01.070 How old is your--roughly your first memory do you think? 29:01.070 --> 29:03.470 How old? Student: 29:03.474 --> 29:04.534 Between one and two. 29:04.529 --> 29:05.829 Professor Paul Bloom: Between one and two? 29:05.829 --> 29:08.919 Anybody think they could beat that? 29:08.920 --> 29:11.200 Same guy? Yeah. 29:11.200 --> 29:12.390 Student: One. 29:12.390 --> 29:16.460 Professor Paul Bloom: One. 29:16.460 --> 29:20.130 [laughter] Anybody else? 29:20.130 --> 29:24.680 The literature is unclear on this because it's very difficult 29:24.678 --> 29:28.998 to test people's recollections of their first memories. 29:29.000 --> 29:31.640 If I'm to ask people about their first memories, 29:31.639 --> 29:33.379 they'll often say, "Oh, yeah. 29:33.380 --> 29:37.990 I remember I was in this room and there was a crib and I'm 29:37.993 --> 29:40.743 going ‘Ga ga, goo goo' [laughter] 29:40.744 --> 29:42.934 and I was on the potty. 29:42.930 --> 29:44.950 I was walking. I was so cute. 29:44.950 --> 29:48.280 I remember it." It's very difficult to tell 29:48.280 --> 29:51.790 and, as we'll discuss in some detail, there are a lot of 29:51.785 --> 29:55.795 reasons to distrust people when they--not that they're lying but 29:55.800 --> 29:59.050 to distrust the accuracy of people's memories. 29:59.049 --> 30:04.229 We also know from studies about trauma where people have 30:04.226 --> 30:08.646 terrible experiences when they're one or two. 30:08.650 --> 30:11.890 Typically, this trauma is not remembered later on. 30:11.890 --> 30:16.020 People know of trauma because they're told about it but they 30:16.019 --> 30:19.379 don't typically remember it with any accuracy. 30:19.380 --> 30:23.150 Even children--older children don't remember back beyond that 30:23.152 --> 30:25.422 age. Nobody knows why childhood 30:25.418 --> 30:28.298 amnesia occurs. Nobody knows why it's very 30:28.295 --> 30:32.315 difficult to recover memories before about the age of three. 30:32.319 --> 30:36.299 One theory is that the retrieval cues change radically. 30:36.299 --> 30:39.339 I had a friend of mine who's a clinical psychologist and he 30:39.339 --> 30:42.379 suggested a new form of therapy where they make these giant 30:42.379 --> 30:45.579 tables and chairs and then they bring you in to the office and 30:45.577 --> 30:48.457 you're standing there with these giant tables and chairs 30:48.459 --> 30:50.609 [laughter] and all these memories of being 30:50.608 --> 30:52.808 a baby would come flooding back. 30:52.809 --> 30:56.159 [laughter] And he dropped out of the field 30:56.155 --> 30:59.745 and-- [laughter] Really, but it's such a cool 30:59.745 --> 31:02.895 idea. Some people think language is 31:02.896 --> 31:05.536 to blame. So a child, a baby, 31:05.539 --> 31:09.919 starts out with no spoken or signed language. 31:09.920 --> 31:13.010 Language comes to be learned at around one, two, 31:13.007 --> 31:16.747 and three, and it might be that the learning of a language 31:16.751 --> 31:18.461 reformats your memory. 31:18.460 --> 31:22.160 And once the memory is reformatted it can't go back to 31:22.164 --> 31:25.664 the previous state prior to language in the end. 31:25.660 --> 31:28.510 It could be neural maturation. 31:28.509 --> 31:32.799 It could be that those memory parts of the brains grow around 31:32.802 --> 31:36.882 age two or three that just weren't there prior to that. 31:36.880 --> 31:39.050 And nobody really knows. 31:39.049 --> 31:44.519 It's a fascinating research area why--about memory changes 31:44.523 --> 31:48.363 early on. Another case of memory failure 31:48.356 --> 31:52.086 is brain damage. And brain damage comes in a 31:52.089 --> 31:53.609 couple of flavors. 31:53.610 --> 31:58.510 There is retrograde amnesia; "retro" for past. 31:58.509 --> 32:02.599 Retrograde amnesia is when you lose some memory of the past. 32:02.599 --> 32:05.889 This could be in a case where you get some sort of head trauma 32:05.885 --> 32:08.735 and you lose memory of your entire episodic memory. 