1 00:00:00,060 --> 00:00:02,400 The following content is provided under a Creative 2 00:00:02,400 --> 00:00:03,790 Commons license. 3 00:00:03,790 --> 00:00:06,030 Your support will help MIT OpenCourseWare 4 00:00:06,030 --> 00:00:10,120 continue to offer high-quality educational resources for free. 5 00:00:10,120 --> 00:00:12,660 To make a donation or to view additional materials 6 00:00:12,660 --> 00:00:16,620 from hundreds of MIT courses, visit MIT OpenCourseWare 7 00:00:16,620 --> 00:00:17,942 at ocw.mit.edu. 8 00:00:20,560 --> 00:00:22,830 PROFESSOR: We've been talking for two days about-- 9 00:00:22,830 --> 00:00:23,770 AUDIENCE: [INAUDIBLE] 10 00:00:23,770 --> 00:00:25,170 PROFESSOR: --attention. 11 00:00:25,170 --> 00:00:27,162 [LAUGHING] 12 00:00:27,162 --> 00:00:28,656 AUDIENCE: Sorry. 13 00:00:28,656 --> 00:00:29,652 PROFESSOR: Jen. 14 00:00:29,652 --> 00:00:30,944 AUDIENCE: [? I don't see it. ?] 15 00:00:30,944 --> 00:00:33,110 PROFESSOR: I would like you to direct your attention 16 00:00:33,110 --> 00:00:34,196 to me and not the ladybug. 17 00:00:34,196 --> 00:00:35,362 AUDIENCE: There's a ladybug? 18 00:00:35,362 --> 00:00:35,862 Where? 19 00:00:38,396 --> 00:00:40,952 AUDIENCE: [INAUDIBLE] pay attention. 20 00:00:40,952 --> 00:00:41,660 PROFESSOR: Right. 21 00:00:41,660 --> 00:00:43,784 So we've been talking for two days about attention. 22 00:00:43,784 --> 00:00:47,480 So attention is this idea that you focus your processing 23 00:00:47,480 --> 00:00:50,390 resources on a subset of all the input that's 24 00:00:50,390 --> 00:00:51,830 coming in at one time. 25 00:00:51,830 --> 00:00:55,220 You can't deal with all the sensory information out there. 26 00:00:55,220 --> 00:00:57,680 You've got to select some portion of it 27 00:00:57,680 --> 00:00:59,497 to direct further. 28 00:00:59,497 --> 00:01:01,080 And we talked yesterday about the idea 29 00:01:01,080 --> 00:01:05,360 that this is a competition between inputs coming in, 30 00:01:05,360 --> 00:01:08,720 that all of the different possible things 31 00:01:08,720 --> 00:01:10,860 you could be attending to compete. 32 00:01:10,860 --> 00:01:13,640 And whichever one gets deemed the most salient, 33 00:01:13,640 --> 00:01:16,160 the most important or relevant to what you're doing 34 00:01:16,160 --> 00:01:20,870 manages to lay claim to the processing level that it needs. 35 00:01:20,870 --> 00:01:25,460 So we're going to shift into a very related topic and a topic 36 00:01:25,460 --> 00:01:27,800 that seems to get more and more related 37 00:01:27,800 --> 00:01:30,020 as we look at modern models. 38 00:01:30,020 --> 00:01:31,770 And we're going to look at working memory. 39 00:01:31,770 --> 00:01:34,680 Working memory is another term for short-term memory. 40 00:01:34,680 --> 00:01:39,770 So what is this working memory business? 41 00:01:39,770 --> 00:01:42,890 So this is this brief, short-term, immediate memory 42 00:01:42,890 --> 00:01:45,550 for data you're currently working with. 43 00:01:45,550 --> 00:01:47,300 Looking up a phone number, hanging onto it 44 00:01:47,300 --> 00:01:52,460 for the time it takes you to punch it into the phone, 45 00:01:52,460 --> 00:01:55,400 or doing mental arithmetic requires 46 00:01:55,400 --> 00:01:57,920 that you hang on to the numbers involved. 47 00:01:57,920 --> 00:01:59,447 Or the store I used to work at, we 48 00:01:59,447 --> 00:02:02,030 had a credit card machine that was separate from the register. 49 00:02:02,030 --> 00:02:03,710 So if someone wanted to pay with a credit card, 50 00:02:03,710 --> 00:02:05,030 you took their card, and you swiped it, 51 00:02:05,030 --> 00:02:06,196 and you typed in the amount. 52 00:02:06,196 --> 00:02:09,110 You'd have to hang on to the total off the register 53 00:02:09,110 --> 00:02:12,200 long enough to punch it into the machine. 54 00:02:12,200 --> 00:02:15,470 All of these are examples of tasks 55 00:02:15,470 --> 00:02:17,780 that use working memory-- short-term memory. 56 00:02:20,510 --> 00:02:23,960 Short-term memory is closely related to attention. 57 00:02:23,960 --> 00:02:25,460 Attending to something and having it 58 00:02:25,460 --> 00:02:28,160 in your working memory are often-- 59 00:02:28,160 --> 00:02:31,115 some people would say always, some people would not-- 60 00:02:31,115 --> 00:02:31,865 equivalent states. 61 00:02:35,958 --> 00:02:37,386 AUDIENCE: What would make someone 62 00:02:37,386 --> 00:02:42,160 have a bad memory, as opposed to a good memory for that-- 63 00:02:42,160 --> 00:02:45,130 PROFESSOR: For short-- so when people talk about memory, 64 00:02:45,130 --> 00:02:47,820 memory researchers say there's two really different things. 65 00:02:47,820 --> 00:02:48,970 There's short-term memory, which is 66 00:02:48,970 --> 00:02:50,845 this immediate hanging on to the stuff you're 67 00:02:50,845 --> 00:02:52,280 working with right now. 68 00:02:52,280 --> 00:02:54,250 And then there's long-term memory, right? 69 00:02:54,250 --> 00:02:57,280 Like remembering what you did last week, or last year, 70 00:02:57,280 --> 00:03:00,130 or something that you learned in third grade. 71 00:03:00,130 --> 00:03:02,500 Long-term memory is really different. 72 00:03:02,500 --> 00:03:04,840 And people's capacities for both of these things-- 73 00:03:04,840 --> 00:03:08,180 short term and long term-- vary among individuals. 74 00:03:08,180 --> 00:03:13,640 But exactly why is not terribly well understood. 75 00:03:13,640 --> 00:03:16,240 We'll talk about this some more when we talk about longer term 76 00:03:16,240 --> 00:03:17,200 memory next week. 77 00:03:20,370 --> 00:03:22,870 You guys might remember that the very first paper we 78 00:03:22,870 --> 00:03:26,036 read the first week of class talked about training people 79 00:03:26,036 --> 00:03:27,660 on this particular working memory task, 80 00:03:27,660 --> 00:03:30,870 and that it was closely related to fluid intelligence, that 81 00:03:30,870 --> 00:03:33,562 improving your working memory-- 82 00:03:33,562 --> 00:03:35,770 or improving on this particular working memory task-- 83 00:03:35,770 --> 00:03:38,530 seemed to improve people's fluid intelligence 84 00:03:38,530 --> 00:03:40,610 on a number of different tests. 85 00:03:40,610 --> 00:03:42,700 So working memory is good stuff. 86 00:03:42,700 --> 00:03:44,200 It correlates with SAT scores. 87 00:03:44,200 --> 00:03:47,125 It correlates with scores on intelligence tests. 88 00:03:47,125 --> 00:03:51,790 It correlates reasonably well with success in school, 89 00:03:51,790 --> 00:03:55,220 about as well as SAT scores do. 90 00:03:55,220 --> 00:04:01,150 OK, so Atkinson and Shiffrin in 1968, 91 00:04:01,150 --> 00:04:03,490 back in the proverbial day, came up 92 00:04:03,490 --> 00:04:06,610 with a model for how memory works that a lot of people 93 00:04:06,610 --> 00:04:08,890 call the modal model. 