1 00:00:00 --> 00:00:01 2 00:00:01 --> 00:00:03 INTRODUCTION: The following content is presented by MIT 3 00:00:03 --> 00:00:06 OpenCourseWare under a Creative Commons license. 4 00:00:06 --> 00:00:08 Additional information about our license and MIT 5 00:00:08 --> 00:00:15 OpenCourseWare in general, is available at OCW.MIT.edu. 6 00:00:15 --> 00:00:23 PROFESSOR: Good afternoon here is the plan for 7 00:00:23 --> 00:00:26 today's lecture. 8 00:00:26 --> 00:00:31 9 00:00:31 --> 00:00:38 There is out there somewhere, unless you're philosopher, 10 00:00:38 --> 00:00:40 a world, it's huge. 11 00:00:40 --> 00:00:43 A lot of stuff out there in the world. 12 00:00:43 --> 00:00:53 Their is inside use somewhere a long term memory, 13 00:00:53 --> 00:00:57 which is also huge. 14 00:00:57 --> 00:01:02 And and hopefully getting huger all the time. 15 00:01:02 --> 00:01:05 16 00:01:05 --> 00:01:14 You cannot transcribe all of that world from the outside 17 00:01:14 --> 00:01:17 into your long term memory, as you may have already discovered 18 00:01:17 --> 00:01:22 if you've had a test this term. 19 00:01:22 --> 00:01:27 And in a move that ought to looks familiar to you by now. 20 00:01:27 --> 00:01:35 We can described that as a bottleneck of some sort. 21 00:01:35 --> 00:01:37 Not the same bottleneck that I was talking about in the 22 00:01:37 --> 00:01:41 context of attention, but a bottleneck nevertheless, that 23 00:01:41 --> 00:01:50 is involved in providing and putting a severe limitation on 24 00:01:50 --> 00:01:55 the ability to encode the world into long term memory. 25 00:01:55 --> 00:01:58 The first part of today's lecture is going to be about 26 00:01:58 --> 00:02:04 this problem of encoding and about the nature of 27 00:02:04 --> 00:02:05 this bottleneck. 28 00:02:05 --> 00:02:12 The second part is it that OK some stuff, a mere trickle of 29 00:02:12 --> 00:02:18 the world, managed to get through this bottleneck. 30 00:02:18 --> 00:02:19 It needs to stay there. 31 00:02:19 --> 00:02:22 It needs to stay there for a long time. 32 00:02:22 --> 00:02:27 You've got things in here, in this long term memory of yours, 33 00:02:27 --> 00:02:35 that are 15, 16 years old, maybe more at this point. 34 00:02:35 --> 00:02:42 How they stay here, how they are made firm, is the 35 00:02:42 --> 00:02:44 problem of consolidation. 36 00:02:44 --> 00:02:47 Which it would say underneath there if it wasn't that 37 00:02:47 --> 00:02:49 I put a screen in front. 38 00:02:49 --> 00:02:54 And finally, it is of absolutely no use to you have 39 00:02:54 --> 00:02:56 -- as you also may have discovered on some exam this 40 00:02:56 --> 00:03:00 term already -- to have a long term memory that is absolutely 41 00:03:00 --> 00:03:04 chock full of marvelous material if you can't 42 00:03:04 --> 00:03:06 get it back out. 43 00:03:06 --> 00:03:11 Unlike the story that we were telling before about 44 00:03:11 --> 00:03:14 perception, where there is a world of bottleneck and then 45 00:03:14 --> 00:03:18 some construct that was your perception of the world. 46 00:03:18 --> 00:03:22 In this case you've got a two way street, that you cannot 47 00:03:22 --> 00:03:25 appreciates the full contents of your long term memory 48 00:03:25 --> 00:03:26 at any given moment. 49 00:03:26 --> 00:03:29 You need to retrieve -- that's what it says on the far board 50 00:03:29 --> 00:03:32 -- you need to be able to retrieve material from long 51 00:03:32 --> 00:03:35 term memory, and bring it back as we will see. 52 00:03:35 --> 00:03:39 In effect to bring it back into the bottleneck. 53 00:03:39 --> 00:03:43 54 00:03:43 --> 00:03:46 Got to be able to get back out into this limited capacity 55 00:03:46 --> 00:03:50 bottleneck where you can do things like write down the 56 00:03:50 --> 00:03:54 answer on a test, or tell me who's running for 57 00:03:54 --> 00:03:57 president this year. 58 00:03:57 --> 00:03:57 Nobody. 59 00:03:57 --> 00:03:59 AUDIENCE: Bush. 60 00:03:59 --> 00:04:00 PROFESSOR: Yeah that's one of them. 61 00:04:00 --> 00:04:01 Bush yeah. 62 00:04:01 --> 00:04:02 AUDIENCE: Kerry 63 00:04:02 --> 00:04:02 PROFESSOR: Kerry OK. 64 00:04:02 --> 00:04:03 And a few other characters too. 65 00:04:03 --> 00:04:04 Right. 66 00:04:04 --> 00:04:05 OK. 67 00:04:05 --> 00:04:10 Presumably except for the most politically obsessed among you, 68 00:04:10 --> 00:04:13 that was not an active thought in your mind until I asked it. 69 00:04:13 --> 00:04:17 But Kerry and Bush live in your long term memory, 70 00:04:17 --> 00:04:18 isn't that a scary thought? 71 00:04:18 --> 00:04:20 AUDIENCE: [LAUGHTER] 72 00:04:20 --> 00:04:23 PROFESSOR: And on demand, you could retrieve that back 73 00:04:23 --> 00:04:28 out into what we will call and your working memory. 74 00:04:28 --> 00:04:31 If we were just thinking about this and encoding terms, we 75 00:04:31 --> 00:04:36 might just call this short term memory -- stuff comes in from 76 00:04:36 --> 00:04:40 the outside, it's maintained for a little while somehow in a 77 00:04:40 --> 00:04:45 short term memory, and then somehow makes it into long term 78 00:04:45 --> 00:04:47 memory in ways that we'll talk about. 79 00:04:47 --> 00:04:52 But if we think about this in a two way, in a bi-directional 80 00:04:52 --> 00:04:56 kind of way, it's better perhaps to think of this 81 00:04:56 --> 00:04:57 as your working memory. 82 00:04:57 --> 00:05:01 Think of it as your computer desktop. 83 00:05:01 --> 00:05:03 It's got access to all the stuff that's on 84 00:05:03 --> 00:05:05 your hard drive. 85 00:05:05 --> 00:05:08 It's got access to the external world. 86 00:05:08 --> 00:05:10 You can tell it's got access to the external -- oh no 87 00:05:10 --> 00:05:11 we can't see it there. 88 00:05:11 --> 00:05:13 Well, you know, it's wireless and so it's 89 00:05:13 --> 00:05:14 got access to the web. 90 00:05:14 --> 00:05:15 And so that's the external world. 91 00:05:15 --> 00:05:18 But only a limited amount of it can be on the 92 00:05:18 --> 00:05:20 desktop any one time. 93 00:05:20 --> 00:05:24 Similarly, only a limited amount of either the world or 94 00:05:24 --> 00:05:27 the contents of your long term memory can be on your 95 00:05:27 --> 00:05:30 mental desktop at one time. 96 00:05:30 --> 00:05:31 That's working memory. 97 00:05:31 --> 00:05:37 That's an important junk of this bottleneck. 98 00:05:37 --> 00:05:43 Well what does it mean to say that a limited capacity 99 00:05:43 --> 00:05:45 bottleneck of some sort? 100 00:05:45 --> 00:05:49 Let me illustrate the limit to that capacity. 101 00:05:49 --> 00:05:53 We saw a particular example of that in what's known in the 102 00:05:53 --> 00:05:58 trade as visual short term memory last week, when I was 103 00:05:58 --> 00:06:01 putting up, you know, four little colored blobs and 104 00:06:01 --> 00:06:04 saying, OK now what changed. 105 00:06:04 --> 00:06:06 And there's a limit of about four objects that you 106 00:06:06 --> 00:06:08 can keep track of. 107 00:06:08 --> 00:06:17 But in trying to ask what's the limit of this bottleneck from 108 00:06:17 --> 00:06:21 the outside world into a more long term varieties of memory, 109 00:06:21 --> 00:06:26 one of the typical measures is what's called digit span. 110 00:06:26 --> 00:06:29 If you end up in the hospital because somebody bopped you 111 00:06:29 --> 00:06:32 over the head, one of the things you neurologist would 112 00:06:32 --> 00:06:34 do is read you a list of numbers, and ask you 113 00:06:34 --> 00:06:35 to repeat them back. 114 00:06:35 --> 00:06:39 And how many you can do is an index of whether you are OK. 115 00:06:39 --> 00:06:42 So let's see if you're OK. 116 00:06:42 --> 00:06:46 Except I won't do this with numbers, I will do 117 00:06:46 --> 00:06:46 it with color names. 118 00:06:46 --> 00:06:48 And we can come back to that in a minute. 119 00:06:48 --> 00:06:50 Where are my color names? 120 00:06:50 --> 00:06:50 Here are my color names. 121 00:06:50 --> 00:06:53 So what I'm going to do is read you a list of color names. 122 00:06:53 --> 00:06:56 I'll then say repeat, and you in glorious unison 123 00:06:56 --> 00:06:58 repeat this back to me. 124 00:06:58 --> 00:06:58 OK. 125 00:06:58 --> 00:07:03 So if I say, red, blue, purple, repeat. 126 00:07:03 --> 00:07:04 You say? 127 00:07:04 --> 00:07:07 AUDIENCE: Red, blue, purple. 128 00:07:07 --> 00:07:08 PROFESSOR: It never fails. 129 00:07:08 --> 00:07:10 AUDIENCE: [LAUGHTER] 130 00:07:10 --> 00:07:11 PROFESSOR: Right, [? Mara, ?] 131 00:07:11 --> 00:07:13 last year too? 132 00:07:13 --> 00:07:15 Always you get the repeat back too. 133 00:07:15 --> 00:07:18 And I'm not sure what it is about this that demands, that 134 00:07:18 --> 00:07:20 at least from some people, the repeats of repeat. 135 00:07:20 --> 00:07:21 But in any case good, good. 136 00:07:21 --> 00:07:25 You're not deeply brain damaged yet. 137 00:07:25 --> 00:07:26 Ready? 138 00:07:26 --> 00:07:27 We'll do some more of these. 139 00:07:27 --> 00:07:34 Green, pink, yellow, white yellow, repeat. 140 00:07:34 --> 00:07:41 AUDIENCE: Green, pink, yellow, white, yellow. 141 00:07:41 --> 00:07:42 PROFESSOR: OK. 142 00:07:42 --> 00:07:43 That was fine. 143 00:07:43 --> 00:07:45 All right, here we go. 144 00:07:45 --> 00:07:48 Oh, I forgot to mention. 145 00:07:48 --> 00:07:52 It turns out to be really easy to repeat back in essentially 146 00:07:52 --> 00:07:56 infinitely long list of these, if you write them down 147 00:07:56 --> 00:07:57 while I'm saying them. 148 00:07:57 --> 00:07:59 AUDIENCE: [LAUGHTER] 149 00:07:59 --> 00:08:01 PROFESSOR: There are a variety of demos that I'm going to do 150 00:08:01 --> 00:08:06 today that are of that form, and it's really boring. 151 00:08:06 --> 00:08:09 So, you know, OK we know you can do that, so don't do 152 00:08:09 --> 00:08:11 that, because that's boring. 153 00:08:11 --> 00:08:12 Be a nice honest person. 154 00:08:12 --> 00:08:14 OK ready? 155 00:08:14 --> 00:08:15 Where was I. 156 00:08:15 --> 00:08:15 I just did 5, right? 157 00:08:15 --> 00:08:16 Let's try 7. 158 00:08:16 --> 00:08:17 Ready? 159 00:08:17 --> 00:08:18 Oh and I told you when I'm going to stop. 160 00:08:18 --> 00:08:20 Oh what the heck. 161 00:08:20 --> 00:08:29 Pink, red, yellow, green, purple, pink, orange, repeat. 162 00:08:29 --> 00:08:35 AUDIENCE: Pink, red, yellow, green, purple, pink, orange. 163 00:08:35 --> 00:08:37 PROFESSOR: It's getting a little weak there at the end. 164 00:08:37 --> 00:08:41 But you will see, if you look at the handout, you will see 165 00:08:41 --> 00:08:45 that the answer to this question is given in the title 166 00:08:45 --> 00:08:47 of George Miller's classic paper on the subject. 167 00:08:47 --> 00:08:51 And is answer is 7 plus or minus 2, and some of you 168 00:08:51 --> 00:08:53 were on the minus side. 169 00:08:53 --> 00:08:56 Oh I should also say -- I should've said this at the 170 00:08:56 --> 00:09:01 beginning of the course -- people in intro psych classes, 171 00:09:01 --> 00:09:03 particularly by the time you get around starting to talk 172 00:09:03 --> 00:09:07 about cognitive things, like memory, take demos very 173 00:09:07 --> 00:09:15 personally and think 7 plus or minus 2, I got minus. 174 00:09:15 --> 00:09:15 I'm doomed. 175 00:09:15 --> 00:09:19 AUDIENCE: [LAUGHTER] 176 00:09:19 --> 00:09:20 PROFESSOR: You're probably fine. 177 00:09:20 --> 00:09:23 I mean if you, if it was the red, yellow, purple, you're 178 00:09:23 --> 00:09:25 saying what was the one after red. 179 00:09:25 --> 00:09:27 You know I admit there might be an issue there. 180 00:09:27 --> 00:09:32 181 00:09:32 --> 00:09:35 The advantage of a 300 person intro class is we're looking at 182 00:09:35 --> 00:09:38 the average, and what you're doing on any given single 183 00:09:38 --> 00:09:43 isolated demo trial is not actually deeply relevant to 184 00:09:43 --> 00:09:46 whether or not you're going to pass physics for instance. 185 00:09:46 --> 00:09:47 OK, one more. 186 00:09:47 --> 00:09:48 Ready? 187 00:09:48 --> 00:09:58 Blue, green, purple, red, pink, yellow, green, blue, 188 00:09:58 --> 00:10:01 white, orange, repeat. 189 00:10:01 --> 00:10:08 AUDIENCE: Blue, green, purple, red, green, yellow 190 00:10:08 --> 00:10:11 [LAUGHTER] 191 00:10:11 --> 00:10:12 PROFESSOR: Yeah, yeah, yeah. 192 00:10:12 --> 00:10:15 193 00:10:15 --> 00:10:16 You can hear. 194 00:10:16 --> 00:10:18 That was 10. 