32:08.740 --> 32:11.940 But typically, if you have any sort of serious 32:11.935 --> 32:16.055 accident that involves you losing consciousness you'll have 32:16.055 --> 32:19.175 a blackout of some period prior to that, 32:19.180 --> 32:20.780 say, blow to the head. 32:20.779 --> 32:25.489 And the reason for this is as you're having these experiences 32:25.494 --> 32:30.214 now they need to kind of get consolidated into your brain. 32:30.210 --> 32:33.320 Your brain needs to rewire and catch up to the experiences 32:33.317 --> 32:36.117 you're having. A sudden blow to the head will 32:36.121 --> 32:40.051 knock you unconscious and then the memories that have happened 32:40.050 --> 32:44.040 immediately prior will not get consolidated and they'll be lost 32:44.042 --> 32:46.712 forever. Another sort of memory is 32:46.710 --> 32:51.100 anterograde amnesia and this was the case of--This happens in 32:51.102 --> 32:52.862 Korsakoff's syndrome. 32:52.859 --> 32:56.749 It happens to a very famous patient known as H.M. 32:56.750 --> 32:59.490 who actually lives in Hartford, Connecticut. 32:59.490 --> 33:04.440 And it happened to Clive Wearing, the film you saw last 33:04.437 --> 33:07.607 class. And this sort of amnesia is a 33:07.610 --> 33:12.070 sort of amnesia where you lose the ability to form new 33:12.070 --> 33:15.270 memories. And so you live in a perpetual 33:15.273 --> 33:18.493 present, unable to accumulate new memories. 33:18.490 --> 33:23.720 But it's actually a little bit more complicated than that. 33:23.720 --> 33:27.110 What happens is--And this was an exciting discovery about 33:27.113 --> 33:30.573 these patients that led to some real insights about normal 33:30.566 --> 33:33.896 memory--What happens is--And this is the brain damage in 33:33.899 --> 33:36.539 these cases, the temporal lobe and the 33:36.541 --> 33:40.081 hippocampus, very useful for spatial memory you'll know. 33:40.079 --> 33:44.359 One discovery made about people who couldn't form new memories 33:44.359 --> 33:48.639 is that they could form new memories, but of certain types. 33:48.640 --> 33:53.120 So for example, this is a task here involving 33:53.120 --> 33:58.110 filling in a star while looking in to a mirror. 33:58.109 --> 34:00.219 And if I asked you to do it you'd find it pretty difficult. 34:00.220 --> 34:02.180 It's just kind of difficult to do. 34:02.180 --> 34:03.520 You'd be clumsy at it. 34:03.519 --> 34:07.029 You bring in an amnesic who can't form new memories and you 34:07.033 --> 34:09.793 say, "Hey. I want you to try something new. 34:09.789 --> 34:11.529 I want you to try this star game." 34:11.530 --> 34:14.360 He'd say, "Okay. I've never seen it before but 34:14.356 --> 34:15.526 I'll do it." Tries it. 34:15.530 --> 34:19.090 Does very badly. You bring him in and over and 34:19.088 --> 34:23.368 over again--Each time he does it he starts off by saying, 34:23.372 --> 34:25.822 "I've never seen this before. 34:25.820 --> 34:29.740 I'll--I'm sure I'll give it a try" but he gets better and 34:29.739 --> 34:32.969 better at it. And this is known as implicit 34:32.968 --> 34:35.618 memory. The claim is that in these 34:35.619 --> 34:39.799 sorts of cases you lose the abilities to form explicit 34:39.796 --> 34:44.836 conscious memories that you're aware of, that you understand. 34:44.840 --> 34:49.650 But some sorts of memories persist and you are able to form 34:49.649 --> 34:51.999 them. This has actually been 34:52.002 --> 34:55.392 illustrated in a couple of dramatic movies, 34:55.385 --> 34:59.005 one of them a very bad dramatic movie [laughs] 34:59.010 --> 35:04.000 where Drew Barrymore loses the ability to form new memories and 35:04.004 --> 35:07.794 somehow falls in love with Adam Sandler. 35:07.789 --> 35:10.199 [laughs] Definitely don't watch that. 35:10.199 --> 35:14.809 But a very good movie called "Memento," which is about a 35:14.805 --> 35:19.575 character who loses his ability to form new memories while 35:19.579 --> 35:23.179 trying to track down his wife's killer. 