94 00:04:08,890 --> 00:04:12,277 It's the statistical use of the term modal. 95 00:04:12,277 --> 00:04:14,110 It's the one that gets cited the most often. 96 00:04:14,110 --> 00:04:16,209 It's not as common now as it was. 97 00:04:16,209 --> 00:04:18,170 But it's still worth thinking about. 98 00:04:18,170 --> 00:04:20,709 So their model says that, OK, we've 99 00:04:20,709 --> 00:04:25,510 got all of this external input coming in. 100 00:04:25,510 --> 00:04:31,649 And it goes into an immediate short-term sensory storage. 101 00:04:31,649 --> 00:04:33,190 You can think of this as being, like, 102 00:04:33,190 --> 00:04:36,400 the primary sensory portions of your brain-- 103 00:04:36,400 --> 00:04:38,660 like primary visual cortex that we talked about, 104 00:04:38,660 --> 00:04:42,430 or auditory cortex, or tactile cortex-- 105 00:04:42,430 --> 00:04:43,950 all of this. 106 00:04:43,950 --> 00:04:45,040 And that's very brief. 107 00:04:45,040 --> 00:04:48,659 And this is probably not within conscious awareness. 108 00:04:48,659 --> 00:04:49,700 And some of it gets lost. 109 00:04:53,050 --> 00:04:58,520 But some of it goes into short-term memory. 110 00:05:03,186 --> 00:05:05,060 And there again, still working with this idea 111 00:05:05,060 --> 00:05:07,185 that short-term memory is information that's, like, 112 00:05:07,185 --> 00:05:08,230 immediately accessible. 113 00:05:08,230 --> 00:05:11,460 You can work on it there. 114 00:05:11,460 --> 00:05:12,960 Atkinson and Shiffrin were really 115 00:05:12,960 --> 00:05:14,730 trying to understand how learning works-- 116 00:05:14,730 --> 00:05:17,760 how this longer-term, get a piece of information, 117 00:05:17,760 --> 00:05:19,540 and be able to recall it forever. 118 00:05:19,540 --> 00:05:21,300 What's the capital of Massachusetts? 119 00:05:21,300 --> 00:05:22,008 AUDIENCE: Boston. 120 00:05:22,008 --> 00:05:22,770 PROFESSOR: Boston. 121 00:05:22,770 --> 00:05:25,750 When did you learn this? 122 00:05:25,750 --> 00:05:27,660 Like yea big? 123 00:05:27,660 --> 00:05:28,464 Back in the day. 124 00:05:28,464 --> 00:05:30,880 So this is something that's just in your long-term memory. 125 00:05:30,880 --> 00:05:31,140 You can get it. 126 00:05:31,140 --> 00:05:32,931 And they were interested in how information 127 00:05:32,931 --> 00:05:34,950 gets to long-term memory. 128 00:05:39,820 --> 00:05:41,920 And they said that this short-term processing 129 00:05:41,920 --> 00:05:47,320 store processes stuff that's going into long-term memory. 130 00:05:47,320 --> 00:05:49,780 You've got to put things into short-term memory first. 131 00:05:49,780 --> 00:05:52,060 There's this sequential model. 132 00:05:52,060 --> 00:05:56,110 Stuff can get lost out of memory at any of these stages, 133 00:05:56,110 --> 00:05:58,840 although once something's really consolidated 134 00:05:58,840 --> 00:06:02,200 into long-term memory, you're not very likely to lose it. 135 00:06:06,160 --> 00:06:09,070 And they said, short-term memory can also pull information back 136 00:06:09,070 --> 00:06:11,530 from long-term memory to work with it 137 00:06:11,530 --> 00:06:12,730 at any given point in time. 138 00:06:12,730 --> 00:06:16,360 So there's connections both ways between those two storage 139 00:06:16,360 --> 00:06:18,940 spaces. 140 00:06:18,940 --> 00:06:21,310 Atkinson and Shiffrin-- right. 141 00:06:21,310 --> 00:06:23,380 So sensory, short term, and long term-- 142 00:06:23,380 --> 00:06:26,190 you can lose stuff at every stage of it. 143 00:06:26,190 --> 00:06:29,360 Once something really makes it into long-term memory, 144 00:06:29,360 --> 00:06:32,800 whether or not it's ever really lost 145 00:06:32,800 --> 00:06:34,570 is a point of theoretical debate. 146 00:06:37,506 --> 00:06:39,380 People will say, but of course stuff is lost. 147 00:06:39,380 --> 00:06:46,092 I can't remember what I did on June 2, 1994-- 148 00:06:46,092 --> 00:06:48,670 [INAUDIBLE] 1998. 149 00:06:48,670 --> 00:06:52,100 So I think you guys were, like, what, 2 in 1994-- 150 00:06:52,100 --> 00:06:53,385 ish? 151 00:06:53,385 --> 00:06:54,760 So young enough that you probably 152 00:06:54,760 --> 00:06:56,260 don't remember anything. 153 00:06:56,260 --> 00:06:59,390 But from, like, childhood-- random day. 154 00:06:59,390 --> 00:07:03,437 But where exactly in the memory process that drop happens 155 00:07:03,437 --> 00:07:05,270 is another thing we'll talk about next week. 156 00:07:05,270 --> 00:07:07,090 So Atkinson and Shiffrin's model says 157 00:07:07,090 --> 00:07:10,000 that short-term memory is this key step in storage 158 00:07:10,000 --> 00:07:11,540 of long-term memories. 159 00:07:11,540 --> 00:07:15,100 And it can also pull information from long-term memory 160 00:07:15,100 --> 00:07:18,370 and use it. 161 00:07:18,370 --> 00:07:19,930 They thought of short-term memory 162 00:07:19,930 --> 00:07:24,250 as being more or less all one unit. 163 00:07:24,250 --> 00:07:26,590 Modern theorists would say there's different kinds 164 00:07:26,590 --> 00:07:28,000 of short-term memory. 165 00:07:28,000 --> 00:07:29,950 Working memory is the more modern term. 166 00:07:35,780 --> 00:07:39,320 This model of how memory works was 167 00:07:39,320 --> 00:07:41,930 one of the very early pieces of theory 168 00:07:41,930 --> 00:07:44,690 in this shift that happened in the middle of the 20th century 169 00:07:44,690 --> 00:07:47,090 from behaviorist perspectives in psychology 170 00:07:47,090 --> 00:07:48,266 to cognitive perspectives. 171 00:07:48,266 --> 00:07:50,390 These guys are really talking about mental events-- 172 00:07:50,390 --> 00:07:52,780 what's happening inside your head-- 173 00:07:52,780 --> 00:07:55,550 in a way that psychologists 10 or 20 years earlier 174 00:07:55,550 --> 00:07:59,240 wouldn't really have wanted to do, which makes it cool. 175 00:07:59,240 --> 00:08:01,730 OK, while we're on working memory, 176 00:08:01,730 --> 00:08:05,210 you guys have also probably encountered this idea before, 177 00:08:05,210 --> 00:08:10,400 that your short-term memory can hold seven, plus or minus two, 178 00:08:10,400 --> 00:08:12,346 pieces of information. 179 00:08:12,346 --> 00:08:16,070 The guy who came up with this is George Miller. 180 00:08:16,070 --> 00:08:20,660 The paper that he wrote in '56 says something like, 181 00:08:20,660 --> 00:08:23,930 I have been being pursued by an integer. 182 00:08:23,930 --> 00:08:25,535 It's the opening sentence. 