195 00:10:18 --> 00:10:21 And you could here it in the group, and you could probably 196 00:10:21 --> 00:10:25 feel it in yourself, that, you know, up to 6, 7, 8, you're 197 00:10:25 --> 00:10:26 yeah, we're good here. 198 00:10:26 --> 00:10:28 Oh man if I start taking in this next one, that 199 00:10:28 --> 00:10:30 I'm losing them here. 200 00:10:30 --> 00:10:31 Something bad is happening. 201 00:10:31 --> 00:10:38 So there's this capacity limit on what you can get into -- 202 00:10:38 --> 00:10:40 well let's used this jargon for the time being -- into some 203 00:10:40 --> 00:10:43 sort of short term memory, and then spit out 204 00:10:43 --> 00:10:44 immediately to me. 205 00:10:44 --> 00:10:49 Now 7 plus or minus to what? 206 00:10:49 --> 00:10:50 Let's try another one. 207 00:10:50 --> 00:10:51 You ready? 208 00:10:51 --> 00:10:54 Same game. 209 00:10:54 --> 00:11:04 Red, red, red, red, blue, blue, blue, blue, green, green, 210 00:11:04 --> 00:11:07 green, green, repeat. 211 00:11:07 --> 00:11:09 AUDIENCE: Red -- 212 00:11:09 --> 00:11:10 [LAUGHTER] 213 00:11:10 --> 00:11:13 PROFESSOR: You know you can do this. 214 00:11:13 --> 00:11:14 That's 12 items. 215 00:11:14 --> 00:11:19 How come I all of a sudden manage to boost your 216 00:11:19 --> 00:11:22 memories so effectively? 217 00:11:22 --> 00:11:25 Yeah classification, or what Miller originally, and the 218 00:11:25 --> 00:11:28 field had stuck with the notion of chunking. 219 00:11:28 --> 00:11:31 If you can break things into meaningful chunks , it's 220 00:11:31 --> 00:11:34 7 plus or minus 2 chunks. 221 00:11:34 --> 00:11:37 And what's important is how you managed to cut the 222 00:11:37 --> 00:11:40 world into chunks. 223 00:11:40 --> 00:11:43 This by the way explains why I don't do this demo in the 224 00:11:43 --> 00:11:49 traditional way of giving you digit strings. 225 00:11:49 --> 00:11:54 Because years ago I discovered that if you give somebody a 10 226 00:11:54 --> 00:11:59 digit, you know, give 300 MIT undergraduates a 10 digit 227 00:11:59 --> 00:12:01 string, you know, a bunch of them crud out on you, but then 228 00:12:01 --> 00:12:04 there's a handful of people who just raddle it right off. 229 00:12:04 --> 00:12:06 And you say oh my goodness, these people have 230 00:12:06 --> 00:12:07 this amazing memory. 231 00:12:07 --> 00:12:09 And you ask them how'd you do it? 232 00:12:09 --> 00:12:11 And they say oh well that was obvious. 233 00:12:11 --> 00:12:13 It was the natural log of 16. 234 00:12:13 --> 00:12:16 AUDIENCE: [LAUGHTER] 235 00:12:16 --> 00:12:20 PROFESSOR: MIT students tend to junk numbers rather more 236 00:12:20 --> 00:12:23 readily than the rest of the population. 237 00:12:23 --> 00:12:27 They don't tend to junk color names particularly adeptly I've 238 00:12:27 --> 00:12:32 found, and so you get the right 7 plus or minus 2 answer. 239 00:12:32 --> 00:12:37 Now this sort of chunking, you should think of this 240 00:12:37 --> 00:12:38 more like a rate right. 241 00:12:38 --> 00:12:42 It's not 7 plus or minus 2 things ever, ever, ever. 242 00:12:42 --> 00:12:46 It's 7 plus or minus 2 things in some unit of time that you 243 00:12:46 --> 00:12:47 can manage to deal with. 244 00:12:47 --> 00:12:51 Otherwise as soon as I had presented 7 plus or minus 2 245 00:12:51 --> 00:12:54 units of information in this lecture, you'd be like dead 246 00:12:54 --> 00:12:55 for the rest of the day. 247 00:12:55 --> 00:12:58 Or maybe, you know, the first lecture shot your brain and 248 00:12:58 --> 00:13:00 that was it for the term. 249 00:13:00 --> 00:13:02 That's obviously not the case. 250 00:13:02 --> 00:13:06 What the question is how big a chunk can you manage to cut 251 00:13:06 --> 00:13:09 things into as it's coming in. 252 00:13:09 --> 00:13:15 One of the places you can see that is in language as you're 253 00:13:15 --> 00:13:19 trying to keep track for instance, what I'm saying. 254 00:13:19 --> 00:13:26 So if I ask you to repeat this sentence, memory is a 255 00:13:26 --> 00:13:29 fascinating topic that would be on the final exam, 256 00:13:29 --> 00:13:31 you would say? 257 00:13:31 --> 00:13:33 AUDIENCE: Memory is a fascinating topic that would 258 00:13:33 --> 00:13:35 be on the final exam. 259 00:13:35 --> 00:13:39 PROFESSOR: But if I lettuce, quadrangle forms only only 260 00:13:39 --> 00:13:42 column bug de la forms aftunic reply quintilian. 261 00:13:42 --> 00:13:43 Kelly You say? 262 00:13:43 --> 00:13:46 AUDIENCE: [LAUGHTER] 263 00:13:46 --> 00:13:49 PROFESSOR: Say it roughly the same number of syllables. 264 00:13:49 --> 00:13:51 But ouughh. 265 00:13:51 --> 00:13:53 You know, you can't repeat that back, because 266 00:13:53 --> 00:13:54 you can't chunk it. 267 00:13:54 --> 00:13:57 Well you can, but the chunks that you were making were too 268 00:13:57 --> 00:14:01 small to survive the rate at which they were 269 00:14:01 --> 00:14:03 being presented. 270 00:14:03 --> 00:14:06 So everybody here, if forced to, would've produce 271 00:14:06 --> 00:14:08 some little bit of it. 272 00:14:08 --> 00:14:13 But people would not be able to produce the whole thing, 273 00:14:13 --> 00:14:16 because you can't chunk it into meaningful chunks that work. 274 00:14:16 --> 00:14:20 This is also the route presumably of the near 275 00:14:20 --> 00:14:25 universal experience that when you go to some other country 276 00:14:25 --> 00:14:28 where they speak another language, that you nominally 277 00:14:28 --> 00:14:31 speak because you took four years of it in high school. 278 00:14:31 --> 00:14:36 You discover that those Parisians or those Spaniards or 279 00:14:36 --> 00:14:40 the Japanese or whoever talk really, really, really fast. 280 00:14:40 --> 00:14:42 What's the chance that everybody else in the 281 00:14:42 --> 00:14:44 world talks really, really, really fast. 282 00:14:44 --> 00:14:46 You know, it's not true of course. 283 00:14:46 --> 00:14:50 And those of you who are non-native English speakers, 284 00:14:50 --> 00:14:54 and have now arrived at MIT, are thinking no, you know, 285 00:14:54 --> 00:14:58 back in, you know, Kazakhstan they all spoke really 286 00:14:58 --> 00:14:59 slowly or whatever. 287 00:14:59 --> 00:15:02 But these English speakers are wackos. 288 00:15:02 --> 00:15:05 The problem is you are in your native tongue, or in any town 289 00:15:05 --> 00:15:07 but you are really fluent in. 290 00:15:07 --> 00:15:12 You are very good at cutting into meaningful chunks rapidly. 291 00:15:12 --> 00:15:15 If you're busy trying to say, what was that word. 292 00:15:15 --> 00:15:15 I know. 293 00:15:15 --> 00:15:17 I damn it that was the week I was asleep in high 294 00:15:17 --> 00:15:18 school or something. 295 00:15:18 --> 00:15:18 AUDIENCE: [LAUGHER] 296 00:15:18 --> 00:15:20 PROFESSOR: No that was the year I was asleep in high school. 297 00:15:20 --> 00:15:20 Anyway. 298 00:15:20 --> 00:15:24 299 00:15:24 --> 00:15:28 The chunks go by too fast like in the nonsense that 300 00:15:28 --> 00:15:34 I was producing for you. 301 00:15:34 --> 00:15:41 You're set up to get chunks through in some domains much 302 00:15:41 --> 00:15:42 better than in others. 303 00:15:42 --> 00:15:46 And in fact, let me illustrate that here. 304 00:15:46 --> 00:15:48 By where did I hide? 305 00:15:48 --> 00:15:50 That looks like it's probably it. 306 00:15:50 --> 00:15:51 All right. 307 00:15:51 --> 00:15:53 Got to memorize some more stuff here. 308 00:15:53 --> 00:15:54 Come on. 309 00:15:54 --> 00:15:55 You want to go. 310 00:15:55 --> 00:15:58 And there we go. 311 00:15:58 --> 00:16:00 Too much light. 312 00:16:00 --> 00:16:04 Let's kill the stage lights. 313 00:16:04 --> 00:16:05 Good. 314 00:16:05 --> 00:16:07 OK. 315 00:16:07 --> 00:16:10 I want you to memorize all these pictures. 316 00:16:10 --> 00:16:12 You ready? 317 00:16:12 --> 00:16:14 There's a picture. 318 00:16:14 --> 00:16:16 And there's a picture. 319 00:16:16 --> 00:16:18 And that's a picture. 320 00:16:18 --> 00:16:21 And there's the back end of a horse or two. 321 00:16:21 --> 00:16:23 And there's a Christmas tree or something. 322 00:16:23 --> 00:16:28 And I don't know I must've rated a housing site. 323 00:16:28 --> 00:16:30 But, you know, these are definitely pictures. 324 00:16:30 --> 00:16:31 And that's a picture. 325 00:16:31 --> 00:16:33 And that looks like a picture. 326 00:16:33 --> 00:16:35 And oh look there's tomatoes. 327 00:16:35 --> 00:16:36 OK. 328 00:16:36 --> 00:16:37 Got them all? 329 00:16:37 --> 00:16:38 AUDIENCE: Yes. 330 00:16:38 --> 00:16:41 PROFESSOR: OK now what I want you to do is, I'm going to show 331 00:16:41 --> 00:16:44 you some pictures that are either old or new, and you just 332 00:16:44 --> 00:16:48 say yes if you've seen it before, no if you havent. 333 00:16:48 --> 00:16:50 AUDIENCE: No. 334 00:16:50 --> 00:16:51 No. 335 00:16:51 --> 00:16:53 Yes. 336 00:16:53 --> 00:16:54 Yes. 337 00:16:54 --> 00:16:55 No. 338 00:16:55 --> 00:17:00 Yes Yes, No. 339 00:17:00 --> 00:17:01 No. 340 00:17:01 --> 00:17:02 Yes. 341 00:17:02 --> 00:17:04 No. 342 00:17:04 --> 00:17:05 PROFESSOR: OK. 343 00:17:05 --> 00:17:10 The important point is that you're very good at this. 344 00:17:10 --> 00:17:15 The interesting side point is that this particular one is 345 00:17:15 --> 00:17:16 flipped left, right, reverse. 346 00:17:16 --> 00:17:18 So if you were sitting there saying why are these people 347 00:17:18 --> 00:17:20 next to me changing their mind. 348 00:17:20 --> 00:17:23 It's because some people noticed that it was 349 00:17:23 --> 00:17:24 left, right, reverse. 350 00:17:24 --> 00:17:29 But if I had lots of time, I could have done this 351 00:17:29 --> 00:17:31 with a lot of pictures. 352 00:17:31 --> 00:17:37 The largest report in the literature is 10,000 pictures 353 00:17:37 --> 00:17:42 shown to the usual collection of college undergraduates. 354 00:17:42 --> 00:17:47 And some days later, the students still remembered 355 00:17:47 --> 00:17:53 I think it's about 80% accuracy on that. 356 00:17:53 --> 00:17:57 If I read you 10,000 color names -- well there aren't 357 00:17:57 --> 00:18:02 10,000 color names -- if I read you 10,000 words or almost any 358 00:18:02 --> 00:18:06 other material that I would care to present to you, you 359 00:18:06 --> 00:18:08 would do nowhere near as well. 360 00:18:08 --> 00:18:11 We're not entirely clear on how you do this. 361 00:18:11 --> 00:18:13 This is quite a remarkable ability. 362 00:18:13 --> 00:18:19 But it does speak to the ability of picture information. 363 00:18:19 --> 00:18:22 Familiarity with a picture. 364 00:18:22 --> 00:18:24 Getting through this bottleneck into some sort of long term 365 00:18:24 --> 00:18:28 memory with remarkable efficiency, that you are 366 00:18:28 --> 00:18:31 somehow specialized for doing that. 367 00:18:31 --> 00:18:34 There's also a little bit of a cheat here. 368 00:18:34 --> 00:18:35 It's only a little bit of a cheat, but it's 369 00:18:35 --> 00:18:36 an interesting cheat. 370 00:18:36 --> 00:18:40 This was a recognition test. 371 00:18:40 --> 00:18:44 The color one, the previous demos were recall test. 372 00:18:44 --> 00:18:47 I was asking you what, you know, pull out of your 373 00:18:47 --> 00:18:48 memory what was there. 374 00:18:48 --> 00:18:51 Here I was saying was this there. 375 00:18:51 --> 00:18:52 And you know this. 376 00:18:52 --> 00:18:57 But, from your test taking experience you know that, OK if 377 00:18:57 --> 00:18:59 I give you a choice between fill in the blank and multiple 378 00:18:59 --> 00:19:01 choice, which do you want to do? 379 00:19:01 --> 00:19:03 AUDIENCE: Multiple choice. 380 00:19:03 --> 00:19:04 PROFESSOR: Multiple choice, because that's 381 00:19:04 --> 00:19:06 a recognition test. 382 00:19:06 --> 00:19:09 Fill in the blank is recall test. 383 00:19:09 --> 00:19:14 The distinction is OK here's your long term memory. 384 00:19:14 --> 00:19:18 Here's the target that you're trying to find in long 385 00:19:18 --> 00:19:20 term memory somewhere. 386 00:19:20 --> 00:19:26 If I'm doing a recall test, I'm saying, you know, go find this 387 00:19:26 --> 00:19:30 thing in all the vast halls of your long term memory, and you 388 00:19:30 --> 00:19:33 send some little probe out into long term memory that 389 00:19:33 --> 00:19:35 may or may not find it. 390 00:19:35 --> 00:19:40 If I say is this in there. 391 00:19:40 --> 00:19:42 That's an obviously simpler matching kind of task. 392 00:19:42 --> 00:19:48 And you are typically much better at doing recognition 393 00:19:48 --> 00:19:54 than you are at doing recall. 394 00:19:54 --> 00:19:59 So you have presumably special purpose mechanisms in your 395 00:19:59 --> 00:20:04 system that are really good at doing some tasks. 