35:23.179 --> 35:26.489 "Memento" is a movie which is fascinating because it's told 35:26.494 --> 35:28.644 backwards. But throughout "Memento" 35:28.643 --> 35:30.903 there's another story told forwards. 35:30.900 --> 35:35.430 And I like this story because it very dramatically illustrates 35:35.433 --> 35:39.893 what does, and what is and is not impaired in cases of severe 35:39.892 --> 35:42.522 memory damage. So, I'm going to show you a 35:42.522 --> 35:45.112 couple of clips that illustrate the disassociation from 35:45.107 --> 35:45.727 "Memento." 35:45.730 --> 35:49.390 35:49.389 --> 35:53.209 Those of you who have seen the movie know that this ends up 35:53.210 --> 35:55.120 quite tragically for Sammy. 35:55.120 --> 35:58.080 I highly recommend the movie. 35:58.079 --> 36:02.049 We've dealt right now with two sorts of failures of memory. 36:02.050 --> 36:06.140 One is everyday failure of memory when you forget. 36:06.139 --> 36:13.549 How many of you remember three or more of the numbers I 36:13.545 --> 36:16.695 originally presented? 36:16.700 --> 36:19.330 Yeah? Go ahead. 36:19.329 --> 36:22.269 Student: [inaudible] Professor Paul Bloom: 36:22.268 --> 36:25.868 Fourteen, 59,11. Is that right? 36:25.870 --> 36:30.970 [laughs] Fine. [laughter] All right. 36:30.969 --> 36:35.739 I'm going to ask you again in a month. 36:35.739 --> 36:37.869 [laughter] Well, people are supposed to 36:37.865 --> 36:40.485 forget [laughter] and some things will--you will 36:40.494 --> 36:42.584 forget. That's normal forgetting. 36:42.579 --> 36:46.949 A second case is forgetting due to brain damage. 36:46.949 --> 36:50.999 Forgetting due to brain damage is exotic and unusual but it's 36:51.003 --> 36:54.453 interesting in that it illustrates some more general 36:54.448 --> 36:56.878 themes about how the mind works. 36:56.880 --> 37:00.730 Remember one theme of this course is we're going to look at 37:00.733 --> 37:03.793 exotic cases like the case of Clive Wearing, 37:03.789 --> 37:07.019 not just because they're interesting in their own right 37:07.023 --> 37:10.263 but sometimes by looking at the extremes we could learn 37:10.256 --> 37:13.066 something about how normal people's normal, 37:13.070 --> 37:16.120 intact minds and brains work. 37:16.119 --> 37:22.609 The third case of forgetting is more interesting and it 37:22.610 --> 37:27.660 actually--Well, I want to do a little trial 37:27.658 --> 37:31.128 here. What I want to do is I want 37:31.133 --> 37:36.193 to--You to listen to three children describe an event that 37:36.186 --> 37:38.716 happened. I want you to come to 37:38.724 --> 37:40.274 some--your own guess. 37:40.269 --> 37:44.019 Imagine you were a judge, you were a childcare worker, 37:44.015 --> 37:48.105 you wanted to see--I want you to be--come to your own guess 37:48.114 --> 37:51.934 about who you believe and what you think happened. 37:51.929 --> 37:52.749 [inaudible] [laughter] 37:52.749 --> 37:53.919 You've heard three children. 37:53.920 --> 37:55.540 Who do you believe? 37:55.539 --> 37:58.679 Who believes--There's three of them, one, two, 37:58.681 --> 38:02.801 three. Who believes the first one? 38:02.800 --> 38:06.580 Who believes the second one? 38:06.580 --> 38:09.200 Who believes the third one? 38:09.200 --> 38:10.580 Sort of an even split. 38:10.580 --> 38:18.540 38:18.539 --> 38:23.329 Twenty-three hundred experts were shown these films and asked 38:23.333 --> 38:27.413 about the different actions, whether or not the person 38:27.413 --> 38:29.773 ripped the book, messed up the bear, 38:29.770 --> 38:33.640 tossed the book in the air and, as you could see, 38:33.643 --> 38:36.813 the majority thought that he did. 38:36.809 --> 38:43.449 This is work done by Steve Ceci who was gracious enough to lend 38:43.446 --> 38:48.046 me the film to use for teaching purposes. 38:48.050 --> 38:50.780 It turns out the second girl was right. 38:50.780 --> 38:54.950 Absolutely nothing happened. 38:54.949 --> 38:56.519 [laughter] The teacher said, 38:56.523 --> 39:00.083 "There is somebody named Sam Stone who's going to come in." 39:00.079 --> 39:07.829 A guy walks in and says, "Hi," walks around and leaves. 