183 00:08:25,535 --> 00:08:27,410 It keeps slipping into my data. 184 00:08:27,410 --> 00:08:28,530 It keeps showing up. 185 00:08:34,390 --> 00:08:38,530 This is a little bit before the Atkinson-Shiffrin model here. 186 00:08:38,530 --> 00:08:40,240 But Miller was one of the first guys 187 00:08:40,240 --> 00:08:43,539 to really pin down not only that short-term memory has 188 00:08:43,539 --> 00:08:48,280 a limited capacity but actually pin down what that capacity is. 189 00:08:53,329 --> 00:08:56,340 He said people can remember seven, plus or minus two, 190 00:08:56,340 --> 00:08:58,350 chunks of information. 191 00:08:58,350 --> 00:09:00,210 So what's a chunk? 192 00:09:04,660 --> 00:09:06,549 If you are given a string of information, 193 00:09:06,549 --> 00:09:08,090 you're probably going to try and make 194 00:09:08,090 --> 00:09:12,440 some kind of organizational sense out of it. 195 00:09:12,440 --> 00:09:17,060 So a chunk is a well-earned cognitive unit made up 196 00:09:17,060 --> 00:09:19,730 of a number of components representing 197 00:09:19,730 --> 00:09:23,156 a frequently occurring and consistent perceptual pattern. 198 00:09:23,156 --> 00:09:24,530 So if, for example, you're trying 199 00:09:24,530 --> 00:09:27,440 to learn your friend's phone number, 200 00:09:27,440 --> 00:09:31,340 and the area code is 617, or 781, 201 00:09:31,340 --> 00:09:35,540 or one of those common Boston and Boston suburbs area codes, 202 00:09:35,540 --> 00:09:39,620 you probably don't need to remember that as 6, 1, 7. 203 00:09:39,620 --> 00:09:44,409 You probably remember the whole 617 part as a unit. 204 00:09:44,409 --> 00:09:46,200 If you grow up in a small town, like I did, 205 00:09:46,200 --> 00:09:49,460 where everyone's phone number starts with a 603-465, where 206 00:09:49,460 --> 00:09:50,930 both the area code and the exchange 207 00:09:50,930 --> 00:09:54,200 are all the same, then again, those are individual chunks. 208 00:09:54,200 --> 00:09:59,020 You don't have to learn each digit individually. 209 00:09:59,020 --> 00:10:01,330 If I ask you to learn a list of words 210 00:10:01,330 --> 00:10:02,747 and then tell me about the letters 211 00:10:02,747 --> 00:10:05,330 and then write them back down, you're probably not remembering 212 00:10:05,330 --> 00:10:06,190 individual letters. 213 00:10:06,190 --> 00:10:07,960 You're remembering the words. 214 00:10:07,960 --> 00:10:11,050 All of these are examples of our brains taking information, 215 00:10:11,050 --> 00:10:14,400 grouping it up to make it into fewer pieces to work with. 216 00:10:20,820 --> 00:10:22,820 Chunking is something that your brain just does. 217 00:10:22,820 --> 00:10:24,486 You don't really have to think about it, 218 00:10:24,486 --> 00:10:26,570 although finding ways to chunk information 219 00:10:26,570 --> 00:10:31,490 helps to remember larger things. 220 00:10:31,490 --> 00:10:35,000 Anyone learn, like, a goofy mnemonic for something ever, 221 00:10:35,000 --> 00:10:37,490 like for, oh, kingdom, phylum-- kings play 222 00:10:37,490 --> 00:10:41,940 chess on fine green sand for the kingdom, phylum, 223 00:10:41,940 --> 00:10:45,290 class, order string of taxonomies or something 224 00:10:45,290 --> 00:10:45,934 like that? 225 00:10:45,934 --> 00:10:48,350 So that's almost an example of taking advantage of the way 226 00:10:48,350 --> 00:10:51,509 that your brain can grab a meaningful unit, 227 00:10:51,509 --> 00:10:53,300 like a sentence, much more easily than just 228 00:10:53,300 --> 00:10:55,010 a random string of words. 229 00:10:55,010 --> 00:10:56,360 AUDIENCE: [INAUDIBLE] 230 00:10:56,360 --> 00:10:58,100 PROFESSOR: Back onto chunking. 231 00:10:58,100 --> 00:11:00,340 So I had a train of thought. 232 00:11:00,340 --> 00:11:01,880 I don't know what it was. 233 00:11:01,880 --> 00:11:02,580 Drat. 234 00:11:02,580 --> 00:11:03,610 I know what I was going to say. 235 00:11:03,610 --> 00:11:05,300 So you know how, like, when you have a long number 236 00:11:05,300 --> 00:11:06,620 that you have to remember for some reason, 237 00:11:06,620 --> 00:11:08,390 like your Social Security number-- they break it up 238 00:11:08,390 --> 00:11:10,040 with the little dashes in the middle? 239 00:11:10,040 --> 00:11:11,873 Part of that is to, instead of just it being 240 00:11:11,873 --> 00:11:14,760 a random string of digits, to give you a piece of chunking, 241 00:11:14,760 --> 00:11:17,330 so you can think about it as three separate parts. 242 00:11:17,330 --> 00:11:18,860 And it's actually easier for people 243 00:11:18,860 --> 00:11:21,175 to remember information when it's broken up this way. 244 00:11:21,175 --> 00:11:22,550 Same thing with the little spaces 245 00:11:22,550 --> 00:11:23,841 on, like, a credit card number. 246 00:11:23,841 --> 00:11:25,340 AUDIENCE: [INAUDIBLE] 1945. 247 00:11:25,340 --> 00:11:26,120 PROFESSOR: End of World War II. 248 00:11:26,120 --> 00:11:26,703 AUDIENCE: Yay. 249 00:11:26,703 --> 00:11:29,181 [INAUDIBLE] 250 00:11:29,181 --> 00:11:30,056 PROFESSOR: All right. 251 00:11:30,056 --> 00:11:30,548 AUDIENCE: [? Bet ?] [? that ?] [? was ?] [? going ?] [? to ?] 252 00:11:30,548 --> 00:11:32,970 bother you. 253 00:11:32,970 --> 00:11:34,220 PROFESSOR: Moving right along. 254 00:11:34,220 --> 00:11:37,820 So when you chunk information, you can hold more of it 255 00:11:37,820 --> 00:11:42,040 in working memory, as we just saw. 256 00:11:42,040 --> 00:11:44,600 Did anyone do worse on the second round of that, 257 00:11:44,600 --> 00:11:46,002 with the numbers clumped up? 258 00:11:46,002 --> 00:11:47,150 AUDIENCE: I did the same. 259 00:11:47,150 --> 00:11:49,327 PROFESSOR: You did the same. 260 00:11:49,327 --> 00:11:51,660 AUDIENCE: I think it's because I wasn't paying attention 261 00:11:51,660 --> 00:11:57,010 to the lecture [INAUDIBLE] 262 00:11:57,010 --> 00:11:59,440 PROFESSOR: Possibly. 263 00:11:59,440 --> 00:12:01,170 So Miller said seven plus or minus 264 00:12:01,170 --> 00:12:04,800 one is working memory's capacity. 265 00:12:04,800 --> 00:12:06,510 There's some more modern numbers showing 266 00:12:06,510 --> 00:12:08,910 that it's closer to, like, three plus or minus one. 267 00:12:08,910 --> 00:12:11,610 And Miller's numbers didn't take into account things 268 00:12:11,610 --> 00:12:14,532 like chunking and rehearsal. 269 00:12:14,532 --> 00:12:16,865 And three plus or minus one-- well, that was four years. 270 00:12:16,865 --> 00:12:18,406 So that's probably pushing the limits 271 00:12:18,406 --> 00:12:21,960 of a lot of people's unrehearsed working memory capabilities. 272 00:12:21,960 --> 00:12:24,606 OK, so things that affect how much 273 00:12:24,606 --> 00:12:26,730 stuff you can stick into your working memory at one 274 00:12:26,730 --> 00:12:28,380 point in time-- 275 00:12:28,380 --> 00:12:30,180 this is pretty variable among people. 276 00:12:30,180 --> 00:12:33,270 It's definitely got a genetic component. 