396 00:20:04 --> 00:20:09 What do you do in the case of those unfortunate tasks. 397 00:20:09 --> 00:20:12 Actually I don't need that anymore at all. 398 00:20:12 --> 00:20:14 In those case of those unfortunate tasks which you 399 00:20:14 --> 00:20:16 were not well designed to do. 400 00:20:16 --> 00:20:21 For instance the memorization of neuroanatomical terms for 401 00:20:21 --> 00:20:24 a midterm in intro psych, which is only the beginning. 402 00:20:24 --> 00:20:28 How many people over here are pre-meds or think 403 00:20:28 --> 00:20:29 they might be pre-meds? 404 00:20:29 --> 00:20:33 You guys are the ones who are going to devote years of your 405 00:20:33 --> 00:20:37 life to the memorization of things that you were 406 00:20:37 --> 00:20:39 not built to memorize. 407 00:20:39 --> 00:20:43 Like where all the bones are, and stuff like that. 408 00:20:43 --> 00:20:45 So how do you do that? 409 00:20:45 --> 00:20:50 Let's try memorizing some nonsense. 410 00:20:50 --> 00:20:54 411 00:20:54 --> 00:20:55 Go to sleep. 412 00:20:55 --> 00:20:56 Why don't you want to go to sleep. 413 00:20:56 --> 00:20:58 AUDIENCE: [LAUGHTER] 414 00:20:58 --> 00:21:01 PROFESSOR: It's like your children. 415 00:21:01 --> 00:21:02 Mara, what do you do? 416 00:21:02 --> 00:21:04 417 00:21:04 --> 00:21:06 I don't want to know OK. 418 00:21:06 --> 00:21:07 You mean I'm supposed to hit it? 419 00:21:07 --> 00:21:09 AUDIENCE: [LAUGHTER] 420 00:21:09 --> 00:21:10 PROFESSOR: Oh OK. 421 00:21:10 --> 00:21:13 All right. 422 00:21:13 --> 00:21:15 It went to sleep. 423 00:21:15 --> 00:21:17 It happens from time to time. 424 00:21:17 --> 00:21:18 Anyway where were we. 425 00:21:18 --> 00:21:20 Oh yes nonsense. 426 00:21:20 --> 00:21:23 This is a line of work that starts really with a guy named 427 00:21:23 --> 00:21:26 Ebbinghaus in Germany in the late 19th century, who wanted 428 00:21:26 --> 00:21:28 to understand memory processes. 429 00:21:28 --> 00:21:33 He's in isolation from these issues about meaning. 430 00:21:33 --> 00:21:38 And so what he did was he got himself actually for starters 431 00:21:38 --> 00:21:43 to memorize nonsense syllables that allegedly had no meaning. 432 00:21:43 --> 00:21:46 It's extremely hard to find material that has no meaning. 433 00:21:46 --> 00:21:50 But I will attempt to read you some now. 434 00:21:50 --> 00:21:52 Grab a hunk of paper. 435 00:21:52 --> 00:21:54 I don't want your writing them down while I read them to you, 436 00:21:54 --> 00:21:58 but I do want you to be in a position to write all the words 437 00:21:58 --> 00:22:02 down when I tell you it's time to write. 438 00:22:02 --> 00:22:08 Oh while you're doing that, let me make my annual disclaimer. 439 00:22:08 --> 00:22:12 MIT is a hugely multicultural institution. 440 00:22:12 --> 00:22:15 You speak a vast range of languages. 441 00:22:15 --> 00:22:18 To my knowledge what I am saying is meaningless. 442 00:22:18 --> 00:22:22 If it turns out that it is meaningful, particularly if it 443 00:22:22 --> 00:22:26 turns out that it's like deeply profane, do let me know later 444 00:22:26 --> 00:22:29 and I will change my nonsense yet again. 445 00:22:29 --> 00:22:32 I have over the years said a number of absolutely 446 00:22:32 --> 00:22:34 remarkable things. 447 00:22:34 --> 00:22:39 Including something in Mali so good that the young women who 448 00:22:39 --> 00:22:43 told me I couldn't say it, wouldn't tell me what 449 00:22:43 --> 00:22:44 it was I had said. 450 00:22:44 --> 00:22:45 But -- 451 00:22:45 --> 00:22:45 AUDIENCE: [LAUGHTER] 452 00:22:45 --> 00:22:48 PROFESSOR: -- they looked at my list and they just 453 00:22:48 --> 00:22:50 crossed these things out. 454 00:22:50 --> 00:22:51 Anyway. 455 00:22:51 --> 00:22:55 So it's mostly gone. 456 00:22:55 --> 00:22:57 So all right. 457 00:22:57 --> 00:22:58 You're ready to write? 458 00:22:58 --> 00:23:02 I'm going to read you a bunch of nonsense syllables, and 459 00:23:02 --> 00:23:06 you're going to write them all down. 460 00:23:06 --> 00:23:09 Ready OK. 461 00:23:09 --> 00:23:11 Fip. 462 00:23:11 --> 00:23:14 Dut. 463 00:23:14 --> 00:23:16 Moche. 464 00:23:16 --> 00:23:19 Yill. 465 00:23:19 --> 00:23:21 Saz. 466 00:23:21 --> 00:23:24 Tirt. 467 00:23:24 --> 00:23:26 Varl. 468 00:23:26 --> 00:23:29 Bince. 469 00:23:29 --> 00:23:32 Jucks. 470 00:23:32 --> 00:23:34 Gouf. 471 00:23:34 --> 00:23:37 Zauce. 472 00:23:37 --> 00:23:39 Rab. 473 00:23:39 --> 00:23:40 OK. 474 00:23:40 --> 00:23:41 Write down everything you can remember. 475 00:23:41 --> 00:23:51 AUDIENCE: [UNINTELLIGIBLE] 476 00:23:51 --> 00:23:52 PROFESSOR: Now don't copy off your neighbor. 477 00:23:52 --> 00:24:03 478 00:24:03 --> 00:24:12 On your hand out on the second page I believe, you will find 479 00:24:12 --> 00:24:17 the axes for what's called the serial position curves. 480 00:24:17 --> 00:24:19 Which is the way we want to represent the data 481 00:24:19 --> 00:24:21 from this experiment. 482 00:24:21 --> 00:24:26 We want to represent percentage recall, which in this case will 483 00:24:26 --> 00:24:29 be in sort of number of hands, as a function of 484 00:24:29 --> 00:24:31 position in the list. 485 00:24:31 --> 00:24:38 Where in the list that I read did the word show up. 486 00:24:38 --> 00:24:40 And I will tell you the answer at ahead of time, and then 487 00:24:40 --> 00:24:44 we'll check if it worked. 488 00:24:44 --> 00:24:50 The answer is that the data will look like this. 489 00:24:50 --> 00:24:56 There will be what's known as a primacy effect, where you'll 490 00:24:56 --> 00:24:59 remember the first words in the list quite well. 491 00:24:59 --> 00:25:07 And a recency affect, with some preservation of the 492 00:25:07 --> 00:25:08 last words in the list. 493 00:25:08 --> 00:25:12 That at least is my assertion. 494 00:25:12 --> 00:25:16 Let us actually check the data here. 495 00:25:16 --> 00:25:19 How many people got fib? 496 00:25:19 --> 00:25:22 497 00:25:22 --> 00:25:22 Alright. 498 00:25:22 --> 00:25:25 Data point turns out to be about there I think. 499 00:25:25 --> 00:25:27 How many people got dut? 500 00:25:27 --> 00:25:30 501 00:25:30 --> 00:25:32 Few, fewer. 502 00:25:32 --> 00:25:35 How many people got -- well you see your falling a 503 00:25:35 --> 00:25:36 little below the line here. 504 00:25:36 --> 00:25:38 I am worried about you guys. 505 00:25:38 --> 00:25:42 How many got tirt? 506 00:25:42 --> 00:25:46 Ouhh how interesting. 507 00:25:46 --> 00:25:49 How many people got varl? 508 00:25:49 --> 00:25:52 Ouhh that's pretty pathetic. 509 00:25:52 --> 00:25:55 How many people got zauce. 510 00:25:55 --> 00:25:56 Oh coming back. 511 00:25:56 --> 00:25:57 Oh that's pretty good actually. 512 00:25:57 --> 00:25:58 That's above the line here. 513 00:25:58 --> 00:26:02 And how many people got rab? 514 00:26:02 --> 00:26:04 That's sort of thereish. 515 00:26:04 --> 00:26:05 OK. 516 00:26:05 --> 00:26:07 What's with tirt? 517 00:26:07 --> 00:26:09 [LAUGHTER] 518 00:26:09 --> 00:26:10 You guys did bet it. 519 00:26:10 --> 00:26:14 Well you know, teaching concourse this morning, I 520 00:26:14 --> 00:26:18 discovered that one of my words has currently acquired a slang 521 00:26:18 --> 00:26:20 meaning that I'd never heard of before. 522 00:26:20 --> 00:26:25 So tirt you just got lucky on, or tirt suddenly means -- 523 00:26:25 --> 00:26:27 AUDIENCE: [UNINTELLIGIBLE] 524 00:26:27 --> 00:26:30 PROFESSOR: What's a tirt? 525 00:26:30 --> 00:26:30 What? 526 00:26:30 --> 00:26:33 AUDIENCE: [UNINTELLIGIBLE] 527 00:26:33 --> 00:26:34 PROFESSOR: Tert. 528 00:26:34 --> 00:26:37 Oh oh tert like as in tertiary or something? 529 00:26:37 --> 00:26:37 AUDIENCE: Yes. 530 00:26:37 --> 00:26:40 PROFESSOR: Oh, see I'm spelling it t-i-r-t, and 531 00:26:40 --> 00:26:42 it never occurred to me. 532 00:26:42 --> 00:26:43 All right, all right. 533 00:26:43 --> 00:26:45 We're going to have to do something about tirt. 534 00:26:45 --> 00:26:46 Anyway. 535 00:26:46 --> 00:26:49 536 00:26:49 --> 00:26:50 OK. 537 00:26:50 --> 00:26:53 The question here is going to be how did you do it? 538 00:26:53 --> 00:26:57 Part of the answer is clearly any word that happened to make 539 00:26:57 --> 00:27:03 some sort of meaningful association to you is going to 540 00:27:03 --> 00:27:07 have a preferential chance of getting into long term memory. 541 00:27:07 --> 00:27:09 Where did my beautiful little map of the world go? 542 00:27:09 --> 00:27:11 It must be underneath there. 543 00:27:11 --> 00:27:15 You go up. 544 00:27:15 --> 00:27:20 One way to think about this is sort of by analogy 545 00:27:20 --> 00:27:21 to immigration. 546 00:27:21 --> 00:27:27 You know here's America, and here are all of our various 547 00:27:27 --> 00:27:29 forebearances and stuff trying to get into the country. 548 00:27:29 --> 00:27:32 And there's this choke point at immigration. 549 00:27:32 --> 00:27:39 You know, if you got your Uncle Charlie already in the country, 550 00:27:39 --> 00:27:44 he can pull you in ways that if you have no relations here 551 00:27:44 --> 00:27:46 it's harder to get in. 552 00:27:46 --> 00:27:49 So if tirt reminded you of tertiary, that's Uncle 553 00:27:49 --> 00:27:52 Tertiary in long term memory. 554 00:27:52 --> 00:27:53 Sort of pulling you through. 555 00:27:53 --> 00:27:57 But the question is what do you do if you don't have that? 556 00:27:57 --> 00:27:59 So the things at the beginning. 557 00:27:59 --> 00:28:02 Anybody got any intuition about what they were doing with fip 558 00:28:02 --> 00:28:06 and dut and stuff like that? 559 00:28:06 --> 00:28:07 Yeah, Yeah, OK. 560 00:28:07 --> 00:28:10 AUDIENCE: I was writing down this fip and dut. 561 00:28:10 --> 00:28:11 PROFESSOR: Oh OK. 562 00:28:11 --> 00:28:13 Apart from the fact that you were -- no, no, you presumably 563 00:28:13 --> 00:28:15 were being a good person and not writing them down when I 564 00:28:15 --> 00:28:16 read them first though right? 565 00:28:16 --> 00:28:17 OK. 566 00:28:17 --> 00:28:19 So what were you doing while you were waiting 567 00:28:19 --> 00:28:23 to dump them back out? 568 00:28:23 --> 00:28:23 Yeah. 569 00:28:23 --> 00:28:24 AUDIENCE: I mean I was going over them in my head. 570 00:28:24 --> 00:28:25 Rehearsing them. 571 00:28:25 --> 00:28:26 PROFESSOR: Yeah. 572 00:28:26 --> 00:28:26 OK. 573 00:28:26 --> 00:28:28 You were sitting there rehearsing. 574 00:28:28 --> 00:28:29 How many people had the experience that they were 575 00:28:29 --> 00:28:33 rehearsing, and discovering that oh man there's more stuff 576 00:28:33 --> 00:28:35 here than I can rehearse? 577 00:28:35 --> 00:28:37 That's the capacity limit again. 578 00:28:37 --> 00:28:40 But there's a rehearsal loop for auditorium material 579 00:28:40 --> 00:28:45 sometimes called a phonological loop. 580 00:28:45 --> 00:28:48 Does it say phonological loop on the handout anywhere? 581 00:28:48 --> 00:28:49 No. 582 00:28:49 --> 00:28:50 Well. 583 00:28:50 --> 00:28:55 You know, phono like phonograph, phonological. 584 00:28:55 --> 00:28:59 Where you repeat this stuff to hold it in this working memory. 585 00:28:59 --> 00:29:02 Keep it alive in working memory. 586 00:29:02 --> 00:29:07 And in that way sort of keep it alive. 587 00:29:07 --> 00:29:11 588 00:29:11 --> 00:29:13 Anybody got a different experience for zauce and 589 00:29:13 --> 00:29:15 rab at the end of there? 590 00:29:15 --> 00:29:18 591 00:29:18 --> 00:29:20 I'm sorry I heard something that sounded promising. 592 00:29:20 --> 00:29:21 Yeah? 593 00:29:21 --> 00:29:21 AUDIENCE: They were just there. 594 00:29:21 --> 00:29:21 PROFESSOR: They 595 00:29:21 --> 00:29:27 were just there is actually the right intuition. 596 00:29:27 --> 00:29:32 It was almost like you could still hear them. 597 00:29:32 --> 00:29:34 They didn't need to be preserved yet. 598 00:29:34 --> 00:29:37 They were still fresh and hadn't rotted. 599 00:29:37 --> 00:29:41 The claim though is that the process that gives you the 600 00:29:41 --> 00:29:43 recency effect is different than the process that gives 601 00:29:43 --> 00:29:45 you the primacy effect. 602 00:29:45 --> 00:29:48 Let's give that a try. 603 00:29:48 --> 00:29:52 What I'll do is I'll read you another list of nonsense 604 00:29:52 --> 00:29:54 like the first one. 605 00:29:54 --> 00:29:57 I'm going to ask you to write them down again, but this time 606 00:29:57 --> 00:30:00 at the end instead of saying write I'll say count. 