39:07.829 --> 39:10.039 [laughter] The first and third children 39:10.037 --> 39:13.517 had their memories implanted, not through any sort of science 39:13.524 --> 39:16.064 fiction means. They had their memories 39:16.055 --> 39:18.565 implanted--Well, they had their memories 39:18.574 --> 39:20.064 implanted like this. 39:20.059 --> 39:23.139 Some of the children would just ask questions. 39:23.139 --> 39:26.159 The interviewer, by the way, was herself unaware 39:26.157 --> 39:29.367 of what happened so the interviewer was a perfectly 39:29.367 --> 39:30.777 naive interviewer. 39:30.780 --> 39:33.570 And it turns out if you just interview children and you ask 39:33.574 --> 39:35.894 them questions about whether the book was ripped, 39:35.888 --> 39:36.898 "Did you see him? 39:36.900 --> 39:38.460 Did he really do it?" 39:38.460 --> 39:40.400 they don't say anything. 39:40.400 --> 39:43.760 They didn't see anything and they won't say anything. 39:43.760 --> 39:47.770 Other children were told about Sam Stone. 39:47.769 --> 39:51.319 They were told a stereotype about Sam Stone – that he's 39:51.321 --> 39:54.941 very clumsy and he tends to rip things and he trips and he 39:54.936 --> 39:57.406 breaks things and he spills things. 39:57.409 --> 39:59.239 And in fact, the third child mentioned that 39:59.242 --> 40:02.052 in passing. He said, "He always does that." 40:02.050 --> 40:06.130 Just knowing this about Sam Stone tends to raise the 40:06.130 --> 40:09.330 proportion of kids who say, for instance, 40:09.329 --> 40:11.569 that he ripped the book. 40:11.570 --> 40:15.210 Other children were given suggestions. 40:15.210 --> 40:17.280 They were given suggestive questioning. 40:17.280 --> 40:21.350 They were a series of leading questions like, 40:21.345 --> 40:23.245 "Oh. Sam Stone came in? 40:23.250 --> 40:25.680 Did he rip a book while he was there?" 40:25.679 --> 40:29.549 And still more children got both. 40:29.550 --> 40:33.070 And in fact, the children you saw were from 40:33.070 --> 40:35.380 this group. They heard Sam Stone being 40:35.376 --> 40:38.136 described as a clumsy fellow and they were given a series of 40:38.136 --> 40:39.396 suggestive questionings. 40:39.400 --> 40:42.530 In this condition they were given several suggestive 40:42.526 --> 40:45.526 questionings over the period of several months. 40:45.530 --> 40:48.800 These children, like the first child and the 40:48.796 --> 40:50.996 third child, are not lying. 40:51.000 --> 40:56.480 They honestly believe that Sam Stone came in and did these 40:56.484 --> 40:59.204 things. Also they believe it and 40:59.197 --> 41:03.197 they're so convincing in their belief that experts, 41:03.199 --> 41:08.409 including police officers and child caseworkers and judges and 41:08.413 --> 41:13.373 lawyers, find these children to be extremely believable. 41:13.369 --> 41:16.689 And I think they probably find them to be extremely believable 41:16.687 --> 41:18.697 because the children are not lying. 41:18.699 --> 41:21.439 They really believe they saw what they saw. 41:21.440 --> 41:23.890 But these memories were implanted. 41:23.889 --> 41:26.949 And Ceci, and many other investigators, 41:26.953 --> 41:31.153 study how memories can be implanted in people's minds 41:31.146 --> 41:35.416 through suggestion and through leading questions. 41:35.420 --> 41:40.060 It turns out that the same sort of experiments and the same sort 41:40.064 --> 41:43.974 of research has been done with considerable success in 41:43.971 --> 41:46.921 implanting false memories in adults. 41:46.920 --> 41:51.370 There are dramatic cases of people remembering terrible 41:51.365 --> 41:54.905 crimes and confessing to them when actually, 41:54.905 --> 41:57.205 they didn't commit them. 41:57.210 --> 42:02.610 And this is not because they are lying. 42:02.610 --> 42:05.470 It's not even because they're, in some obvious sense, 42:05.469 --> 42:07.779 deranged or schizophrenic or delusional. 