277 00:12:33,270 --> 00:12:35,730 But even for an individual, you can 278 00:12:35,730 --> 00:12:42,960 change how much stuff you can store by chunking, by how 279 00:12:42,960 --> 00:12:45,750 fast these things can actually be spoken out loud, 280 00:12:45,750 --> 00:12:49,560 which is an unintuitive result. 281 00:12:49,560 --> 00:12:52,440 But people tend to recall about one and a half seconds' 282 00:12:52,440 --> 00:12:55,000 worth of material. 283 00:12:55,000 --> 00:12:57,930 So if I show you a list of simple nouns or, let's take, 284 00:12:57,930 --> 00:12:58,920 barn animals-- 285 00:12:58,920 --> 00:13:02,880 a cat, a dog, a cow, a chicken-- 286 00:13:02,880 --> 00:13:05,759 and have you remember a list of four or five of these, 287 00:13:05,759 --> 00:13:07,800 you'll probably be able to remember about as many 288 00:13:07,800 --> 00:13:10,410 of these words as you could pronounce 289 00:13:10,410 --> 00:13:12,510 in about a second and a half. 290 00:13:12,510 --> 00:13:15,809 And much more than that, you just can't hang on to. 291 00:13:15,809 --> 00:13:17,850 And this has been pretty well documented, in part 292 00:13:17,850 --> 00:13:20,110 because it's such a weird result. 293 00:13:20,110 --> 00:13:22,530 People have tested it for color names, for shape names, 294 00:13:22,530 --> 00:13:24,810 for a whole bunch of different kinds of nouns, 295 00:13:24,810 --> 00:13:28,620 even just for nonsense words or nonsense syllables. 296 00:13:28,620 --> 00:13:30,930 Someone did a really interesting study, 297 00:13:30,930 --> 00:13:34,620 looking at how many numbers people could recall compared 298 00:13:34,620 --> 00:13:37,470 across English speakers-- 299 00:13:37,470 --> 00:13:41,280 most English digits have one syllable-- 300 00:13:41,280 --> 00:13:45,210 Spanish and Hebrew speakers, both of which 301 00:13:45,210 --> 00:13:49,080 tend to even out to around two syllables per number, close to; 302 00:13:49,080 --> 00:13:50,250 and Arabic speakers. 303 00:13:50,250 --> 00:13:53,670 And Arabic actually averages to a little 304 00:13:53,670 --> 00:13:55,980 over two syllables per number. 305 00:13:55,980 --> 00:13:58,770 And just checked digit recollection 306 00:13:58,770 --> 00:14:01,140 across different people whose native language 307 00:14:01,140 --> 00:14:02,670 was these different things. 308 00:14:02,670 --> 00:14:05,010 And they found that people had the longest recollection 309 00:14:05,010 --> 00:14:06,870 for numbers in English out of these four 310 00:14:06,870 --> 00:14:08,280 and the shortest for Arabic. 311 00:14:08,280 --> 00:14:11,090 And Spanish and Hebrew were somewhere in the middle. 312 00:14:11,090 --> 00:14:15,230 So even though it's the exact same semantic content-- 313 00:14:15,230 --> 00:14:19,010 I'm still asking you to think about a 7 or a 3 or something-- 314 00:14:19,010 --> 00:14:23,700 and I think they were all written in numerals, 315 00:14:23,700 --> 00:14:25,800 people's performance varied just by what language 316 00:14:25,800 --> 00:14:27,216 they were speaking, by how long it 317 00:14:27,216 --> 00:14:33,714 took to pronounce that, which is a nifty result. 318 00:14:33,714 --> 00:14:35,598 AUDIENCE: [INAUDIBLE] 319 00:14:36,182 --> 00:14:37,390 PROFESSOR: Shorter or longer? 320 00:14:37,390 --> 00:14:38,306 AUDIENCE: [INAUDIBLE]. 321 00:14:38,306 --> 00:14:40,607 Well, Japanese numbers are pronounced shorter 322 00:14:40,607 --> 00:14:42,340 than [INAUDIBLE] numbers. 323 00:14:42,340 --> 00:14:43,825 [INAUDIBLE] 324 00:14:43,825 --> 00:14:45,810 PROFESSOR: [COUNTING IN JAPANESE] 325 00:14:45,810 --> 00:14:49,830 But it holds even, like, within single syllable. 326 00:14:49,830 --> 00:14:52,980 Syllables with short vowels, you can pack more in than syllables 327 00:14:52,980 --> 00:14:53,737 with long vowels. 328 00:14:53,737 --> 00:14:55,570 And you'll see that kind of difference, too. 329 00:14:55,570 --> 00:14:57,950 AUDIENCE: There aren't many, like, long vowels [INAUDIBLE].. 330 00:14:57,950 --> 00:14:58,616 PROFESSOR: Yeah? 331 00:14:58,616 --> 00:14:59,362 Cool. 332 00:14:59,362 --> 00:15:01,570 AUDIENCE: Well, what if someone was, like, bilingual? 333 00:15:01,570 --> 00:15:03,290 Would it be based on the language 334 00:15:03,290 --> 00:15:06,130 they're more comfortable with? 335 00:15:06,130 --> 00:15:08,590 PROFESSOR: Yeah, or whichever language they're using one 336 00:15:08,590 --> 00:15:11,810 they encode the data in the first place. 337 00:15:11,810 --> 00:15:12,932 I don't know. 338 00:15:12,932 --> 00:15:15,360 I'm not bilingual, not even close. 339 00:15:15,360 --> 00:15:18,416 I don't have a good perspective on this. 340 00:15:18,416 --> 00:15:20,344 AUDIENCE: There's people-- like, I know 341 00:15:20,344 --> 00:15:21,790 there's people [INAUDIBLE]. 342 00:15:28,375 --> 00:15:31,000 PROFESSOR: Everyone here, like, had to memorize your timetables 343 00:15:31,000 --> 00:15:32,166 in about third grade, right? 344 00:15:32,166 --> 00:15:36,770 You know 2 times 2 is 4, and 2 times 3 is 6, and 2 times 4 345 00:15:36,770 --> 00:15:38,849 is 8, right? 346 00:15:38,849 --> 00:15:41,390 Anyone here still run through that little thing in their head 347 00:15:41,390 --> 00:15:44,510 when they're doing out a multiplication problem? 348 00:15:44,510 --> 00:15:45,390 6 times 4. 349 00:15:45,390 --> 00:15:46,540 And 6 times 4 is 24. 350 00:15:46,540 --> 00:15:47,510 OK, so that's 4. 351 00:15:47,510 --> 00:15:50,330 And find yourself running through these same things. 352 00:15:50,330 --> 00:15:52,130 There's some interesting research somebody 353 00:15:52,130 --> 00:15:56,500 did because in Chinese, the standard way of doing a times 354 00:15:56,500 --> 00:15:57,440 table-- 355 00:15:57,440 --> 00:16:00,935 and this is what I was told, OK, this is second-hand knowledge-- 356 00:16:00,935 --> 00:16:03,275 is instead of saying 1 times 1 is 2-- 357 00:16:03,275 --> 00:16:04,775 AUDIENCE: They just do [INAUDIBLE].. 358 00:16:04,775 --> 00:16:07,130 PROFESSOR: 1, 1, 2, or 1 times 1 is 1. 359 00:16:07,130 --> 00:16:08,630 I may not be fibbing to you guys. 360 00:16:08,630 --> 00:16:09,660 And so it's faster. 361 00:16:09,660 --> 00:16:12,590 And so they're actually noticeably measurably faster 362 00:16:12,590 --> 00:16:14,810 on a lot of these quick arithmetic tasks 363 00:16:14,810 --> 00:16:17,660 simply because the verbal encoding of the information 364 00:16:17,660 --> 00:16:22,170 is that much shorter, which is interesting. 