607 00:30:00 --> 00:30:04 When I say count, I want you to count backwards from 608 00:30:04 --> 00:30:07 431 by 3's out loud. 609 00:30:07 --> 00:30:09 You know all in beautiful unison. 610 00:30:09 --> 00:30:11 Right got that? 611 00:30:11 --> 00:30:14 That would make a lot of noise, so when I do 612 00:30:14 --> 00:30:15 this, keep an eye on me. 613 00:30:15 --> 00:30:18 When I do this, write everything down. 614 00:30:18 --> 00:30:21 OK? 615 00:30:21 --> 00:30:22 OK. 616 00:30:22 --> 00:30:23 They kind of got the instructions. 617 00:30:23 --> 00:30:25 All right, where's the rest of my nonsense here? 618 00:30:25 --> 00:30:28 619 00:30:28 --> 00:30:29 AUDIENCE: [UNINTELLIGIBLE] 620 00:30:29 --> 00:30:30 PROFESSOR: OK. 621 00:30:30 --> 00:30:33 622 00:30:33 --> 00:30:34 OK ready? 623 00:30:34 --> 00:30:38 624 00:30:38 --> 00:30:40 Sip. 625 00:30:40 --> 00:30:43 Forth. 626 00:30:43 --> 00:30:46 Lig. 627 00:30:46 --> 00:30:48 Vop. 628 00:30:48 --> 00:30:51 Hearn. 629 00:30:51 --> 00:30:53 Mope. 630 00:30:53 --> 00:30:56 Jick. 631 00:30:56 --> 00:30:59 Tinned. 632 00:30:59 --> 00:31:01 Mez. 633 00:31:01 --> 00:31:04 Wamp. 634 00:31:04 --> 00:31:07 Flob. 635 00:31:07 --> 00:31:09 Gone. 636 00:31:09 --> 00:31:09 OK. 637 00:31:09 --> 00:31:11 Count 431. 638 00:31:11 --> 00:31:11 AUDIENCE: 639 00:31:11 --> 00:31:13 431. 640 00:31:13 --> 00:31:18 [UNINTELLIGIBLE] 641 00:31:18 --> 00:31:19 PROFESSOR: OK. 642 00:31:19 --> 00:31:20 Write them all down. 643 00:31:20 --> 00:31:21 The giggles would probably do just fine. 644 00:31:21 --> 00:31:41 AUDIENCE: [LAUGHTER] 645 00:31:41 --> 00:31:41 PROFESSOR: OK. 646 00:31:41 --> 00:31:46 The claim here is that what that should do, did you 647 00:31:46 --> 00:31:49 feel yourself rehearsing the first ones again? 648 00:31:49 --> 00:31:50 AUDIENCE: Yeah. 649 00:31:50 --> 00:31:53 PROFESSOR: The claim is that the primacy affect, which is 650 00:31:53 --> 00:31:57 due to some sort of rehearsal into long term memory. 651 00:31:57 --> 00:32:01 That that primacy effect should still be there. 652 00:32:01 --> 00:32:05 But the recency effect due to some sort of it's just 653 00:32:05 --> 00:32:09 thereness, should have been disrupted by the counting and 654 00:32:09 --> 00:32:10 the chordals and stuff like that. 655 00:32:10 --> 00:32:11 OK. 656 00:32:11 --> 00:32:14 So let's check here, where's my list gone to? 657 00:32:14 --> 00:32:15 Here list. 658 00:32:15 --> 00:32:17 I should remember this by now. 659 00:32:17 --> 00:32:17 OK. 660 00:32:17 --> 00:32:21 How many people got sip? 661 00:32:21 --> 00:32:21 And look. 662 00:32:21 --> 00:32:24 That's almost exactly the same number. 663 00:32:24 --> 00:32:27 How many people got forth? 664 00:32:27 --> 00:32:27 Ooh. 665 00:32:27 --> 00:32:30 A lot of forths. 666 00:32:30 --> 00:32:34 How many people got jick? 667 00:32:34 --> 00:32:35 Ouhh deeply pathetic. 668 00:32:35 --> 00:32:37 AUDIENCE: [LAUGHTER] 669 00:32:37 --> 00:32:42 PROFESSOR: How many people got mez? 670 00:32:42 --> 00:32:43 Few more. 671 00:32:43 --> 00:32:45 How many people got flob? 672 00:32:45 --> 00:32:49 Ouhh that's pretty pathetic. 673 00:32:49 --> 00:32:52 How many people got gone? 674 00:32:52 --> 00:32:53 No gone? 675 00:32:53 --> 00:32:55 Gone was pretty well gone. 676 00:32:55 --> 00:32:57 These were supposed to be circles, So they'd be 677 00:32:57 --> 00:32:58 a different data set. 678 00:32:58 --> 00:33:02 So anyway, nice primacy effect. 679 00:33:02 --> 00:33:04 Recency effect gone. 680 00:33:04 --> 00:33:04 AUDIENCE: [UNINTELLIGIBLE] 681 00:33:04 --> 00:33:05 PROFESSOR: Yeah. 682 00:33:05 --> 00:33:06 There are other words in there. 683 00:33:06 --> 00:33:10 This is the phenomenon of MIT students really wanting to do 684 00:33:10 --> 00:33:12 well on these sort of things right? 685 00:33:12 --> 00:33:13 AUDIENCE: [LAUGHTER] 686 00:33:13 --> 00:33:14 PROFESSOR: Man, he didn't ask me. 687 00:33:14 --> 00:33:17 It's like, you know, you have the same experience 688 00:33:17 --> 00:33:18 for keeps on the exam. 689 00:33:18 --> 00:33:21 You sit there and you study chapter 3 until you're blue. 690 00:33:21 --> 00:33:25 And oh man they only asked about chapter 2. 691 00:33:25 --> 00:33:26 So yeah, yeah, yeah. 692 00:33:26 --> 00:33:28 I'm glad you got whatever it was wamp or -- 693 00:33:28 --> 00:33:30 AUDIENCE: [LAUGHTER] 694 00:33:30 --> 00:33:34 PROFESSOR: The worst thing about this you realize is that 695 00:33:34 --> 00:33:37 the stuff that you rehearsed into long term memory may stay 696 00:33:37 --> 00:33:39 there for a surprisingly long time. 697 00:33:39 --> 00:33:42 I had a guy come up to me on the subway some years back. 698 00:33:42 --> 00:33:45 Say to me, you don't know me right? 699 00:33:45 --> 00:33:48 I said that's right I don't know you. 700 00:33:48 --> 00:33:50 He said fip, dut. 701 00:33:50 --> 00:33:53 AUDIENCE: [LAUGHTER] 702 00:33:53 --> 00:33:55 PROFESSOR: I said you are a very strange person. 703 00:33:55 --> 00:33:59 But I now know where I know you from, or don't you from 704 00:33:59 --> 00:34:04 or something like that. 705 00:34:04 --> 00:34:13 You may have now warped your brain on ongoing basis. 706 00:34:13 --> 00:34:13 No. 707 00:34:13 --> 00:34:15 I don't want you to go up. 708 00:34:15 --> 00:34:16 Get back here. 709 00:34:16 --> 00:34:16 Stop. 710 00:34:16 --> 00:34:17 Come back down. 711 00:34:17 --> 00:34:19 You go up. 712 00:34:19 --> 00:34:19 All right. 713 00:34:19 --> 00:34:25 So back on the first page of the handout, you have what was 714 00:34:25 --> 00:34:28 sometimes called a standard model for getting from the 715 00:34:28 --> 00:34:32 outside world -- whoops my bottleneck disappeared again 716 00:34:32 --> 00:34:35 -- into long term memory. 717 00:34:35 --> 00:34:41 Which was stuff comes into a short term memory. 718 00:34:41 --> 00:34:43 It needs to be preserved in short term memory by 719 00:34:43 --> 00:34:45 something like rehearsal. 720 00:34:45 --> 00:34:48 If it lasts long enough in short term memory, it gets 721 00:34:48 --> 00:34:54 into long term memory where it manages to stay. 722 00:34:54 --> 00:34:58 There are certain problems with this kind of a model, which you 723 00:34:58 --> 00:35:00 ought to be able to anticipate now. 724 00:35:00 --> 00:35:02 But we can illustrate easily enough. 725 00:35:02 --> 00:35:06 How many people here had breakfast? 726 00:35:06 --> 00:35:06 OK. 727 00:35:06 --> 00:35:09 Person with the MIT shirt on there. 728 00:35:09 --> 00:35:10 What did you have for breakfast? 729 00:35:10 --> 00:35:11 AUDIENCE: Bagel. 730 00:35:11 --> 00:35:12 PROFESSOR: Bagels. 731 00:35:12 --> 00:35:12 Anything else? 732 00:35:12 --> 00:35:14 AUDIENCE: Bagel and cream cheese. 733 00:35:14 --> 00:35:15 PROFESSOR: Bagels and cream cheese. 734 00:35:15 --> 00:35:15 Oh OK. 735 00:35:15 --> 00:35:15 That's good. 736 00:35:15 --> 00:35:19 We want a little bit of a memory load here. 737 00:35:19 --> 00:35:21 Because I need to be able to explain that the reason that 738 00:35:21 --> 00:35:24 she knows that she had bagels and cream cheese for breakfast 739 00:35:24 --> 00:35:29 is that on the way out of her living arrangements, she was 740 00:35:29 --> 00:35:31 going bagels and cream cheese, bagels and cream cheese, 741 00:35:31 --> 00:35:32 Bagel and cream cheese. 742 00:35:32 --> 00:35:33 AUDIENCE: [LAUGHTER] 743 00:35:33 --> 00:35:34 PROFESSOR: Right? 744 00:35:34 --> 00:35:35 OK. 745 00:35:35 --> 00:35:37 That's good. 746 00:35:37 --> 00:35:38 No. 747 00:35:38 --> 00:35:38 She wasn't doing that. 748 00:35:38 --> 00:35:40 You weren't doing that, but you remember what you 749 00:35:40 --> 00:35:44 had for breakfast too. 750 00:35:44 --> 00:35:48 Obviously this rehearsal thing while it's important for 751 00:35:48 --> 00:35:52 getting nonsense from the world into long term memory, is 752 00:35:52 --> 00:35:56 not the only way to get through this bottleneck. 753 00:35:56 --> 00:36:01 Again things with meaning manage to get there 754 00:36:01 --> 00:36:02 more readily. 755 00:36:02 --> 00:36:13 And we can see that that fact tells us something perhaps 756 00:36:13 --> 00:36:19 about the way that our long term memory is put together. 757 00:36:19 --> 00:36:21 Now here what we're talking about is what's know as 758 00:36:21 --> 00:36:24 explicit long term memories as distinct from implicit. 759 00:36:24 --> 00:36:27 Explicit long term memory are the memories that you can get 760 00:36:27 --> 00:36:29 to if I ask you about them. 761 00:36:29 --> 00:36:32 That you can recover. 762 00:36:32 --> 00:36:35 So if I say, you know, who's running for President, you have 763 00:36:35 --> 00:36:37 an explicit memory that the answer is Kerry/Bush and 764 00:36:37 --> 00:36:42 whoever else you can name down the line there. 765 00:36:42 --> 00:36:50 If I asked you how do you say the word president, what are 766 00:36:50 --> 00:36:53 the tongue motions that are required. 767 00:36:53 --> 00:36:56 If you say president, you can now monitor it and figure 768 00:36:56 --> 00:36:56 out what they are. 769 00:36:56 --> 00:37:01 But you have no notion ahead of time of how you do that. 770 00:37:01 --> 00:37:03 You certainly remember how to do it in the sense that you 771 00:37:03 --> 00:37:05 can produce it on demand. 772 00:37:05 --> 00:37:06 That would be an implicit memory. 773 00:37:06 --> 00:37:09 And how you ride a bicycle. 774 00:37:09 --> 00:37:12 There are a whole slew of things like motor memories 775 00:37:12 --> 00:37:13 that are implicit. 776 00:37:13 --> 00:37:15 But here what we're talking about is getting into an 777 00:37:15 --> 00:37:17 explicit long term memory. 778 00:37:17 --> 00:37:24 It's useful to divide explicit long term memory into episodic 779 00:37:24 --> 00:37:33 memories for the episodes of your life, and semantic 780 00:37:33 --> 00:37:37 memories, your body of knowledge and facts and 781 00:37:37 --> 00:37:38 things of that sort. 782 00:37:38 --> 00:37:41 There not like walled off from each other. 783 00:37:41 --> 00:37:49 If I say, what's the capital of the United States, you dig 784 00:37:49 --> 00:37:52 out of your semantic memory the answer Washington DC. 785 00:37:52 --> 00:37:55 If you've been to Washington, you might then start digging 786 00:37:55 --> 00:37:59 out associated episodic memories of being 787 00:37:59 --> 00:38:02 there and so on. 788 00:38:02 --> 00:38:07 But within semantic memory, the notion here -- remember the 789 00:38:07 --> 00:38:11 Uncle Charlie or whoever he was notion, that things can reach 790 00:38:11 --> 00:38:18 out to the world and help pull things into long term memory -- 791 00:38:18 --> 00:38:23 is related to the way that we think that your long term up 792 00:38:23 --> 00:38:25 memory might be organized. 793 00:38:25 --> 00:38:28 And one of the popular ways of thinking about this is sort of 794 00:38:28 --> 00:38:34 a giant network of associations called a semantic 795 00:38:34 --> 00:38:35 network often. 796 00:38:35 --> 00:38:39 797 00:38:39 --> 00:38:43 If I say cat, the first word that comes to your mind is? 798 00:38:43 --> 00:38:44 AUDIENCE: Meow. 799 00:38:44 --> 00:38:44 PROFESSOR: Meow. 800 00:38:44 --> 00:38:45 AUDIENCE: Dog. 801 00:38:45 --> 00:38:46 PROFESSOR: Dog. 802 00:38:46 --> 00:38:47 AUDIENCE: Mouse. 803 00:38:47 --> 00:38:47 PROFESSOR: OK. 804 00:38:47 --> 00:38:53 So there's a note in your long term memory that is cat in 805 00:38:53 --> 00:38:56 some sense in this story. 806 00:38:56 --> 00:39:07 It's close neighbors are things like dog and meow and -- 807 00:39:07 --> 00:39:10 what did we have -- mouse. 808 00:39:10 --> 00:39:12 And now it's important to realize this is not some sort 809 00:39:12 --> 00:39:16 of Linnaean taxonomy of, you know, species and geneus and 810 00:39:16 --> 00:39:18 all that sort of good stuff. 811 00:39:18 --> 00:39:20 You know, even though it's going to be connected with 812 00:39:20 --> 00:39:26 animal up here somewhere, animal and fur and stuff. 813 00:39:26 --> 00:39:28 But you might also have connections that 814 00:39:28 --> 00:39:32 go like meow, mow. 815 00:39:32 --> 00:39:37 AUDIENCE: [LAUGHTER] 816 00:39:37 --> 00:39:38 PROFESSOR: Actually that goes back here. 817 00:39:38 --> 00:39:41 I actually have a cat whose name is Chairman Mao by 818 00:39:41 --> 00:39:43 roughly this association. 819 00:39:43 --> 00:39:44 AUDIENCE: [LAUGHTER] 820 00:39:44 --> 00:39:47 PROFESSOR: You know, then, you know, China. 821 00:39:47 --> 00:39:51 AUDIENCE: [LAUGHTER] 822 00:39:51 --> 00:39:53 PROFESSOR: Plates. 823 00:39:53 --> 00:39:54 AUDIENCE: [LAUGHTER] 824 00:39:54 --> 00:40:00 PROFESSOR: So, you know, nobody has the notion that you could 825 00:40:00 --> 00:40:05 somehow map this out for any given individual. 