42:07.780 --> 42:12.300 Rather, they have persuaded themselves, or more often been 42:12.302 --> 42:16.352 persuaded by others, that these things have actually 42:16.348 --> 42:19.208 happened. Psychologists have studied in 42:19.212 --> 42:22.752 the laboratory how one could do this, how one can implant 42:22.750 --> 42:24.520 memories in other people. 42:24.519 --> 42:26.739 And some things are sort of standard. 42:26.739 --> 42:31.459 Suppose I was to tell you a story about a trip I took to the 42:31.460 --> 42:36.100 dentist or a visit I took to--or a time when I ate out at a 42:36.099 --> 42:39.859 restaurant and I'm to omit certain details. 42:39.860 --> 42:42.650 I omit the fact that I paid the bill in a restaurant, 42:42.652 --> 42:45.662 let's say or I finished the meal and then I went home. 42:45.659 --> 42:48.939 Still, you will tend to fill in the blanks. 42:48.940 --> 42:52.330 You'll tend to fill in the blanks with things you know. 42:52.329 --> 42:54.069 So, you might remember this later saying, 42:54.071 --> 42:56.021 "Okay. He told me he finished eating, 42:56.022 --> 42:58.962 paid the bill and left," because paying the bill is what 42:58.960 --> 43:00.350 you do in a restaurant. 43:00.350 --> 43:01.710 This is benign enough. 43:01.710 --> 43:03.140 You fill in the blanks. 43:03.139 --> 43:11.739 You also can integrate suppositions made by others. 43:11.739 --> 43:15.279 And the clearest case of this is eyewitness testimony. 43:15.280 --> 43:18.650 And the best research on this has been done by Elizabeth 43:18.653 --> 43:21.233 Loftus who has done a series of studies, 43:21.230 --> 43:25.600 some discussed in the textbook, showing how people's memories 43:25.597 --> 43:28.287 can be swayed by leading questions. 43:28.290 --> 43:30.780 And it can be extremely subtle. 43:30.780 --> 43:34.370 In one experiment, the person was just asked in 43:34.371 --> 43:38.661 the course of a series of questions--shown a scene where 43:38.664 --> 43:42.104 there's a car accident and asked either, 43:42.099 --> 43:44.519 "Did you see a broken headlight?" 43:44.519 --> 43:48.439 or "Did you see the broken headlight?" 43:48.440 --> 43:52.070 The ‘the' presupposes that there was a broken headlight and 43:52.067 --> 43:55.387 in fact, the people told--asked, "Did you see the broken 43:55.393 --> 43:58.253 headlight?" later on are more likely to 43:58.249 --> 44:01.099 remember one. It creates an image and they 44:01.101 --> 44:03.241 fill it in. In another study, 44:03.244 --> 44:06.514 she would show film segments and then ask, 44:06.509 --> 44:10.969 "Did you see the children getting on the school bus?" 44:10.969 --> 44:13.829 Now, there was no school bus but people who hear that 44:13.832 --> 44:16.972 question later on when asked, "Did you see a school bus in 44:16.969 --> 44:19.149 the film?" are more likely to say yes. 44:19.150 --> 44:22.290 In another study, she would show people film 44:22.288 --> 44:26.518 segments and ask them either, "How fast were the cars going 44:26.520 --> 44:28.710 when they hit each other?" 44:28.710 --> 44:34.730 or "How fast were the cars going when they smashed into 44:34.729 --> 44:38.509 each other?" A week later she'd bring people 44:38.513 --> 44:42.603 back in to the laboratory and ask, "Did you see any broken 44:42.602 --> 44:45.292 glass?" Those who hear a smash tend to 44:45.285 --> 44:49.345 see the broken glass more than those who hear a hit because the 44:49.346 --> 44:51.766 question has changed their memory, 44:51.769 --> 44:56.339 making it more of a dramatic event. 44:56.340 --> 45:01.330 Hypnosis is the clearest case where there's a sort of 45:01.328 --> 45:06.608 reconstructive effort led by--led as a result of leading 45:06.605 --> 45:09.095 and probing questions. 45:09.099 --> 45:14.119 Some of you are readily hypnotizable and you can be 45:14.123 --> 45:17.423 hypnotized. And what we would learn about a 45:17.418 --> 45:21.198 past event from hypnotizing you will not necessarily then be 45:21.202 --> 45:23.642 inaccurate. What hypnotizing does is it 45:23.635 --> 45:25.905 makes people very willing to cooperate. 