365 00:16:22,170 --> 00:16:24,830 I don't think you're likely to see the way that we teach 366 00:16:24,830 --> 00:16:27,560 timetables in American schools changing just for that anytime 367 00:16:27,560 --> 00:16:31,160 soon, but a good example of partly just how 368 00:16:31,160 --> 00:16:34,210 tied to the language we speak we are. 369 00:16:34,210 --> 00:16:38,030 OK, one more thing that affects your working memory capability 370 00:16:38,030 --> 00:16:40,400 is how similar the things that you're 371 00:16:40,400 --> 00:16:45,200 trying to put into your working memory are. 372 00:16:45,200 --> 00:16:51,050 If I ask you to remember a lot of words for parts of a car 373 00:16:51,050 --> 00:16:54,230 and then test you with words that were and were not 374 00:16:54,230 --> 00:16:57,170 in the original list, you're going 375 00:16:57,170 --> 00:17:00,310 to be prone to thinking-- if I ask you to memorize 376 00:17:00,310 --> 00:17:04,790 seat belt, hubcap, hood, trunk, and then ask you if engine was 377 00:17:04,790 --> 00:17:07,250 in the list, you're likely to think that engine was, 378 00:17:07,250 --> 00:17:09,650 even though it wasn't ever. 379 00:17:09,650 --> 00:17:11,690 Likewise, if I give you a set of four words that 380 00:17:11,690 --> 00:17:13,369 are all parts of a car and then throw in 381 00:17:13,369 --> 00:17:15,026 chocolate along with them, you're 382 00:17:15,026 --> 00:17:16,400 more likely to remember chocolate 383 00:17:16,400 --> 00:17:17,540 than any of the other ones. 384 00:17:17,540 --> 00:17:19,859 The fact that it is different causes 385 00:17:19,859 --> 00:17:21,800 it to stand out in some way, and you're 386 00:17:21,800 --> 00:17:24,150 more likely to recall it. 387 00:17:24,150 --> 00:17:28,089 So these are all things that affect what gets held onto well 388 00:17:28,089 --> 00:17:31,400 in working memory, how much stuff you can put there. 389 00:17:31,400 --> 00:17:38,280 But what, anyways, is working memory? 390 00:17:38,280 --> 00:17:41,150 Remember, Atkinson and Shiffrin said short-term memory is just 391 00:17:41,150 --> 00:17:42,330 all one box. 392 00:17:42,330 --> 00:17:45,370 And this is a very linear system. 393 00:17:45,370 --> 00:17:49,654 Well, some guy in the '70s-- 394 00:17:49,654 --> 00:17:50,610 oh, backing up. 395 00:17:50,610 --> 00:17:53,600 OK, one more thing I wanted to talk about this 396 00:17:53,600 --> 00:17:56,272 is what happens when you're losing stuff out of memory? 397 00:17:56,272 --> 00:17:58,730 And this is something that most people who study short-term 398 00:17:58,730 --> 00:17:59,300 memory-- 399 00:17:59,300 --> 00:18:02,390 working memory-- and people who study long-term memory 400 00:18:02,390 --> 00:18:03,660 argue about. 401 00:18:03,660 --> 00:18:05,870 Does information that is in storage 402 00:18:05,870 --> 00:18:07,430 just drop off over time? 403 00:18:07,430 --> 00:18:10,920 Does it just fall out of your brain somehow-- 404 00:18:10,920 --> 00:18:13,550 the representation naturally decays? 405 00:18:13,550 --> 00:18:20,810 Or does other information that's coming in interfere 406 00:18:20,810 --> 00:18:23,910 with it-- mess it up? 407 00:18:23,910 --> 00:18:27,050 All right, so these guys were really 408 00:18:27,050 --> 00:18:30,680 trying to put together a model that is about how you learn, 409 00:18:30,680 --> 00:18:34,040 how you get information from your sensory inputs 410 00:18:34,040 --> 00:18:37,710 into long-term memory. 411 00:18:37,710 --> 00:18:40,599 And more recently, scientists have been like, 412 00:18:40,599 --> 00:18:41,640 hey look, working memory. 413 00:18:41,640 --> 00:18:44,149 Working memory is required for a whole bundle 414 00:18:44,149 --> 00:18:45,440 of different things that we do. 415 00:18:45,440 --> 00:18:46,730 It's required for math. 416 00:18:46,730 --> 00:18:48,200 It's required for counting. 417 00:18:48,200 --> 00:18:51,420 It's required for reading or holding a conversation. 418 00:18:51,420 --> 00:18:55,701 It's required for following driving directions. 419 00:18:55,701 --> 00:18:57,200 All of these are things where you've 420 00:18:57,200 --> 00:19:00,710 got to hold information where it's accessible, 421 00:19:00,710 --> 00:19:02,987 where you can work with it and use it to influence 422 00:19:02,987 --> 00:19:04,070 other things you're doing. 423 00:19:16,030 --> 00:19:23,190 So Baddeley and Hitch-- the current canonical model 424 00:19:23,190 --> 00:19:25,800 of short-term memory was come up with by this guy 425 00:19:25,800 --> 00:19:28,370 named Alan Baddeley and Christopher Hitch, 426 00:19:28,370 --> 00:19:30,270 although Baddeley's the guy who's 427 00:19:30,270 --> 00:19:31,980 really still working on it. 428 00:19:31,980 --> 00:19:33,390 And these guys said, OK, so if we 429 00:19:33,390 --> 00:19:37,210 want to figure out how short-term memory works, 430 00:19:37,210 --> 00:19:41,880 we've got to be able to really define what it does. 431 00:19:41,880 --> 00:19:44,550 It's not just a box into which information 432 00:19:44,550 --> 00:19:46,230 is put before it can get [? passed ?] 433 00:19:46,230 --> 00:19:47,950 to long-term memory. 434 00:19:47,950 --> 00:19:50,350 Short-term memory-- working memory-- 435 00:19:50,350 --> 00:19:52,830 holds several pieces of information 436 00:19:52,830 --> 00:19:55,920 that may or may not be related in some way 437 00:19:55,920 --> 00:19:58,020 so they can be worked with and processed. 438 00:19:58,020 --> 00:20:01,170 So the stuff that you're actively thinking about, 439 00:20:01,170 --> 00:20:04,110 actively reasoning about, actively 440 00:20:04,110 --> 00:20:06,900 applying cognitive capabilities to 441 00:20:06,900 --> 00:20:08,500 is stuff that's in working memory. 442 00:20:08,500 --> 00:20:10,980 Working memory is like a workbench where 443 00:20:10,980 --> 00:20:12,538 you can work with information. 444 00:20:15,082 --> 00:20:16,790 So Baddeley and Hitch came up with what's 445 00:20:16,790 --> 00:20:21,050 pretty much the canonical model of working memory. 446 00:20:21,050 --> 00:20:24,980 And they said that working memory-- 447 00:20:24,980 --> 00:20:27,020 and they came up with this name that is working, 448 00:20:27,020 --> 00:20:30,350 rather than short-term-- 449 00:20:30,350 --> 00:20:32,001 who's drawing stuff on my boards? 450 00:20:32,001 --> 00:20:32,500 Huh? 451 00:20:32,500 --> 00:20:33,300 Huh? 452 00:20:33,300 --> 00:20:36,155 Who got yellow chalk? 453 00:20:36,155 --> 00:20:37,970 I want yellow chalk. 454 00:20:37,970 --> 00:20:43,150 OK, so they said there's three main parts 455 00:20:43,150 --> 00:20:54,150 to working memory, that there's this phonological part 456 00:20:54,150 --> 00:20:56,660 over here. 457 00:20:56,660 --> 00:20:58,530 And there's this visuospatial part-- 458 00:21:03,470 --> 00:21:07,026 and I'm writing that too high up-- go. 459 00:21:07,026 --> 00:21:08,650 And they refer to the visuospatial part 460 00:21:08,650 --> 00:21:10,500 as a sketch pad. 461 00:21:10,500 --> 00:21:12,090 Some people call it a scratch pad. 462 00:21:12,090 --> 00:21:13,506 Some people just call it a module. 