826 00:40:05 --> 00:40:08 It's going to be a vastly complex and continuously 827 00:40:08 --> 00:40:12 changing set of associations. 828 00:40:12 --> 00:40:16 But there is evidence for this notion of proximity 829 00:40:16 --> 00:40:18 by association. 830 00:40:18 --> 00:40:23 By meaning in long term memory. 831 00:40:23 --> 00:40:28 And one of the ways you might find that out is by doing what 832 00:40:28 --> 00:40:32 are called priming experiments, which I might as well mention 833 00:40:32 --> 00:40:36 because I see that's -- is that on the handout properly. 834 00:40:36 --> 00:40:38 Well if it isn't, it should be. 835 00:40:38 --> 00:40:41 It probably says priming somewhere. 836 00:40:41 --> 00:40:44 Priming experiment. 837 00:40:44 --> 00:40:45 Evidence from priming. 838 00:40:45 --> 00:40:46 Look at that. 839 00:40:46 --> 00:40:49 840 00:40:49 --> 00:40:57 On my computer screen here, I'll put up a string of 841 00:40:57 --> 00:41:01 letters, and you tell me whether or not it's a word. 842 00:41:01 --> 00:41:01 Right? 843 00:41:01 --> 00:41:04 So I put up. 844 00:41:04 --> 00:41:05 And you say? 845 00:41:05 --> 00:41:06 AUDIENCE: [UNINTELLIGIBLE] 846 00:41:06 --> 00:41:07 PROFESSOR: No. 847 00:41:07 --> 00:41:10 Even though it's now in your long term memory right. 848 00:41:10 --> 00:41:11 But if I put up. 849 00:41:11 --> 00:41:15 850 00:41:15 --> 00:41:16 Yes. 851 00:41:16 --> 00:41:17 OK now. 852 00:41:17 --> 00:41:27 If the next word was something like truck, unless you have a 853 00:41:27 --> 00:41:29 dog in your truck or something like that. 854 00:41:29 --> 00:41:33 855 00:41:33 --> 00:41:36 Well all right, let's compare this. 856 00:41:36 --> 00:41:40 The next word might be truck, or the next word might be cat. 857 00:41:40 --> 00:41:42 You'll be faster to say cat. 858 00:41:42 --> 00:41:44 You'll be faster to confirm that cat is a word 859 00:41:44 --> 00:41:46 than truck is a word. 860 00:41:46 --> 00:41:48 Why is that? 861 00:41:48 --> 00:41:52 Well seemingly when you go in to your semantic network to 862 00:41:52 --> 00:42:01 discover that dog is a word, you light up the dog node, and 863 00:42:01 --> 00:42:05 the activities spreads -- it's called spreading activation -- 864 00:42:05 --> 00:42:10 away from that node to the neighboring nodes. 865 00:42:10 --> 00:42:15 As a result, when you get the word cat, cat's already 866 00:42:15 --> 00:42:17 a little activated. 867 00:42:17 --> 00:42:19 The bell's been rung a little bit already. 868 00:42:19 --> 00:42:23 And your quicker to confirm that cat is a word than truck. 869 00:42:23 --> 00:42:27 Which is you know, down here in, you know, in the 870 00:42:27 --> 00:42:29 next county somewhere. 871 00:42:29 --> 00:42:33 872 00:42:33 --> 00:42:36 In principle I suppose you could map out somebody's 873 00:42:36 --> 00:42:39 semantic network with a technique like this. 874 00:42:39 --> 00:42:42 But it would be a lifetime's endeavor ever, and not 875 00:42:42 --> 00:42:43 clear what the point is. 876 00:42:43 --> 00:42:47 But you could you can use this sort of a technique to show 877 00:42:47 --> 00:42:50 that some things live closer to each other than other things. 878 00:42:50 --> 00:42:55 Oh and by the way, when you go and reach in to retrieve -- 879 00:42:55 --> 00:42:56 this is sort of jumping to the retrieval part -- but when 880 00:42:56 --> 00:43:04 you go reach in to retrieve something from one node, you 881 00:43:04 --> 00:43:07 might goof and pull something from the neighboring node. 882 00:43:07 --> 00:43:11 And you know this, or you have experienced this if you are not 883 00:43:11 --> 00:43:13 an only child perhaps, even if you are. 884 00:43:13 --> 00:43:15 How many do you have siblings? 885 00:43:15 --> 00:43:16 Keep your hands up. 886 00:43:16 --> 00:43:19 How many of those have ever had your parents call you 887 00:43:19 --> 00:43:21 by the sibling's name? 888 00:43:21 --> 00:43:22 Almost everybody. 889 00:43:22 --> 00:43:23 Including some people who didn't have their 890 00:43:23 --> 00:43:24 hands up before. 891 00:43:24 --> 00:43:24 AUDIENCE: [LAUGHTER] 892 00:43:24 --> 00:43:27 PROFESSOR: Who didn't know that they had siblings 893 00:43:27 --> 00:43:30 until that happened. 894 00:43:30 --> 00:43:32 Or worse, you know, you get called by the name of the 895 00:43:32 --> 00:43:35 dog or something like that. 896 00:43:35 --> 00:43:39 This is not because whatever you may believe, you know, 897 00:43:39 --> 00:43:42 that your parents are unusually dim people. 898 00:43:42 --> 00:43:48 It's because all those terms presumably live very close to 899 00:43:48 --> 00:43:50 each in this semantic memory. 900 00:43:50 --> 00:43:53 And when particularly under some pressure you reach into 901 00:43:53 --> 00:43:56 grab kid number -- all my kids have the same name. 902 00:43:56 --> 00:44:01 Their name is Ben Phillip Simon, whatever your name is. 903 00:44:01 --> 00:44:06 And it works for all of them. 904 00:44:06 --> 00:44:08 Now there are certain drawbacks to this. 905 00:44:08 --> 00:44:13 It turns out to be decidedly unfortunate if in a moment of 906 00:44:13 --> 00:44:17 passion you murmur the name of the last girlfriend. 907 00:44:17 --> 00:44:17 AUDIENCE: [LAUGHTER] 908 00:44:17 --> 00:44:19 PROFESSOR: No matter how close they live in 909 00:44:19 --> 00:44:21 your semantic network. 910 00:44:21 --> 00:44:24 911 00:44:24 --> 00:44:28 We can explain it, but we can't excuse it if you know what I. 912 00:44:28 --> 00:44:31 913 00:44:31 --> 00:44:38 So big world, small desktop, small working memory, 914 00:44:38 --> 00:44:41 big long term memory. 915 00:44:41 --> 00:44:44 916 00:44:44 --> 00:44:48 What's going on when the stuff in long term memory gets turned 917 00:44:48 --> 00:44:51 into something long term? 918 00:44:51 --> 00:44:56 It's really there for a long, long time. 919 00:44:56 --> 00:45:01 If you get a chance, David Pillemer's article that's in 920 00:45:01 --> 00:45:05 the website under the memory category about very long term 921 00:45:05 --> 00:45:10 memories, is fun to read as he goes and probes people's oldest 922 00:45:10 --> 00:45:15 memories from way, way back. 923 00:45:15 --> 00:45:17 How does stuff get consolidated. 924 00:45:17 --> 00:45:19 Well we know some things about this. 925 00:45:19 --> 00:45:20 We know quite a lot about this. 926 00:45:20 --> 00:45:24 But I want to tell you one bit, a couple of bits, that are 927 00:45:24 --> 00:45:27 important on a sort of a physiological front. 928 00:45:27 --> 00:45:30 So here's the brain again. 929 00:45:30 --> 00:45:32 Underneath the temporal lobe, not in the cortex of the 930 00:45:32 --> 00:45:40 temporal lobe, but underneath the temporal lobe struc the 931 00:45:40 --> 00:45:46 hippocampus, and to a lesser extent the amygdala. 932 00:45:46 --> 00:45:48 I'll just mention. 933 00:45:48 --> 00:45:51 But really we're talking about hippocampus here. 934 00:45:51 --> 00:45:55 Parts of the so called Limbic System that are vitally 935 00:45:55 --> 00:45:58 important in this process of consolidation. 936 00:45:58 --> 00:46:05 And we know that in the first instance from perhaps the 937 00:46:05 --> 00:46:10 most famous patient in the neuropsychological literature. 938 00:46:10 --> 00:46:15 He's a patient known in the literature as HM. 939 00:46:15 --> 00:46:18 He's the inspiration for the film Memento if 940 00:46:18 --> 00:46:18 you saw Memento. 941 00:46:18 --> 00:46:21 942 00:46:21 --> 00:46:28 When he was a young man he had a bike accident that was 943 00:46:28 --> 00:46:33 probably the cause of his epilepsy. 944 00:46:33 --> 00:46:36 So he was an epileptic. 945 00:46:36 --> 00:46:39 You may recall I said earlier in the course that epilepsy is 946 00:46:39 --> 00:46:43 a sort of an electrical storm often started from a damaged 947 00:46:43 --> 00:46:46 piece of tissue in the brain. 948 00:46:46 --> 00:46:49 If it's not responding to drugs, which this was not, one 949 00:46:49 --> 00:46:53 of the treatment is to go and try to excise, to take 950 00:46:53 --> 00:46:55 out, the generator. 951 00:46:55 --> 00:47:02 The evidence in HM's case pointed to structures deep 952 00:47:02 --> 00:47:03 in the temporal lobe. 953 00:47:03 --> 00:47:06 Particularly the hippocampus and amygdala. 954 00:47:06 --> 00:47:12 And what was done in the late 50's, in the late 50's or 955 00:47:12 --> 00:47:14 early 60's, anybody remember? 956 00:47:14 --> 00:47:20 Anyway a long time ago, at this point, up in Montreal was that 957 00:47:20 --> 00:47:23 these structures were largely destroyed on 958 00:47:23 --> 00:47:25 both sides of his brain. 959 00:47:25 --> 00:47:25 That's important. 960 00:47:25 --> 00:47:29 This is a bilateral lesion. 961 00:47:29 --> 00:47:33 And this did have the effect of largely controlling 962 00:47:33 --> 00:47:34 his seizures. 963 00:47:34 --> 00:47:36 The seizure problem went away. 964 00:47:36 --> 00:47:42 However, he's a one of a kind patient, because the side 965 00:47:42 --> 00:47:46 effects, the unintentional consequences of this lesion, 966 00:47:46 --> 00:47:50 was so devastating that nobody could ethically 967 00:47:50 --> 00:47:52 ever do this again. 968 00:47:52 --> 00:47:56 At least not to a human patient. 969 00:47:56 --> 00:47:59 970 00:47:59 --> 00:48:03 So he's still alive. 971 00:48:03 --> 00:48:06 And in fact, he's in a nursing facility near Boston. 972 00:48:06 --> 00:48:12 Because he was brought to Boston essentially as a 973 00:48:12 --> 00:48:15 psychological subject. 974 00:48:15 --> 00:48:17 Been unable to live on his own since the surgery. 975 00:48:17 --> 00:48:18 What's his problem? 976 00:48:18 --> 00:48:19 Well, he really has two. 977 00:48:19 --> 00:48:25 But the most relevant one for the present purposes is that he 978 00:48:25 --> 00:48:31 is simply unable to form new explicit long term memories. 979 00:48:31 --> 00:48:35 980 00:48:35 --> 00:48:41 Take his interaction with a once upon a time TA in this 981 00:48:41 --> 00:48:44 course name John Gabrieli now a Professor at Stanford soon 982 00:48:44 --> 00:48:47 to be a Professor back here at MIT. 983 00:48:47 --> 00:48:50 But who did, I think he did his doctral work studying HM. 984 00:48:50 --> 00:48:52 But he certainly studied HM as a graduate student. 985 00:48:52 --> 00:48:59 He'd come into the lab to meet HM. 986 00:48:59 --> 00:49:01 Say hi, you know, have we've met before. 987 00:49:01 --> 00:49:02 HM said no. 988 00:49:02 --> 00:49:03 I don't think so. 989 00:49:03 --> 00:49:06 He's perfectly intact short term memory, which is fine for 990 00:49:06 --> 00:49:08 conducting a conversation. 991 00:49:08 --> 00:49:14 Also decently intact memories, this long term memory from 992 00:49:14 --> 00:49:16 before the operation. 993 00:49:16 --> 00:49:19 So it's not that we've somehow wiped out his 994 00:49:19 --> 00:49:20 storehouse of memories. 995 00:49:20 --> 00:49:23 It's the ability to put new stuff in here that's gone. 996 00:49:23 --> 00:49:25 So in comes Gabrieli. 997 00:49:25 --> 00:49:26 Have we ever met before? 998 00:49:26 --> 00:49:26 No. 999 00:49:26 --> 00:49:31 Hi I'm John Gabrieli, you know, it's nice to meet you. 1000 00:49:31 --> 00:49:32 I'm HM, well I suppose he didn't call himself 1001 00:49:32 --> 00:49:34 HM, but anyway. 1002 00:49:34 --> 00:49:38 You use initials to protect the privacy of the patient. 1003 00:49:38 --> 00:49:40 Not necessarily the initials of the patient. 1004 00:49:40 --> 00:49:42 Something to label them. 1005 00:49:42 --> 00:49:45 So anyway, fine have a nice conversation. 1006 00:49:45 --> 00:49:46 Gabrieli goes out. 1007 00:49:46 --> 00:49:47 Comes back in. 1008 00:49:47 --> 00:49:49 Says hi, have we ever met before. 1009 00:49:49 --> 00:49:50 No I don't think so. 1010 00:49:50 --> 00:49:51 Well hi. 1011 00:49:51 --> 00:49:52 I'm John Gabrieli nice to meet. 1012 00:49:52 --> 00:49:54 Have the same conversation. 1013 00:49:54 --> 00:49:56 You can do this over and over again. 1014 00:49:56 --> 00:50:00 1015 00:50:00 --> 00:50:03 In a more control kind of way, do a digit span 1016 00:50:03 --> 00:50:04 kind of experiment. 1017 00:50:04 --> 00:50:06 Give him a few digits to remember. 1018 00:50:06 --> 00:50:07 No problem. 1019 00:50:07 --> 00:50:09 He's got the same 7 plus or minus 2 that the rest 1020 00:50:09 --> 00:50:12 of us have pretty much. 1021 00:50:12 --> 00:50:15 Distract him momentarily, and it's just gone. 1022 00:50:15 --> 00:50:17 You've had this experience yourself if you try to get 1023 00:50:17 --> 00:50:21 from the phone book to the phone with a phone number. 