45:25.909 --> 45:28.299 Unfortunately, it isn't as if there is a 45:28.297 --> 45:31.967 memory storage there where you could just go through and look 45:31.971 --> 45:34.421 as in the movies where you just say, 45:34.420 --> 45:35.530 "What's the license plate?" 45:35.530 --> 45:38.250 The person's hypnotized and then the flashback comes in and 45:38.254 --> 45:40.184 then they zoom in on the license plate. 45:40.180 --> 45:42.190 Memory doesn't work that way. 45:42.190 --> 45:45.840 What happens is--What somebody will do in a hypnotizable state 45:45.839 --> 45:48.889 is they'll be very eager to please the hypnotist. 45:48.890 --> 45:50.910 And so they'll make stuff up. 45:50.909 --> 45:54.009 And people under hypnosis just make stuff up. 45:54.010 --> 45:57.710 And they do very enthusiastically and very 45:57.706 --> 46:00.136 believably make stuff up. 46:00.139 --> 46:03.529 This is particularly the case with hypnotic regression when we 46:03.526 --> 46:06.136 ask you to go back to your sixth birthday party, 46:06.136 --> 46:08.246 for instance. And what's great as a 46:08.247 --> 46:11.577 developmental psychologist is if I ask you to go back to your 46:11.583 --> 46:15.033 fourth birthday party and you're hypnotizable you'll be oh, 46:15.030 --> 46:18.490 just like a four-year-old except you won't be like a real 46:18.491 --> 46:20.741 four-year-old. What you'll be like is an 46:20.742 --> 46:23.672 adult's notion of what a four-year-old is supposed to be. 46:23.670 --> 46:28.170 In fact, this has happened in the extreme case with hypnotic 46:28.174 --> 46:32.534 regression where people claim to speak languages like from 46:32.525 --> 46:35.095 ancient Egypt. And linguists love these 46:35.098 --> 46:37.988 studies because you don't--of course you don't really sound 46:37.993 --> 46:40.693 like you're speaking a language from ancient Egypt. 46:40.690 --> 46:45.900 What you sound is like a North American who believes he's 46:45.901 --> 46:50.461 speaking a language from ancient Egypt so they're, 46:50.462 --> 46:52.512 "nonsense sounds." 46:52.510 --> 46:54.470 [laughter] And so what it makes you 46:54.471 --> 46:57.011 is--Hypnotism brings out the actor in you. 46:57.010 --> 47:01.020 It makes you want to give a persuasive account of what 47:01.018 --> 47:04.568 happened. And so hypnotism is just an 47:04.570 --> 47:10.920 extreme form of what normally happens in eyewitness testimony. 47:10.920 --> 47:11.870 Repressed memories. 47:11.869 --> 47:16.079 We could devote a class--We could devote a semester to the 47:16.081 --> 47:20.591 very heated debate in the United States mostly about repressed 47:20.589 --> 47:24.469 memories. There are many adults who have 47:24.469 --> 47:29.529 claimed to have experienced traumatic sexual abuse. 47:29.530 --> 47:32.300 In some cases, this is unexceptional from a 47:32.295 --> 47:33.805 memory point of view. 47:33.809 --> 47:35.349 People know this happened to them. 47:35.349 --> 47:38.119 They've always known it happens to them and then they tell 47:38.116 --> 47:41.216 people about it. But there's a subset of cases 47:41.215 --> 47:45.895 where people have had no memory up to a point of what happened 47:45.895 --> 47:48.715 to them. Then they go to a psychologist 47:48.717 --> 47:52.127 or a psychiatrist; they undergo questioning, 47:52.133 --> 47:54.993 often using hypnotic techniques; 47:54.989 --> 47:59.949 and then they recover a memory of past traumatic sexual abuse. 47:59.949 --> 48:02.689 And what this is--what makes this so debatable, 48:02.687 --> 48:04.767 and there is a debate about this. 48:04.769 --> 48:07.169 I don't want to try to preclude it one way or another. 