463 00:21:16,970 --> 00:21:19,990 And then there's a central executive component. 464 00:21:25,430 --> 00:21:30,890 And the central executive is connected to the other parts. 465 00:21:30,890 --> 00:21:33,900 It's also connected to long-term memory. 466 00:21:37,870 --> 00:21:39,840 It's also connected to sensory input. 467 00:21:46,760 --> 00:21:51,250 So these three parts, together, comprise a model 468 00:21:51,250 --> 00:21:57,280 for working memory that allows it to actively hold, change, 469 00:21:57,280 --> 00:22:00,320 modify, process information. 470 00:22:16,000 --> 00:22:20,310 So let's talk first about the phonological loop-- 471 00:22:20,310 --> 00:22:22,530 this verbal part. 472 00:22:22,530 --> 00:22:28,650 So phonological is phonology, is a linguistics subfield. 473 00:22:28,650 --> 00:22:32,350 It's the study of the sounds in a particular language. 474 00:22:32,350 --> 00:22:36,360 So phonological means, like, auditory sound based. 475 00:22:36,360 --> 00:22:40,180 So why did they-- 476 00:22:40,180 --> 00:22:40,680 backing up. 477 00:22:40,680 --> 00:22:42,150 I keep getting ahead of myself with this stuff 478 00:22:42,150 --> 00:22:42,870 because I think it's cool. 479 00:22:42,870 --> 00:22:44,460 OK, so why did Baddeley and Hitch 480 00:22:44,460 --> 00:22:47,070 think that there's three parts to working memory? 481 00:22:47,070 --> 00:22:50,820 Well, so for example, they had this 1974 paper 482 00:22:50,820 --> 00:22:53,610 where they had subjects memorize a string of digits. 483 00:22:53,610 --> 00:22:57,930 Between, I think, two and eight digits were presented. 484 00:22:57,930 --> 00:23:03,420 So you might have to memorize 3, 1, 7, or 8, 2, 4, 5, 0, 485 00:23:03,420 --> 00:23:05,560 or something. 486 00:23:05,560 --> 00:23:07,060 And they had subjects rehearse them. 487 00:23:07,060 --> 00:23:09,060 So that would be like saying, you read 3, 1, 7. 488 00:23:09,060 --> 00:23:11,910 And you [INAUDIBLE] 3, 1, 7, 3, 1, 7, 3, 1, 7. 489 00:23:11,910 --> 00:23:12,601 I can do this. 490 00:23:12,601 --> 00:23:14,100 3, 1, 7, because as long as you just 491 00:23:14,100 --> 00:23:15,516 keep saying it to yourself, you're 492 00:23:15,516 --> 00:23:17,460 not going to forget it, right? 493 00:23:17,460 --> 00:23:19,500 And at the same time, they had them 494 00:23:19,500 --> 00:23:23,800 perform a spatial reasoning task, which was pretty basic. 495 00:23:23,800 --> 00:23:30,390 They just showed them two letters, like on a screen. 496 00:23:30,390 --> 00:23:39,870 And they asked them X follows Y. And they 497 00:23:39,870 --> 00:23:44,730 had to push either a button for Yes or a button for No. 498 00:23:44,730 --> 00:23:46,530 And they had them do this at the same time 499 00:23:46,530 --> 00:23:48,738 that they were still rehearsing the string of digits. 500 00:23:51,790 --> 00:23:54,670 And what you would imagine is that the people who are hanging 501 00:23:54,670 --> 00:23:57,400 on to a longer string of digits-- 502 00:23:57,400 --> 00:23:59,770 an eight-digit string-- 503 00:23:59,770 --> 00:24:02,650 would be poorer at this than people who are 504 00:24:02,650 --> 00:24:05,086 hanging on to a shorter one. 505 00:24:05,086 --> 00:24:10,430 But what they found is that people in the short 506 00:24:10,430 --> 00:24:14,830 and in the long conditions were equally accurate 507 00:24:14,830 --> 00:24:17,770 on the spatial task. 508 00:24:17,770 --> 00:24:24,220 So no matter how long a digit string you're rehearsing, 509 00:24:24,220 --> 00:24:28,090 you can still do this "which letters in which order" task. 510 00:24:28,090 --> 00:24:32,295 And they were only a smidgen slower 511 00:24:32,295 --> 00:24:34,840 at answering yes or no to this task. 512 00:24:34,840 --> 00:24:37,900 So we talked yesterday about divided attention. 513 00:24:37,900 --> 00:24:41,290 You can see some of a divided attention effect here. 514 00:24:41,290 --> 00:24:43,840 People who are holding onto longer digit strings 515 00:24:43,840 --> 00:24:48,110 are performing poorer on this spatial task, 516 00:24:48,110 --> 00:24:51,400 but not by as much as anybody expected. 517 00:24:51,400 --> 00:24:53,140 And Baddeley and Hitch used this data 518 00:24:53,140 --> 00:24:57,880 to argue that these two tasks are 519 00:24:57,880 --> 00:25:00,190 using different parts of your working memory, 520 00:25:00,190 --> 00:25:02,080 that the rehearsal of the digit string, 521 00:25:02,080 --> 00:25:05,980 where you're going 8, 4, 3, 5, 0, 8, 4, 3, 5, 0-- 522 00:25:05,980 --> 00:25:08,390 not even out loud, right, but in your head-- 523 00:25:08,390 --> 00:25:10,390 and this spatial task, where they're 524 00:25:10,390 --> 00:25:12,670 thinking about which order the letters are in 525 00:25:12,670 --> 00:25:15,320 are using different parts of your working memory. 526 00:25:15,320 --> 00:25:18,370 And so you're not seeing as much interference between them 527 00:25:18,370 --> 00:25:19,620 as you might otherwise expect. 528 00:25:26,070 --> 00:25:28,140 OK, so the first part we're going to talk about 529 00:25:28,140 --> 00:25:30,450 is this phonological loop part. 530 00:25:30,450 --> 00:25:34,380 So this stores a limited number of sounds 531 00:25:34,380 --> 00:25:36,150 for a short period of time. 532 00:25:39,480 --> 00:25:43,350 If you are thinking about what remembering a digit string 533 00:25:43,350 --> 00:25:45,810 feels like, if you read a string of numbers, 534 00:25:45,810 --> 00:25:50,640 you probably feel like you hear yourself reading them aloud 535 00:25:50,640 --> 00:25:54,000 as you read them, even if you're reading silently. 536 00:25:54,000 --> 00:25:55,710 And as you rehearse it, you probably 537 00:25:55,710 --> 00:25:58,240 feel like you're saying-- 538 00:25:58,240 --> 00:26:03,310 "saying"-- them over and over subvocally in your head. 539 00:26:03,310 --> 00:26:06,720 That's this phonological loop going on. 540 00:26:06,720 --> 00:26:10,050 So as long as you're rehearsing a digit string like that, 541 00:26:10,050 --> 00:26:12,330 your recall for it is, as far as anyone can tell, 542 00:26:12,330 --> 00:26:13,390 pretty much endless. 543 00:26:13,390 --> 00:26:18,180 You can hold this eight-digit string for about 20 minutes. 