1024 00:50:21 --> 00:50:24 And somebody says, didd you do x, y and z? 1025 00:50:24 --> 00:50:25 Ahh. 1026 00:50:25 --> 00:50:27 There goes the phone number. 1027 00:50:27 --> 00:50:27 AUDIENCE: [LAUGHTER] 1028 00:50:27 --> 00:50:28 PROFESSOR: 1029 00:50:28 --> 00:50:32 That's his continuous experience, and has been for 1030 00:50:32 --> 00:50:34 50 years at this point. 1031 00:50:34 --> 00:50:36 A few things have gotten in. 1032 00:50:36 --> 00:50:38 He happens to know that men have landed on the 1033 00:50:38 --> 00:50:41 moon for instance. 1034 00:50:41 --> 00:50:43 Perhaps because they're a little bits of the structure 1035 00:50:43 --> 00:50:45 that are maintained perhaps from another route. 1036 00:50:45 --> 00:50:47 We're not quite sure. 1037 00:50:47 --> 00:50:52 But basically no new episodic memories. 1038 00:50:52 --> 00:50:57 What this is telling us is that the hippocampus in particular, 1039 00:50:57 --> 00:51:00 other research indicates is critical for the act 1040 00:51:00 --> 00:51:02 of consolidation. 1041 00:51:02 --> 00:51:04 It is not the memory itself. 1042 00:51:04 --> 00:51:10 It is the mechanism that allows you to make those memories 1043 00:51:10 --> 00:51:11 solid in some fashion. 1044 00:51:11 --> 00:51:11 Yes? 1045 00:51:11 --> 00:51:14 AUDIENCE: But, does he know he has this disorder? 1046 00:51:14 --> 00:51:18 PROFESSOR: He knows he's got a memory problem. 1047 00:51:18 --> 00:51:19 And this actually is a good point. 1048 00:51:19 --> 00:51:23 He ties into his other problem, which is in some sense 1049 00:51:23 --> 00:51:25 an off setting benefits. 1050 00:51:25 --> 00:51:30 The Limbic System, certainly the amygdala, very important 1051 00:51:30 --> 00:51:32 in the mediation of emotion. 1052 00:51:32 --> 00:51:34 His experience of emotion is greatly flattened. 1053 00:51:34 --> 00:51:38 Flattened affect in the jargon of the trade. 1054 00:51:38 --> 00:51:40 So. 1055 00:51:40 --> 00:51:40 Which is good. 1056 00:51:40 --> 00:51:42 Because he does know that there's something wrong, but it 1057 00:51:42 --> 00:51:44 doesn't bother him that much. 1058 00:51:44 --> 00:51:46 You know what bothers him? 1059 00:51:46 --> 00:51:50 What bothers him is looking in the mirror. 1060 00:51:50 --> 00:51:52 Now why should that be? 1061 00:51:52 --> 00:51:53 Well imagine this. 1062 00:51:53 --> 00:51:57 His lesion happened when he's in his 20s. 1063 00:51:57 --> 00:52:00 He's still got the long term memories that 1064 00:52:00 --> 00:52:02 he had in his 20s. 1065 00:52:02 --> 00:52:03 Basically your age. 1066 00:52:03 --> 00:52:07 Now you go back to your room, you look in the mirror, and 1067 00:52:07 --> 00:52:11 what looks out at you is a 70 year old guy or women 1068 00:52:11 --> 00:52:12 as the case may 1069 00:52:12 --> 00:52:12 be. 1070 00:52:12 --> 00:52:14 I suppose it doesn't particularly matter. 1071 00:52:14 --> 00:52:16 It's going to be a disaster under either 1072 00:52:16 --> 00:52:17 circumstance right? 1073 00:52:17 --> 00:52:21 You're going to think I'm in a sci-fi movie, and 1074 00:52:21 --> 00:52:23 I don't have the part. 1075 00:52:23 --> 00:52:26 You know the good guy with the laser gun part 1076 00:52:26 --> 00:52:27 that I really wanted. 1077 00:52:27 --> 00:52:30 You know I'm in big trouble here. 1078 00:52:30 --> 00:52:35 And this is disturbing to him. 1079 00:52:35 --> 00:52:37 Not as disturbing as it would be to you again, because his 1080 00:52:37 --> 00:52:40 emotional responses has been very severely blunted. 1081 00:52:40 --> 00:52:46 So much so that he needs to be asked whether or not he's sick. 1082 00:52:46 --> 00:52:48 You know do you have a stomach ache? 1083 00:52:48 --> 00:52:50 No that you mentioned it, yes I do. 1084 00:52:50 --> 00:52:52 It's not that he couldn't feel the pain, it's just without the 1085 00:52:52 --> 00:52:56 emotional overlay so, you know. 1086 00:52:56 --> 00:52:57 I see blue. 1087 00:52:57 --> 00:52:58 I have pain what's the big deal? 1088 00:52:58 --> 00:52:58 AUDIENCE: [LAUGHTER] 1089 00:52:58 --> 00:52:59 PROFESSOR: Yes? 1090 00:52:59 --> 00:53:00 AUDIENCE: [UNINTELLIGIBLE] 1091 00:53:00 --> 00:53:01 motor memory 1092 00:53:01 --> 00:53:03 PROFESSOR: Yes. 1093 00:53:03 --> 00:53:05 Nice point. 1094 00:53:05 --> 00:53:09 So for example, I don't know how many of you have hung out 1095 00:53:09 --> 00:53:14 at -- you know get rid of the brain here. 1096 00:53:14 --> 00:53:16 Who needs the brain? 1097 00:53:16 --> 00:53:21 So any you ever done a mirror maze? 1098 00:53:21 --> 00:53:22 Some people may have actually built one of 1099 00:53:22 --> 00:53:23 these once upon a time. 1100 00:53:23 --> 00:53:26 So imagine you have like a star. 1101 00:53:26 --> 00:53:26 Oops. 1102 00:53:26 --> 00:53:29 Imagine you could actually draw a star. 1103 00:53:29 --> 00:53:32 AUDIENCE: [LAUGHTER] 1104 00:53:32 --> 00:53:33 PROFESSOR: All right. 1105 00:53:33 --> 00:53:33 Well. 1106 00:53:33 --> 00:53:36 1107 00:53:36 --> 00:53:39 So you make this out of like aluminum foil, and you go and 1108 00:53:39 --> 00:53:40 trace it with a metal stylus. 1109 00:53:40 --> 00:53:44 And if you go off the aluminum foil it makes a nasty noise. 1110 00:53:44 --> 00:53:47 Because you're breaking a circuit or something like that. 1111 00:53:47 --> 00:53:49 Anybody ever build one of those? 1112 00:53:49 --> 00:53:51 Oh what's the world coming to. 1113 00:53:51 --> 00:53:52 It's easy right? 1114 00:53:52 --> 00:53:53 Anybody can do this. 1115 00:53:53 --> 00:53:58 What people can't do regular readily is do this looking 1116 00:53:58 --> 00:54:01 at the image in a mirror. 1117 00:54:01 --> 00:54:01 Right? 1118 00:54:01 --> 00:54:05 Because then think they got [UNINTELLIGIBLE]. 1119 00:54:05 --> 00:54:07 AUDIENCE: [LAUGHTER] 1120 00:54:07 --> 00:54:10 PROFESSOR: But you learn to do this. 1121 00:54:10 --> 00:54:14 And HM has been trained in one set of experiments to do this. 1122 00:54:14 --> 00:54:18 In multiple sessions he got better. 1123 00:54:18 --> 00:54:20 If you asked him have you ever done this before on the 1124 00:54:20 --> 00:54:22 nth time that he did it. 1125 00:54:22 --> 00:54:23 The answer is no. 1126 00:54:23 --> 00:54:25 I don't think so. 1127 00:54:25 --> 00:54:27 And he goes [UNINTELLIGIBLE]. 1128 00:54:27 --> 00:54:29 Hey, you know, that was really good. 1129 00:54:29 --> 00:54:31 Most people asked how did you do that? 1130 00:54:31 --> 00:54:32 Yeah. 1131 00:54:32 --> 00:54:33 I've always been good at that kind of thing. 1132 00:54:33 --> 00:54:35 Or, you know, just lucky. 1133 00:54:35 --> 00:54:40 No episodic memory, no explicit memory of ever having done it. 1134 00:54:40 --> 00:54:43 But yes a motor memory that allows them to do it. 1135 00:54:43 --> 00:54:50 Similarly, your sense of familiarity runs on 1136 00:54:50 --> 00:54:51 different pathways. 1137 00:54:51 --> 00:54:56 So if I bumped into you wandering around the halls, I 1138 00:54:56 --> 00:55:01 may well have the sense that I've seen you somewhere before. 1139 00:55:01 --> 00:55:08 And I might guess that that's, you know, I've seen you in 1140 00:55:08 --> 00:55:10 intro psych, because that's where I see a lot of people 1141 00:55:10 --> 00:55:13 whose names I don't really quite know and stuff like that. 1142 00:55:13 --> 00:55:14 So that would work. 1143 00:55:14 --> 00:55:17 And HM gets that sense of familiarity too. 1144 00:55:17 --> 00:55:23 So after a while actually with Gabrieli, Gabrieli comes in, 1145 00:55:23 --> 00:55:24 have we ever met before? 1146 00:55:24 --> 00:55:24 Yeah. 1147 00:55:24 --> 00:55:25 You looks familiar. 1148 00:55:25 --> 00:55:26 Do you know my name? 1149 00:55:26 --> 00:55:26 Know you're name? 1150 00:55:26 --> 00:55:28 Forgotten your name. 1151 00:55:28 --> 00:55:29 Who do you think I am? 1152 00:55:29 --> 00:55:31 I think you're probably somebody from my 1153 00:55:31 --> 00:55:33 high school class. 1154 00:55:33 --> 00:55:34 Now why would that be. 1155 00:55:34 --> 00:55:36 Well Gabrieli at the time was a young 20 1156 00:55:36 --> 00:55:37 something year old guy. 1157 00:55:37 --> 00:55:42 This guy's got a 20 something year old long term memory. 1158 00:55:42 --> 00:55:44 A reasonable guess of somebody who looks more or less 1159 00:55:44 --> 00:55:48 familiar, but isn't in my current collection of 1160 00:55:48 --> 00:55:49 people I know well. 1161 00:55:49 --> 00:55:52 It's yeah maybe from high school, or something like that. 1162 00:55:52 --> 00:55:58 We're about the same age as the 70 year old HM to the 20 1163 00:55:58 --> 00:56:01 something year old Gabrieli. 1164 00:56:01 --> 00:56:04 So those sort of things he does learn. 1165 00:56:04 --> 00:56:09 It's that new explicit long term memories that require the 1166 00:56:09 --> 00:56:14 hippocampus to be consolidated. 1167 00:56:14 --> 00:56:15 How long does this take? 1168 00:56:15 --> 00:56:16 Let me say a quick word about that, and then 1169 00:56:16 --> 00:56:20 we'll take a quick break. 1170 00:56:20 --> 00:56:24 You can measure the time course of consolidation by doing an 1171 00:56:24 --> 00:56:28 experiment of the following sort. 1172 00:56:28 --> 00:56:31 You can draw bad rats. 1173 00:56:31 --> 00:56:34 1174 00:56:34 --> 00:56:36 OK there's a rat. 1175 00:56:36 --> 00:56:37 He's on a little pedestal. 1176 00:56:37 --> 00:56:43 If you were a rat on a small pedestal not too far above 1177 00:56:43 --> 00:56:46 the floor, what are you going to do? 1178 00:56:46 --> 00:56:48 AUDIENCE: Jump off. 1179 00:56:48 --> 00:56:49 PROFESSOR: Jump off, right. 1180 00:56:49 --> 00:56:49 I'm a rat. 1181 00:56:49 --> 00:56:51 Man I got to go look around at stuff. 1182 00:56:51 --> 00:56:54 So I jump off. 1183 00:56:54 --> 00:56:58 Well what they didn't tell me back at home was that this 1184 00:56:58 --> 00:57:01 particular floor is electrified. 1185 00:57:01 --> 00:57:06 So when I jump off my little toes get an unpleasant, 1186 00:57:06 --> 00:57:08 not damaging, but an unpleasant shock. 1187 00:57:08 --> 00:57:12 So now the guy puts me back on here. 1188 00:57:12 --> 00:57:12 What do I do? 1189 00:57:12 --> 00:57:14 AUDIENCE: Jump off. 1190 00:57:14 --> 00:57:15 PROFESSOR: I don't jump off. 1191 00:57:15 --> 00:57:16 I'm not a stupid rat. 1192 00:57:16 --> 00:57:17 I'm just a rat. 1193 00:57:17 --> 00:57:19 And so I stay there. 1194 00:57:19 --> 00:57:21 OK, well. 1195 00:57:21 --> 00:57:26 I'm a rat who's in an experiment. 1196 00:57:26 --> 00:57:29 And the question is what percentage of the 1197 00:57:29 --> 00:57:37 time do I jump off? 1198 00:57:37 --> 00:57:41 1199 00:57:41 --> 00:57:48 And the variable here is after I jump off, and after I get my 1200 00:57:48 --> 00:57:52 toes shocked, I'm also going to get a dose of what's called 1201 00:57:52 --> 00:57:55 electroconvulsive shock. 1202 00:57:55 --> 00:57:59 Which is basically electrical current run through the brain. 1203 00:57:59 --> 00:58:03 You know you can kill people or animals that way. 1204 00:58:03 --> 00:58:05 Obviously that's not the point here. 1205 00:58:05 --> 00:58:08 This is to disrupt the on going electrical 1206 00:58:08 --> 00:58:10 activity of the brain. 1207 00:58:10 --> 00:58:12 Not to scramble the structure. 1208 00:58:12 --> 00:58:14 Not to cook proteins. 1209 00:58:14 --> 00:58:20 But just to disrupt any ongoing electrical activity. 1210 00:58:20 --> 00:58:24 So this is the delay to ECS. 1211 00:58:24 --> 00:58:26 1212 00:58:26 --> 00:58:30 Rat steps down when does he get his ECS? 1213 00:58:30 --> 00:58:37 If he gets the ECS immediately, put him back on the platform. 1214 00:58:37 --> 00:58:39 Basically a 100% of the time the rat will step down 1215 00:58:39 --> 00:58:40 the next time to. 1216 00:58:40 --> 00:58:41 Why? 1217 00:58:41 --> 00:58:45 Not because the ECS is like fun or something like that. 1218 00:58:45 --> 00:58:49 But because the ECS has wiped out the memory. 1219 00:58:49 --> 00:58:53 The memory is not consolidated and the rat simply doesn't know 1220 00:58:53 --> 00:58:55 that his little feet got fried. 1221 00:58:55 --> 00:58:58 And he steps down again. 1222 00:58:58 --> 00:59:02 As the delay gets longer, you get an essentially exponential 1223 00:59:02 --> 00:59:09 looking curve with a time course in this case of seconds 1224 00:59:09 --> 00:59:13 that represents the time of consolidation of that memory. 1225 00:59:13 --> 00:59:16 At least to the point of supporting the behavior 1226 00:59:16 --> 00:59:18 of not stepping down. 1227 00:59:18 --> 00:59:22 but it's probably better to think it's not the case that 1228 00:59:22 --> 00:59:25 all memories consolidate within seconds. 