48:07.170 --> 48:10.130 What makes this debatable is some psychologists believe that, 48:10.134 --> 48:12.844 in at least some cases, these memories are real and 48:12.839 --> 48:16.159 they have been repressed through a Freudian mechanism – that 48:16.158 --> 48:18.878 they're too terrible to bring to consciousness, 48:18.880 --> 48:22.110 and the therapy brings them out into real life. 48:22.110 --> 48:26.300 But most psychologists believe that these memories cannot be 48:26.295 --> 48:30.685 trusted, that these memories are created through the actions of 48:30.693 --> 48:33.653 the therapist. And so, there's actually 48:33.650 --> 48:37.920 considerable psychological and legal battles over the veracity 48:37.920 --> 48:41.630 of the therapists where women who have claimed to have 48:41.630 --> 48:44.520 sexual--be sexually abused, for instance, 48:44.524 --> 48:47.394 have pressed criminal charges against their fathers on the 48:47.391 --> 48:48.751 basis of false memories. 48:48.750 --> 48:52.300 Similarly, people who have been accused of sexual abuse have 48:52.304 --> 48:55.864 pressed criminal charges against psychiatrists claiming that 48:55.859 --> 48:58.929 these psychiatrists have implanted the memories into 48:58.932 --> 49:00.802 their sons and daughters. 49:00.800 --> 49:04.940 It is controversial whether memories are ever repressed. 49:04.940 --> 49:09.000 What isn't controversial is that, for at least some cases, 49:08.998 --> 49:12.058 you can implant false memories in people, 49:12.059 --> 49:15.329 not because you're a sinister or evil person but because you 49:15.329 --> 49:17.379 really believed something happened. 49:17.380 --> 49:20.540 And you talked to them about it and then you caused these 49:20.544 --> 49:22.244 memories to come into being. 49:22.239 --> 49:25.059 A final case is flashbulb memories. 49:25.059 --> 49:26.709 I asked this early in the semester. 49:26.710 --> 49:27.680 I'll ask it again. 49:27.679 --> 49:33.619 How many of you remember where you were on September 11,2001? 49:33.619 --> 49:37.589 Is there anybody who doesn't remember where they were on 49:37.594 --> 49:39.044 September 11,2001? 49:39.040 --> 49:40.030 It would be interesting. 49:40.030 --> 49:44.890 It was a socially relevant event, but here's the problem 49:44.885 --> 49:47.705 with these flashbulb memories. 49:47.710 --> 49:51.030 Flashbulb memories are the idea that these memories being so 49:51.028 --> 49:54.228 vivid, and they are vivid for many of us--exactly where we 49:54.233 --> 49:58.103 stood, what happened; well, they can't really be 49:58.098 --> 50:00.478 trusted. And here is why not. 50:00.480 --> 50:05.080 Because they are such important events, I bet many of you have 50:05.079 --> 50:09.079 actually heard the question before, "Where were you on 50:09.075 --> 50:10.805 September 11^(th)?" 50:10.810 --> 50:12.820 and talked about it. 50:12.820 --> 50:17.330 What happens in these conversations is stories change. 50:17.329 --> 50:20.829 I have my--I knew where I was on September 11^(th). 50:20.830 --> 50:22.810 My wife knew where she was. 50:22.809 --> 50:25.899 But I spent as much time listening to her talk about it 50:25.903 --> 50:28.083 as I spent time me talking about it. 50:28.079 --> 50:31.249 And now maybe my memory is actually of her experience and 50:31.247 --> 50:34.137 not mine. It's not--For all of these 50:34.144 --> 50:38.524 cases, the temptation you have to resist is saying, 50:38.519 --> 50:41.969 "Yes. I know memories can be swayed. 50:41.969 --> 50:45.079 I know they could be distorted and everything but, 50:45.075 --> 50:47.795 you see, I really am sure that happened." 50:47.800 --> 50:50.910 You have to resist that temptation because there are so 50:50.906 --> 50:53.776 many cases we know, including the tape of the girls 50:53.782 --> 50:56.362 that we just saw, where people are entirely sure 50:56.355 --> 50:59.125 things happened. And we know full well that they 50:59.134 --> 51:02.494 didn't exist. Being sure is no guarantee that 51:02.491 --> 51:05.811 a memory isn't false, reconstructed or even 51:05.812 --> 51:09.242 implanted. So, this part of memory has 51:09.240 --> 51:10.930 three main morals. 51:10.929 --> 51:12.829 There are many types of memories. 51:12.829 --> 51:15.639 I talked about short-term memory, long-term memory. 51:15.639 --> 51:18.969 I talked about implicit memory and explicit memory. 51:18.969 --> 51:21.659 These are sort of separable sort of memories. 