544 00:26:18,180 --> 00:26:19,980 And certain kinds of errors in recall 545 00:26:19,980 --> 00:26:22,539 can get traced back to confusions 546 00:26:22,539 --> 00:26:23,580 in the phonological loop. 547 00:26:23,580 --> 00:26:24,720 If you're remembering letters, you 548 00:26:24,720 --> 00:26:26,720 might get two letters that rhyme with each other 549 00:26:26,720 --> 00:26:29,196 mixed up because they sound the same. 550 00:26:29,196 --> 00:26:30,570 So certain kinds of errors can be 551 00:26:30,570 --> 00:26:33,600 traced to acoustical confusions in the phonological loop. 552 00:26:33,600 --> 00:26:36,240 Take a look at your string of letters and numbers 553 00:26:36,240 --> 00:26:38,059 from 10 minutes ago. 554 00:26:38,059 --> 00:26:39,600 Did anyone have any mistakes in there 555 00:26:39,600 --> 00:26:42,232 where they got two letters that sounded the same mixed up 556 00:26:42,232 --> 00:26:43,440 or a letter that replaced it? 557 00:26:43,440 --> 00:26:44,165 One, yeah. 558 00:26:44,165 --> 00:26:45,918 AUDIENCE: Nearly, but I switched it back 559 00:26:45,918 --> 00:26:49,710 because after I thought about it, I kind wasn't sure. 560 00:26:49,710 --> 00:26:50,831 But I switched it back. 561 00:26:50,831 --> 00:26:51,330 So-- 562 00:26:51,330 --> 00:26:52,205 PROFESSOR: Which two? 563 00:26:52,205 --> 00:26:53,094 What two letters? 564 00:26:53,094 --> 00:26:54,750 AUDIENCE: X and S. 565 00:26:54,750 --> 00:26:55,560 PROFESSOR: X and S? 566 00:26:55,560 --> 00:26:57,280 You guys can hear the acoustical confusion there. 567 00:26:57,280 --> 00:26:58,140 AUDIENCE: But I still switched it back. 568 00:26:58,140 --> 00:26:59,590 PROFESSOR: You caught it. 569 00:26:59,590 --> 00:27:02,344 AUDIENCE: I almost did B and V, but then I [INAUDIBLE].. 570 00:27:02,344 --> 00:27:03,510 PROFESSOR: Same thing, yeah. 571 00:27:03,510 --> 00:27:04,640 AUDIENCE: I got one and nine-- 572 00:27:04,640 --> 00:27:05,220 PROFESSOR: One and nine? 573 00:27:05,220 --> 00:27:06,679 AUDIENCE: [INAUDIBLE] I and Y. 574 00:27:06,679 --> 00:27:07,470 PROFESSOR: I and Y? 575 00:27:07,470 --> 00:27:09,330 One and nine-- they've both got the N in there. 576 00:27:09,330 --> 00:27:10,080 They're both numbers. 577 00:27:10,080 --> 00:27:12,413 AUDIENCE: [? I ?] [? nearly ?] [? got 1 and ?] [? 4s. ?] 578 00:27:12,413 --> 00:27:14,460 PROFESSOR: Yeah, it's certainly not the only way 579 00:27:14,460 --> 00:27:16,260 that you can make working memory errors. 580 00:27:16,260 --> 00:27:18,051 But it's definitely something that happens. 581 00:27:20,700 --> 00:27:24,120 Your classic Freudian slip is sort of related to this. 582 00:27:24,120 --> 00:27:27,367 You've got two things that sound enough alike, 583 00:27:27,367 --> 00:27:29,200 and you've got both of them kind of in mind. 584 00:27:29,200 --> 00:27:32,120 And whoops, the wrong thing just came out. 585 00:27:32,120 --> 00:27:36,650 OK, so we've got a phonological loop. 586 00:27:36,650 --> 00:27:37,570 Here we go. 587 00:27:37,570 --> 00:27:39,250 All right, read the digits below. 588 00:27:39,250 --> 00:27:41,375 Close your eyes, and try to remember them silently. 589 00:27:41,375 --> 00:27:43,260 So this is that silent rehearsal thing. 590 00:27:46,424 --> 00:27:48,450 I'm getting all out of order today. 591 00:27:48,450 --> 00:27:50,274 I'm frazzled. 592 00:27:50,274 --> 00:27:52,190 AUDIENCE: Wait, when should we close our eyes? 593 00:27:52,190 --> 00:27:53,564 PROFESSOR: Once you've read them. 594 00:27:53,564 --> 00:27:55,690 So read them to yourself, and then close your eyes. 595 00:27:55,690 --> 00:27:56,314 AUDIENCE: Wait. 596 00:27:56,314 --> 00:27:57,100 [INAUDIBLE] again. 597 00:28:06,350 --> 00:28:08,510 PROFESSOR: OK. 598 00:28:08,510 --> 00:28:10,854 Anyone think they remember them? 599 00:28:10,854 --> 00:28:11,354 Sara? 600 00:28:11,354 --> 00:28:14,240 AUDIENCE: [INAUDIBLE] 601 00:28:14,240 --> 00:28:16,490 PROFESSOR: Sounds good. 602 00:28:16,490 --> 00:28:19,850 So you read them. 603 00:28:19,850 --> 00:28:22,610 People tend to report they were experienced hearing themselves 604 00:28:22,610 --> 00:28:26,180 say them, not aloud, but imagining that you 605 00:28:26,180 --> 00:28:27,562 hear yourself say them. 606 00:28:27,562 --> 00:28:30,020 And as you rehearse them, you hear yourself repeating them. 607 00:28:30,020 --> 00:28:32,420 You feel like you keep speaking them. 608 00:28:32,420 --> 00:28:34,340 This is what's commonly reported as what 609 00:28:34,340 --> 00:28:37,440 that experience feels like. 610 00:28:37,440 --> 00:28:39,280 AUDIENCE: You can look at the patterns, 611 00:28:39,280 --> 00:28:41,730 and you'll find out that 5 is two less than 7, 612 00:28:41,730 --> 00:28:43,360 and 9 is two more than 7. 613 00:28:43,360 --> 00:28:47,093 And then you find out that 4, 1, 3, 2 is actually a pattern 614 00:28:47,093 --> 00:28:49,440 because they're all in the first four digits 615 00:28:49,440 --> 00:28:50,993 except it's switched around. 616 00:28:50,993 --> 00:28:51,820 AUDIENCE: Yeah. 617 00:28:51,820 --> 00:28:52,630 PROFESSOR: I do that with numbers 618 00:28:52,630 --> 00:28:55,171 when I know I'm not going to get to continually rehearse them 619 00:28:55,171 --> 00:28:55,882 like that. 620 00:28:55,882 --> 00:28:58,090 I do something very similar with, like, phone numbers 621 00:28:58,090 --> 00:29:00,434 that I want to learn, although I've totally 622 00:29:00,434 --> 00:29:02,350 gotten out of the habit of actually memorizing 623 00:29:02,350 --> 00:29:04,540 phone numbers anymore. 624 00:29:04,540 --> 00:29:06,042 Yeah. 625 00:29:06,042 --> 00:29:07,000 I don't know anybody's. 626 00:29:07,000 --> 00:29:08,500 And it's so funny because I remember 627 00:29:08,500 --> 00:29:10,541 being in, like, middle school and not having one. 628 00:29:10,541 --> 00:29:13,060 And I probably knew 25 phone numbers [SNAPS] like that. 629 00:29:13,060 --> 00:29:14,220 I'm not sure I-- 630 00:29:14,220 --> 00:29:15,610 I know my parents'. 631 00:29:15,610 --> 00:29:16,600 I know my husband's. 632 00:29:16,600 --> 00:29:16,950 AUDIENCE: [INAUDIBLE] 633 00:29:16,950 --> 00:29:18,040 PROFESSOR: I know mine. 634 00:29:18,040 --> 00:29:19,760 [INTERPOSING VOICES] 635 00:29:20,620 --> 00:29:22,420 PROFESSOR: And so it's interesting. 636 00:29:22,420 --> 00:29:25,120 Anyway, so what you see in that is 637 00:29:25,120 --> 00:29:27,750 that there's two different subcomponents 638 00:29:27,750 --> 00:29:29,387 of this phonological storage idea. 639 00:29:29,387 --> 00:29:30,970 There's this phonological store, which 640 00:29:30,970 --> 00:29:34,460 is that reading it out loud to yourself effect. 641 00:29:34,460 --> 00:29:38,200 That's this input coming in. 642 00:29:38,200 --> 00:29:40,645 And then there's this articulatory rehearsal process. 643 00:29:43,114 --> 00:29:44,530 And that second one is when you're 644 00:29:44,530 --> 00:29:46,821 sitting there trying to remember that string of digits. 645 00:29:46,821 --> 00:29:49,330 And you keep repeating it to yourself. 646 00:29:49,330 --> 00:29:54,470 The best example of this I know of is if you're sitting, 647 00:29:54,470 --> 00:29:57,580 and you're concentrating heavily on something. 