1229 00:59:25 --> 00:59:29 There's evidence from human studies. 1230 00:59:29 --> 00:59:31 Now why would you do human electroconvulsive 1231 00:59:31 --> 00:59:33 shock studies? 1232 00:59:33 --> 00:59:37 The reason is that electroconvulsive shock in 1233 00:59:37 --> 00:59:42 humans is electroconvulsive therapy. 1234 00:59:42 --> 00:59:45 AUDIENCE: [LAUGHTER] 1235 00:59:45 --> 00:59:47 PROFESSOR: It sounds very odd. 1236 00:59:47 --> 00:59:47 It sounds very odd. 1237 00:59:47 --> 00:59:51 And those of you who've seen the now quite old movie to 1238 00:59:51 --> 00:59:53 kill, Not To Kill A Mockingbird, One Flew 1239 00:59:53 --> 00:59:54 Over The Cuckoo's Nest. 1240 00:59:54 --> 00:59:56 One of those bird movies. 1241 00:59:56 --> 01:00:00 One Flew Over The Cuckoo's Nest probably have a dim view of 1242 01:00:00 --> 01:00:03 electroconvulsive therapy. 1243 01:00:03 --> 01:00:05 But that's badly earned. 1244 01:00:05 --> 01:00:09 Int he case of intractable both depression. 1245 01:00:09 --> 01:00:14 Depression that's not being broken up successfully by 1246 01:00:14 --> 01:00:18 medication or by psychotherapy. 1247 01:00:18 --> 01:00:22 Severe depression of that sort carries with it a very serious 1248 01:00:22 --> 01:00:24 risk of mortality from suicide. 1249 01:00:24 --> 01:00:27 So this is not a, you know, casual disorder. 1250 01:00:27 --> 01:00:31 It turns out that electroconvulsive therapy is 1251 01:00:31 --> 01:00:37 often very effective in breaking up such depressions. 1252 01:00:37 --> 01:00:42 And is actually is useful part of the spectrum of treatments 1253 01:00:42 --> 01:00:47 that are available in this case. 1254 01:00:47 --> 01:00:50 Like most medical treatment, it's not entirely cost free. 1255 01:00:50 --> 01:00:55 It is the case that there are memory consequences of 1256 01:00:55 --> 01:00:56 electroconvulsive therapy. 1257 01:00:56 --> 01:00:59 It's part of what actually gives it a somewhat 1258 01:00:59 --> 01:01:01 bad reputation. 1259 01:01:01 --> 01:01:06 And you can actually see those memory consequences. 1260 01:01:06 --> 01:01:08 You can see traces of them going back years. 1261 01:01:08 --> 01:01:14 1262 01:01:14 --> 01:01:17 This is in studies for instance of memory for 1263 01:01:17 --> 01:01:19 TV shows seen only once. 1264 01:01:19 --> 01:01:21 You know sort of thing that you might dimly remember under 1265 01:01:21 --> 01:01:23 the best of circumstances. 1266 01:01:23 --> 01:01:27 These rather limited kind of memories can be damaged 1267 01:01:27 --> 01:01:29 by a electroconvulsive therapy years later. 1268 01:01:29 --> 01:01:34 As if this exponential has a very long tail. 1269 01:01:34 --> 01:01:39 And when you want to call it consolidated is perhaps a 1270 01:01:39 --> 01:01:41 function of may be something like how important 1271 01:01:41 --> 01:01:42 the memory is. 1272 01:01:42 --> 01:01:47 If I do this again, I'm going to be hurt right now. 1273 01:01:47 --> 01:01:50 That memory can be made solid enough relatively quickly that 1274 01:01:50 --> 01:01:52 you don't go and do it again. 1275 01:01:52 --> 01:01:57 On the other hand, if the memory is, was it Bridgette 1276 01:01:57 --> 01:02:02 Loves Bernie, or Benji. 1277 01:02:02 --> 01:02:04 All right that kind of good. 1278 01:02:04 --> 01:02:06 You know no stakes kind of memory is apparently 1279 01:02:06 --> 01:02:08 still vulnerable. 1280 01:02:08 --> 01:02:10 At least still partially vulnerable going back 1281 01:02:10 --> 01:02:11 for a very long time. 1282 01:02:11 --> 01:02:13 The hippocampus seems to be behaving like something 1283 01:02:13 --> 01:02:15 like scaffolding. 1284 01:02:15 --> 01:02:19 It holds the bits of the memory in place until it's firm 1285 01:02:19 --> 01:02:21 enough to stand on its own. 1286 01:02:21 --> 01:02:28 And then you can really call it in the sense of sort 1287 01:02:28 --> 01:02:30 of a permanent memory. 1288 01:02:30 --> 01:02:34 Now that doesn't solve the issue of getting 1289 01:02:34 --> 01:02:36 stuff out of memory. 1290 01:02:36 --> 01:02:41 And I will say a word or two about that momentarily. 1291 01:02:41 --> 01:02:43 But first you can wipe out your short term memory by 1292 01:02:43 --> 01:02:44 stretching for a second. 1293 01:02:44 --> 01:02:58 1294 01:02:58 --> 01:04:23 [SIDE___CONVERSA TION___WITH___STUDENT] 1295 01:04:23 --> 01:04:23 OK. 1296 01:04:23 --> 01:04:28 1297 01:04:28 --> 01:04:32 Looking at the handout, let me just make sure that I 1298 01:04:32 --> 01:04:35 explained the jargon here. 1299 01:04:35 --> 01:04:39 So a loss of memory as everybody knows is an amnesia. 1300 01:04:39 --> 01:04:46 A loss of memory for events prior to the point of injury 1301 01:04:46 --> 01:04:49 would be a retrograde amnesia. 1302 01:04:49 --> 01:04:52 That's not what HM has. 1303 01:04:52 --> 01:04:55 HM has an anterograde amnesia. 1304 01:04:55 --> 01:04:59 A problem with his memory subsequent to the lesion, 1305 01:04:59 --> 01:05:03 subsequent to the injury. 1306 01:05:03 --> 01:05:08 We won't do this is a demo, but if I were to take a baseball 1307 01:05:08 --> 01:05:11 bat and pop her on the head. 1308 01:05:11 --> 01:05:13 Well apart from the fact that I probably wouldn't 1309 01:05:13 --> 01:05:15 be back on a Thursday 1310 01:05:15 --> 01:05:18 AUDIENCE: [LAUGHTER] 1311 01:05:18 --> 01:05:23 PROFESSOR: When she came to, she might not remember having 1312 01:05:23 --> 01:05:27 actually been hit, because the set of memories from 1313 01:05:27 --> 01:05:31 immediately before that whop would have been wiped out in 1314 01:05:31 --> 01:05:32 a retrograde amnesia. 1315 01:05:32 --> 01:05:37 It is also possible that she wouldn't remember things for a 1316 01:05:37 --> 01:05:40 period of time after she had regained consciousness, if I 1317 01:05:40 --> 01:05:42 managed to knock her out completely. 1318 01:05:42 --> 01:05:44 There might be a period of time after she regained 1319 01:05:44 --> 01:05:49 consciousness before her memory would -- she might have been 1320 01:05:49 --> 01:05:54 talking and wandering around and say, I'm fine, I'm fine. 1321 01:05:54 --> 01:05:56 But have no recollection of that either. 1322 01:05:56 --> 01:06:00 That would be an anterograde amnesia. 1323 01:06:00 --> 01:06:03 Sorry I didn't really mean to hit you on the head. 1324 01:06:03 --> 01:06:04 You'll be fine. 1325 01:06:04 --> 01:06:07 1326 01:06:07 --> 01:06:12 So let me say a few words in the remaining time about 1327 01:06:12 --> 01:06:17 retrieval from memory. 1328 01:06:17 --> 01:06:18 Like any of these topics, there's a 1329 01:06:18 --> 01:06:19 great deal to be said. 1330 01:06:19 --> 01:06:22 But probably the most important thing to understand about 1331 01:06:22 --> 01:06:26 retrieval from memory is that retrieval particularly if any 1332 01:06:26 --> 01:06:35 rich sort of memory is not the recovery, the replay, of the 1333 01:06:35 --> 01:06:46 tape for some true guaranteed image of what you experienced 1334 01:06:46 --> 01:06:49 or what you learned or something like that. 1335 01:06:49 --> 01:06:53 This sort of image here of a scaffolding holding together 1336 01:06:53 --> 01:06:58 that memory is also an image of a notion that any memory is 1337 01:06:58 --> 01:07:02 probably stored multipley and in different bits of brain. 1338 01:07:02 --> 01:07:04 Different aspects of different bits of brain. 1339 01:07:04 --> 01:07:07 And if you want to recall it, that what you're doing 1340 01:07:07 --> 01:07:11 is reconstructing it. 1341 01:07:11 --> 01:07:15 You can probably get some intuition about this if you 1342 01:07:15 --> 01:07:21 asked yourself about famous stories about you in your 1343 01:07:21 --> 01:07:28 childhood that are famous in your family that you remember. 1344 01:07:28 --> 01:07:32 Not just the stories they tell about you from when you hit 1345 01:07:32 --> 01:07:33 with the bat and stuff like that. 1346 01:07:33 --> 01:07:36 But stories that you remember is sort of shaping incidents 1347 01:07:36 --> 01:07:37 in your childhood. 1348 01:07:37 --> 01:07:42 But that also get retold every time the family 1349 01:07:42 --> 01:07:44 gets together, right. 1350 01:07:44 --> 01:07:48 So in the Jewish tradition is a holiday called Passover. 1351 01:07:48 --> 01:07:51 The tradition is that you drink four cups of wine. 1352 01:07:51 --> 01:07:53 It's not really a brilliant tradition if you happen to 1353 01:07:53 --> 01:07:56 be about four years old. 1354 01:07:56 --> 01:08:01 And I vividly remember my sister getting into the nice 1355 01:08:01 --> 01:08:06 sweet wine that you drink and conching out at the table. 1356 01:08:06 --> 01:08:09 And I know whose house we were at. 1357 01:08:09 --> 01:08:13 This is a tale that has been retold to my sister's no doubt 1358 01:08:13 --> 01:08:18 the light every year for the intervening 40 plus years. 1359 01:08:18 --> 01:08:23 And it is absolutely unclear to me given what I subsequently 1360 01:08:23 --> 01:08:26 learned about memory, whether what I am remembering is the 1361 01:08:26 --> 01:08:29 original event at this point or reconstruction based on 1362 01:08:29 --> 01:08:31 all the family stories. 1363 01:08:31 --> 01:08:33 I think I can still remember the original thing, but there's 1364 01:08:33 --> 01:08:34 no real way of knowing. 1365 01:08:34 --> 01:08:41 What you are remembering is a reconstruction. 1366 01:08:41 --> 01:08:44 Before I talk more about that, let me load you up 1367 01:08:44 --> 01:08:47 with some more words. 1368 01:08:47 --> 01:08:50 These aren't going to be nonsense unless 1369 01:08:50 --> 01:08:51 I can't find them. 1370 01:08:51 --> 01:08:52 There we go. 1371 01:08:52 --> 01:08:54 There's some nice words. 1372 01:08:54 --> 01:08:56 Which words do I want to use. 1373 01:08:56 --> 01:08:57 Oh these are good words. 1374 01:08:57 --> 01:08:59 OK. 1375 01:08:59 --> 01:09:01 I'm going to read you a list of words, and then we'll 1376 01:09:01 --> 01:09:02 do a recognition test. 1377 01:09:02 --> 01:09:05 Don't have to write anything down this time. 1378 01:09:05 --> 01:09:07 I'm going to read you a list of words, and you're just going to 1379 01:09:07 --> 01:09:10 tell me the word was on the list or the word was 1380 01:09:10 --> 01:09:11 not on the list. 1381 01:09:11 --> 01:09:13 Ready? 1382 01:09:13 --> 01:09:16 So try to remember all of these. 1383 01:09:16 --> 01:09:19 Fury. 1384 01:09:19 --> 01:09:21 Rage. 1385 01:09:21 --> 01:09:24 Enrage. 1386 01:09:24 --> 01:09:26 Carpet. 1387 01:09:26 --> 01:09:28 Club. 1388 01:09:28 --> 01:09:31 Emotion. 1389 01:09:31 --> 01:09:33 Ire. 1390 01:09:33 --> 01:09:35 Mountain. 1391 01:09:35 --> 01:09:38 Fight. 1392 01:09:38 --> 01:09:40 Fear. 1393 01:09:40 --> 01:09:42 Mean. 1394 01:09:42 --> 01:09:45 Mad. 1395 01:09:45 --> 01:09:47 Place. 1396 01:09:47 --> 01:09:49 Hate. 1397 01:09:49 --> 01:09:51 And road. 1398 01:09:51 --> 01:09:51 OK. 1399 01:09:51 --> 01:09:54 You got all those nicely stored in memory? 1400 01:09:54 --> 01:09:55 OK. 1401 01:09:55 --> 01:09:57 So let's see, where's my test list here? 1402 01:09:57 --> 01:09:59 Looks like my test list. 1403 01:09:59 --> 01:10:01 OK. 1404 01:10:01 --> 01:10:02 Did you hear the word fight? 1405 01:10:02 --> 01:10:03 AUDIENCE: Yes. 1406 01:10:03 --> 01:10:04 PROFESSOR: Science? 1407 01:10:04 --> 01:10:05 AUDIENCE: No. 1408 01:10:05 --> 01:10:07 PROFESSOR: Rage? 1409 01:10:07 --> 01:10:08 AUDIENCE: Yes. 1410 01:10:08 --> 01:10:09 PROFESSOR: Fear? 1411 01:10:09 --> 01:10:10 AUDIENCE: Yes. 1412 01:10:10 --> 01:10:11 PROFESSOR: Emotion? 1413 01:10:11 --> 01:10:12 AUDIENCE: Yes. 1414 01:10:12 --> 01:10:14 PROFESSOR: Wrath? 1415 01:10:14 --> 01:10:15 AUDIENCE: No. 1416 01:10:15 --> 01:10:15 PROFESSOR: Club? 1417 01:10:15 --> 01:10:16 AUDIENCE: Yes. 1418 01:10:16 --> 01:10:17 PROFESSOR: Anger? 1419 01:10:17 --> 01:10:18 AUDIENCE: [UNINTELLIGIBLE] 1420 01:10:18 --> 01:10:22 PROFESSOR: Well. 1421 01:10:22 --> 01:10:24 1422 01:10:24 --> 01:10:25 Road? 1423 01:10:25 --> 01:10:26 AUDIENCE: Yes. 1424 01:10:26 --> 01:10:27 PROFESSOR: Ire? 1425 01:10:27 --> 01:10:28 AUDIENCE: Yes. 1426 01:10:28 --> 01:10:30 PROFESSOR: Hatred? 1427 01:10:30 --> 01:10:30 AUDIENCE: No. 1428 01:10:30 --> 01:10:33 PROFESSOR: OK. 1429 01:10:33 --> 01:10:34 How about enrage? 1430 01:10:34 --> 01:10:35 AUDIENCE: Yes. 1431 01:10:35 --> 01:10:36 PROFESSOR: OK. 1432 01:10:36 --> 01:10:42 so there were a couple in there where you could hear people 1433 01:10:42 --> 01:10:49 hesitating, or you could hear a diversion of opinion. 