51:21.659 --> 51:23.759 You could break one while having the other one impaired. 51:23.760 --> 51:28.040 Arguably, there are brain systems dedicated to memory for 51:28.044 --> 51:32.104 faces, memory for everyday objects, memory for spatial 51:32.098 --> 51:36.068 locations. The key to remembering is 51:36.073 --> 51:39.793 organization and understanding. 51:39.789 --> 51:44.159 Introduction to "X" courses, including Introduction to 51:44.156 --> 51:48.106 Psychology courses, are among the hardest courses 51:48.111 --> 51:51.571 at Yale. And the reason why is there is 51:51.569 --> 51:56.589 just a lot of material that is diverse and you have to command 51:56.590 --> 51:58.730 each aspect separately. 51:58.730 --> 52:01.950 The easiest courses at Yale tend to be highfalutin seminars 52:01.949 --> 52:05.279 where you kind of have enough of a background that everything 52:05.279 --> 52:07.499 is--can be clear and understandable. 52:07.500 --> 52:10.880 The more you understand something, the easier you'll 52:10.881 --> 52:12.871 remember it. And finally, 52:12.866 --> 52:16.296 you can't trust some of your memories. 52:16.300 --> 52:24.010 Your reading response for this week is you have to use your 52:24.008 --> 52:28.658 powers for good and not for evil, 52:28.659 --> 52:31.359 [laughter] though if you manage to succeed 52:31.362 --> 52:33.672 at this I will be very impressed. 52:33.670 --> 52:35.910 [laughter] But you have to describe, 52:35.905 --> 52:38.905 based on the lecture materials and the readings, 52:38.908 --> 52:41.078 how to implant a false memory. 52:41.080 --> 52:47.010 52:47.010 --> 52:48.210 We have a few minutes. 52:48.210 --> 52:52.850 Any questions on memory. 52:52.850 --> 52:55.600 Yes. Student: [inaudible] 52:55.604 --> 52:58.164 Professor Paul Bloom: Uh huh. 52:58.160 --> 53:02.270 Hey. Please-- Student: 53:02.269 --> 53:07.829 [inaudible] Is that long-term sensory 53:07.834 --> 53:10.634 memory? Professor Paul Bloom: 53:10.625 --> 53:12.665 The example is, "What sort of memory is it when 53:12.672 --> 53:14.232 you know how to play the piano?" 53:14.230 --> 53:15.940 And it's a very good question. 53:15.940 --> 53:19.130 It is long-term memory because you might know how to do a 53:19.133 --> 53:22.673 concerto or a song and then you have it stored in your head and 53:22.668 --> 53:24.548 you carry it around with you. 53:24.550 --> 53:27.200 You'll remember it a year from now, two years from now. 53:27.199 --> 53:30.699 It is long-term memory but it is also an excellent example of 53:30.698 --> 53:34.428 implicit memory because you know how to do it but you could do it 53:34.430 --> 53:36.880 unconsciously without attending to it. 53:36.880 --> 53:40.310 It's not sensory but it's as if, put it crudely, 53:40.311 --> 53:43.451 that your fingers know and not your mind. 53:43.449 --> 53:46.409 We have time for one more question. 53:46.410 --> 53:48.650 Yes. Student: [inaudible] 53:48.651 --> 53:51.031 Professor Paul Bloom: The question is about 53:51.032 --> 53:52.152 photographic memory. 53:52.150 --> 53:55.780 There are a lot of claims about photographic memory. 53:55.780 --> 53:59.870 My understanding is they do not tend to be substantiated. 53:59.869 --> 54:03.639 Sometimes photographic memory, and this came up when we talked 54:03.638 --> 54:07.158 about autism a few classes ago, is linked with savant-like 54:07.160 --> 54:09.150 skills. People who have severe 54:09.152 --> 54:12.552 impairments in some ways may have photographic memories in 54:12.553 --> 54:14.573 others. I am not convinced that 54:14.570 --> 54:17.970 photographic memory in the sense that you see something, 54:17.966 --> 54:21.236 you take a picture of it, you hold it in memory really 54:21.239 --> 54:23.949 exists. I think there may be one or two 54:23.948 --> 54:27.628 case studies that suggest it might be real but I think it's 54:27.630 --> 54:29.350 controversial. Okay. 54:29.349 --> 54:32.469 We have a guest lecturer on Wednesday. 54:32.469 --> 59:42.999 Dean Peter Salovey will talk to us about love.