1434 01:10:49 --> 01:10:52 And those are the critical elements there. 1435 01:10:52 --> 01:10:55 The most critical element in this case is the word anger, 1436 01:10:55 --> 01:10:59 which a fair number of people asserted it was on the list. 1437 01:10:59 --> 01:11:00 It was not in fact on the list. 1438 01:11:00 --> 01:11:03 And even those people who were asserting that it was not on 1439 01:11:03 --> 01:11:08 the list, there was a measurable what's 1440 01:11:08 --> 01:11:10 going on there. 1441 01:11:10 --> 01:11:13 This is in effect known as the Deese-Roediger-McDermott 1442 01:11:13 --> 01:11:17 effect or on the handout the Deese-Roediger-McDermott demo. 1443 01:11:17 --> 01:11:21 Roediger and Mcdermott discovered it. 1444 01:11:21 --> 01:11:24 And then like all good phenomena in psychology 1445 01:11:24 --> 01:11:27 discovered that Deese had published it 10 years prior. 1446 01:11:27 --> 01:11:30 And so it now goes by all of their names. 1447 01:11:30 --> 01:11:34 In any case, what they did, they were deliberately looking 1448 01:11:34 --> 01:11:36 for this, what they did was they -- where did my 1449 01:11:36 --> 01:11:37 semantic network go? 1450 01:11:37 --> 01:11:39 Oh here's my semantic network. 1451 01:11:39 --> 01:11:44 What they did was they took a word and they asked people to 1452 01:11:44 --> 01:11:46 give the associates to it. 1453 01:11:46 --> 01:11:49 So let's take the word anger. 1454 01:11:49 --> 01:11:52 What words come to mind when you think of the word anger? 1455 01:11:52 --> 01:11:57 Well fight, rage, enrage, emotion, ire. 1456 01:11:57 --> 01:11:58 This collection of words. 1457 01:11:58 --> 01:12:04 You then use a training list of words that has the associates, 1458 01:12:04 --> 01:12:06 but not the target word. 1459 01:12:06 --> 01:12:11 And what you're in effect doing is saying, animal, 1460 01:12:11 --> 01:12:16 meow, mouse, dog, fur. 1461 01:12:16 --> 01:12:20 And each time you do that, the spreading activation gets all 1462 01:12:20 --> 01:12:24 sorts of stuff, but all of these are pointing towards cat. 1463 01:12:24 --> 01:12:27 And then later I say well was cats on the list? 1464 01:12:27 --> 01:12:30 And you go and look at that note and lo and behold it's 1465 01:12:30 --> 01:12:32 glowing softly in your mind. 1466 01:12:32 --> 01:12:36 And you say, yeah, yeah. 1467 01:12:36 --> 01:12:38 That was there. 1468 01:12:38 --> 01:12:42 And if you take one of other, it didn't particularly work in 1469 01:12:42 --> 01:12:44 this case, ire sometimes works for instance. 1470 01:12:44 --> 01:12:48 You take a word like ire people are no more sure that ire was 1471 01:12:48 --> 01:12:53 on the list than they are that anger was on the list. 1472 01:12:53 --> 01:12:57 People are thoroughly in these experiments confused about 1473 01:12:57 --> 01:13:01 whether or not the target word was on the list. 1474 01:13:01 --> 01:13:07 Because you don't have access to the truth about your memore. 1475 01:13:07 --> 01:13:08 You have the access to what you can manage to 1476 01:13:08 --> 01:13:10 reconstruct about it. 1477 01:13:10 --> 01:13:11 Now that's fine. 1478 01:13:11 --> 01:13:17 That sounds benign enough as a lab demo, but let's suppose 1479 01:13:17 --> 01:13:21 that you're not just sitting in some nice intro psych 1480 01:13:21 --> 01:13:25 class, but you're sitting in the witness box. 1481 01:13:25 --> 01:13:27 And somebody is saying, where were you on the 1482 01:13:27 --> 01:13:30 night of May 5th? 1483 01:13:30 --> 01:13:32 And expecting you to tell the truth. 1484 01:13:32 --> 01:13:34 Well of course you're going to try to tell the truth, but the 1485 01:13:34 --> 01:13:40 truth in this case is going to get shaped by the nature of 1486 01:13:40 --> 01:13:44 your particular semantic network, among other things. 1487 01:13:44 --> 01:13:46 Let's talk about that part a little bit first. 1488 01:13:46 --> 01:13:51 Famous experiment done first in the 40's replicated many times 1489 01:13:51 --> 01:13:53 since in many different ways, but let me describe 1490 01:13:53 --> 01:13:56 the original version. 1491 01:13:56 --> 01:14:00 You, and in this case you are a white male, are 1492 01:14:00 --> 01:14:05 shown a depiction of an altercation on a subway. 1493 01:14:05 --> 01:14:07 Two guys get into a fight. 1494 01:14:07 --> 01:14:10 In the 40's it was done as a hand drawing cartoon. 1495 01:14:10 --> 01:14:12 It's been replicated with, you know, snazzy video. 1496 01:14:12 --> 01:14:14 It doesn't matter. 1497 01:14:14 --> 01:14:18 Two guys get into a fight, or get into a sort of 1498 01:14:18 --> 01:14:19 shouting kind of argument. 1499 01:14:19 --> 01:14:24 At some point one guy pulls a knife, and then sometimes 1500 01:14:24 --> 01:14:27 subsequent we'll ask you about this, OK. 1501 01:14:27 --> 01:14:34 1502 01:14:34 --> 01:14:35 Well the question you're going to be asked is 1503 01:14:35 --> 01:14:37 who pulled the knife. 1504 01:14:37 --> 01:14:41 And the variable of interest is the race of the participants. 1505 01:14:41 --> 01:14:44 So you got a nice little 2 by 2 here. 1506 01:14:44 --> 01:14:48 They can both be caucasian, they can both be African 1507 01:14:48 --> 01:14:51 American, or it can be crossed. 1508 01:14:51 --> 01:14:55 The interesting and disturbing finding is the number of times 1509 01:14:55 --> 01:15:01 when the knife is pulled by the white guy, it ends up in memory 1510 01:15:01 --> 01:15:04 in the hand of the black guy. 1511 01:15:04 --> 01:15:07 Now as I say, this has been replicated all 1512 01:15:07 --> 01:15:08 sorts of different ways. 1513 01:15:08 --> 01:15:10 This is not a, you know, white guys are all 1514 01:15:10 --> 01:15:13 biggits experiment. 1515 01:15:13 --> 01:15:17 This is we all have biases, and the interesting and disturbing 1516 01:15:17 --> 01:15:22 thing is that they can actually interfere with what we think 1517 01:15:22 --> 01:15:25 of as, you know, our nice clear memory. 1518 01:15:25 --> 01:15:27 These people weren't sitting there saying, you know, oh yeah 1519 01:15:27 --> 01:15:32 I'm being paid, you know, 1940 $0.10 an hour to be in 1520 01:15:32 --> 01:15:33 this psych experiment. 1521 01:15:33 --> 01:15:37 Why don't I see if I can stick it to an imaginary black guy. 1522 01:15:37 --> 01:15:39 No. 1523 01:15:39 --> 01:15:44 They were presumably being as honest as they could be. 1524 01:15:44 --> 01:15:48 But the structure of their memory. 1525 01:15:48 --> 01:15:52 Which included things like who do I think might pull a knife, 1526 01:15:52 --> 01:15:56 influenced what they recalled, what they pulled out of memory. 1527 01:15:56 --> 01:15:58 And we can impose this structure from the 1528 01:15:58 --> 01:16:00 outside as well. 1529 01:16:00 --> 01:16:03 So suppose we do the following experiment. 1530 01:16:03 --> 01:16:06 This is an experiment also done many different 1531 01:16:06 --> 01:16:10 ways at this point. 1532 01:16:10 --> 01:16:13 Elizabeth Loftus is the name associated with 1533 01:16:13 --> 01:16:16 this line of work. 1534 01:16:16 --> 01:16:18 You're going to see another film. 1535 01:16:18 --> 01:16:20 This time you're going to watch a car crash. 1536 01:16:20 --> 01:16:25 The red car is going to get into an accident 1537 01:16:25 --> 01:16:27 with the blue car. 1538 01:16:27 --> 01:16:29 All right. 1539 01:16:29 --> 01:16:29 OK. 1540 01:16:29 --> 01:16:30 You see this. 1541 01:16:30 --> 01:16:31 You're going to be asked about it. 1542 01:16:31 --> 01:16:33 We'll show all of you guys the same film. 1543 01:16:33 --> 01:16:37 And now we'll ask the question three different ways. 1544 01:16:37 --> 01:16:46 At what speed did the red car bump into the blue car? 1545 01:16:46 --> 01:16:51 At what speed did the red car hit the blue car? 1546 01:16:51 --> 01:16:55 At what speed did the red car smashed into the blue car? 1547 01:16:55 --> 01:16:57 That's the only manipulation here. 1548 01:16:57 --> 01:17:01 You all saw the same video. 1549 01:17:01 --> 01:17:04 So what do you figure the difference is? 1550 01:17:04 --> 01:17:07 What's the result look like? 1551 01:17:07 --> 01:17:12 So who gives the higher speed responses? 1552 01:17:12 --> 01:17:12 AUDIENCE: The smashed guy. 1553 01:17:12 --> 01:17:14 PROFESSOR: The smashed guys Lawyers know this. 1554 01:17:14 --> 01:17:20 This is why courts attempts to avoid leading question. 1555 01:17:20 --> 01:17:22 But it's very hard to avoid. 1556 01:17:22 --> 01:17:27 But, you know, that's a pretty subtle kind of lead. 1557 01:17:27 --> 01:17:28 But it's not a subtle kind of effect. 1558 01:17:28 --> 01:17:32 It's about as I recall a 15 MPH kind of effect, which is a sort 1559 01:17:32 --> 01:17:37 of thing that has an impact, you should pardon the 1560 01:17:37 --> 01:17:40 expression, on what a jury is going to think about whether or 1561 01:17:40 --> 01:17:44 not you, the driver of that red car, was at fault or not. 1562 01:17:44 --> 01:17:49 If you were screaming through the intersection at 45 MPH 1563 01:17:49 --> 01:17:52 that's worse than if you're screaming through the 1564 01:17:52 --> 01:17:55 intersection at 30 MPH or something like that. 1565 01:17:55 --> 01:18:02 So your memory can be influenced by the way that you 1566 01:18:02 --> 01:18:04 are asked about that memory. 1567 01:18:04 --> 01:18:08 This shows up not just on the stand. 1568 01:18:08 --> 01:18:12 Well actually a version of this shows up beautifully in the 1569 01:18:12 --> 01:18:18 experience of most people at some point who come to MIT. 1570 01:18:18 --> 01:18:23 Most people who get to MIT are smart people, and they would 1571 01:18:23 --> 01:18:25 do well to remember that. 1572 01:18:25 --> 01:18:28 Because most people sometime, like within the first few weeks 1573 01:18:28 --> 01:18:32 of being at MIT, receive a data point of some sort that 1574 01:18:32 --> 01:18:35 suggests that this might not be true. 1575 01:18:35 --> 01:18:36 Right? 1576 01:18:36 --> 01:18:38 Like the first calculus test or something like that. 1577 01:18:38 --> 01:18:42 You've never gotten a score below 99.8 on a math 1578 01:18:42 --> 01:18:43 test in your life. 1579 01:18:43 --> 01:18:46 You get back the first test in 18 or whatever it was, 1580 01:18:46 --> 01:18:51 and you know, 10 cool. 1581 01:18:51 --> 01:18:53 Oh man it's not 10 out of 10 is it? 1582 01:18:53 --> 01:18:54 AUDIENCE: [LAUGHTER] 1583 01:18:54 --> 01:18:55 PROFESSOR: You know this is really bad. 1584 01:18:55 --> 01:18:58 And so that's a depressing data point. 1585 01:18:58 --> 01:19:02 But a surprising number of people come to the conclusion 1586 01:19:02 --> 01:19:05 from that, that they not only are stupid but 1587 01:19:05 --> 01:19:07 always were stupid. 1588 01:19:07 --> 01:19:15 And probably unlovely and generally, you know, bad, 1589 01:19:15 --> 01:19:16 horrible individuals. 1590 01:19:16 --> 01:19:21 And you should remember this is one of these 1591 01:19:21 --> 01:19:25 sort of context effects. 1592 01:19:25 --> 01:19:26 That may be a little over stating it. 1593 01:19:26 --> 01:19:29 But look you know -- and this has actually been tested with 1594 01:19:29 --> 01:19:32 clinical populations of depressive people who go from 1595 01:19:32 --> 01:19:34 being depressed to being undepressed. 1596 01:19:34 --> 01:19:38 Suppose you ask somebody who's depressed clinically or 1597 01:19:38 --> 01:19:40 otherwise, you wake up in the morning it's pouring 1598 01:19:40 --> 01:19:42 cold rain out there. 1599 01:19:42 --> 01:19:45 The problems that did you get done because you fell asleep 1600 01:19:45 --> 01:19:47 in there, and there are now drool marks on it. 1601 01:19:47 --> 01:19:50 And not only is there drook mark, there's a little heart 1602 01:19:50 --> 01:19:53 shaped note next to it saying, I'm leaving 1603 01:19:53 --> 01:19:54 you for your roommate. 1604 01:19:54 --> 01:19:55 AUDIENCE: [LAUGHTER] 1605 01:19:55 --> 01:19:57 PROFESSOR: Right? 1606 01:19:57 --> 01:20:05 So if I asked you at that point how was your childhood? 1607 01:20:05 --> 01:20:09 Not how do you feel, but what kind of childhood did you have? 1608 01:20:09 --> 01:20:12 You had not maybe a miserable childhood. 1609 01:20:12 --> 01:20:15 But to exaggerate, you know, you grew up in a closet. 1610 01:20:15 --> 01:20:18 1611 01:20:18 --> 01:20:21 Later on if you're feeling better I ask you about 1612 01:20:21 --> 01:20:22 the same childhood. 1613 01:20:22 --> 01:20:25 The childhood has magically improved. 1614 01:20:25 --> 01:20:26 All right. 1615 01:20:26 --> 01:20:30 We'll see you on Thursday if I remember correctly. 1616 01:20:30 --> 01:20:31