648 00:29:57,580 --> 00:30:01,144 And somebody says something to you, and you don't catch it. 649 00:30:01,144 --> 00:30:02,560 And you look up and you say, what? 650 00:30:02,560 --> 00:30:04,030 And about the time you actually say 651 00:30:04,030 --> 00:30:07,210 what, that whatever it was that they just said to you 652 00:30:07,210 --> 00:30:09,960 reprocesses and makes itself into your brain. 653 00:30:09,960 --> 00:30:12,910 It's this kind of delayed hearing effect. 654 00:30:12,910 --> 00:30:14,950 Happens to me all the time. 655 00:30:14,950 --> 00:30:17,620 That's so this auditory information 656 00:30:17,620 --> 00:30:20,020 is held in the phonological store there. 657 00:30:20,020 --> 00:30:23,200 The phonological store, as far as anyone in this model, 658 00:30:23,200 --> 00:30:24,520 isn't semantic. 659 00:30:24,520 --> 00:30:28,452 It's holding stuff just based on sound. 660 00:30:28,452 --> 00:30:30,160 And so the auditory information coming in 661 00:30:30,160 --> 00:30:32,620 can be held for just a few seconds 662 00:30:32,620 --> 00:30:35,950 until it can get passed to the speech-processing portions 663 00:30:35,950 --> 00:30:36,520 of your mind. 664 00:30:43,340 --> 00:30:47,167 The phonological loop is used heavily when you're counting. 665 00:30:47,167 --> 00:30:49,250 Look around and try and count the number of chairs 666 00:30:49,250 --> 00:30:49,791 in this room. 667 00:30:52,580 --> 00:30:53,530 AUDIENCE: [INAUDIBLE] 668 00:30:53,530 --> 00:30:57,890 PROFESSOR: Hear yourself saying, 1, 2, 3, 4, 5. 669 00:30:57,890 --> 00:30:59,240 [INTERPOSING VOICES] 670 00:30:59,240 --> 00:31:02,270 PROFESSOR: Yes, go ahead. 671 00:31:02,270 --> 00:31:04,560 Now, under your breath to yourself, 672 00:31:04,560 --> 00:31:09,012 say, 6, 8, 6, 8, 6, 8. 673 00:31:09,012 --> 00:31:10,970 Just keep saying the numbers, and try and count 674 00:31:10,970 --> 00:31:13,029 at the same time. 675 00:31:13,029 --> 00:31:13,570 AUDIENCE: Oh. 676 00:31:15,805 --> 00:31:17,930 PROFESSOR: All right, I won't make you do a number. 677 00:31:17,930 --> 00:31:20,480 Just say, like, the, the, the, the, the under your breath 678 00:31:20,480 --> 00:31:21,230 and try and count. 679 00:31:23,910 --> 00:31:26,650 Fast-- say it fast. 680 00:31:26,650 --> 00:31:28,610 AUDIENCE: Ugh. 681 00:31:28,610 --> 00:31:29,610 PROFESSOR: Is it harder? 682 00:31:32,592 --> 00:31:34,580 AUDIENCE: Was it in the reading, this thing? 683 00:31:34,580 --> 00:31:36,068 Oh, I thought [? it was, like, ?] 684 00:31:36,068 --> 00:31:36,568 [? a law. ?] 685 00:31:36,568 --> 00:31:39,060 [INTERPOSING VOICES] 686 00:31:39,060 --> 00:31:40,820 PROFESSOR: Yeah, you can do it with laws. 687 00:31:40,820 --> 00:31:43,032 The 6, 8, 6, 8 is one that my friend 688 00:31:43,032 --> 00:31:44,990 who does attention and working memory research. 689 00:31:44,990 --> 00:31:47,900 All of these, by having you say something or even just 690 00:31:47,900 --> 00:31:52,210 say it under your breath, you're filling up this working memory 691 00:31:52,210 --> 00:31:53,290 phonological space. 692 00:31:53,290 --> 00:31:56,680 You're demanding that it just work on what you're saying. 693 00:31:56,680 --> 00:32:01,940 And you can't let it do the counting or the rehearsal that 694 00:32:01,940 --> 00:32:03,480 are so important to it. 695 00:32:03,480 --> 00:32:07,880 So hard to count when your speech 696 00:32:07,880 --> 00:32:09,150 centers are doing other stuff. 697 00:32:11,970 --> 00:32:12,860 Come on. 698 00:32:12,860 --> 00:32:13,710 Space-- there we go. 699 00:32:13,710 --> 00:32:16,890 OK, so the other big half here is this idea 700 00:32:16,890 --> 00:32:21,150 of a visuospatial sketchpad. 701 00:32:21,150 --> 00:32:23,900 What is a visuospatial sketchpad? 702 00:32:23,900 --> 00:32:27,780 It stores, again, a limited amount of visual information 703 00:32:27,780 --> 00:32:30,130 or spatial information. 704 00:32:30,130 --> 00:32:33,060 So, for example, picture yourself 705 00:32:33,060 --> 00:32:35,580 in a room you know your way around very well-- maybe 706 00:32:35,580 --> 00:32:37,470 your room at home. 707 00:32:37,470 --> 00:32:44,730 And think about what's hanging on the walls. 708 00:32:44,730 --> 00:32:49,217 Just walk around your room and count them off. 709 00:32:49,217 --> 00:32:49,800 Can you do it? 710 00:32:49,800 --> 00:32:53,640 AUDIENCE: Hey, I should really clean. 711 00:32:53,640 --> 00:32:55,110 PROFESSOR: So in a task like this, 712 00:32:55,110 --> 00:32:58,350 where you're envisioning a space, 713 00:32:58,350 --> 00:33:01,770 you're using your visuospatial information 714 00:33:01,770 --> 00:33:05,479 to pull up what's basically a visual recollection. 715 00:33:05,479 --> 00:33:07,020 Remember that working memory not only 716 00:33:07,020 --> 00:33:08,270 takes sensory input coming in. 717 00:33:08,270 --> 00:33:10,980 It also pulls things from long-term storage. 718 00:33:10,980 --> 00:33:14,550 The appearance of your room is probably a long-term storage 719 00:33:14,550 --> 00:33:15,450 piece of information. 720 00:33:15,450 --> 00:33:16,570 And you can pull it up. 721 00:33:16,570 --> 00:33:18,570 You can picture your room from different angles. 722 00:33:18,570 --> 00:33:19,740 You can stand at the door and think 723 00:33:19,740 --> 00:33:21,510 about what it looks like, and picture yourself 724 00:33:21,510 --> 00:33:22,968 walking across the room and looking 725 00:33:22,968 --> 00:33:24,810 at it from the other angle. 726 00:33:24,810 --> 00:33:27,350 So the visuospatial sketchpad lets us have this data. 727 00:33:27,350 --> 00:33:28,844 We can modify it. 728 00:33:28,844 --> 00:33:29,760 We can move it around. 729 00:33:32,490 --> 00:33:36,200 Another good example of this is, OK, so picture a letter 730 00:33:36,200 --> 00:33:38,070 D, right? 731 00:33:38,070 --> 00:33:41,960 Rotate it 90 degrees to the right. 732 00:33:41,960 --> 00:33:44,040 AUDIENCE: You mean clockwise? 733 00:33:44,040 --> 00:33:46,650 PROFESSOR: Top to the right, yeah, clockwise. 734 00:33:46,650 --> 00:33:49,590 Put a number 4 above it. 735 00:33:49,590 --> 00:33:52,050 Take away the horizontal segment of the 4 736 00:33:52,050 --> 00:33:55,260 that's to the right of the vertical part. 737 00:33:55,260 --> 00:33:56,800 What object are you left with? 738 00:33:56,800 --> 00:33:58,520 AUDIENCE: [INAUDIBLE] 739 00:33:58,520 --> 00:34:00,510 PROFESSOR: Imagine a D-- 740 00:34:00,510 --> 00:34:01,220 capital D. 741 00:34:01,220 --> 00:34:03,302 AUDIENCE: Wait, like a 4 that's [INAUDIBLE]?? 742 00:34:03,302 --> 00:34:04,010 PROFESSOR: Mm-hm. 743 00:34:04,010 --> 00:34:05,051 AUDIENCE: Oh, a sailboat. 744 00:34:05,051 --> 00:34:05,836 PROFESSOR: Yeah. 745 00:34:05,836 --> 00:34:06,336 Good. 746 00:34:06,336 --> 00:34:06,919 AUDIENCE: Yay. 747 00:34:10,010 --> 00:34:11,260 PROFESSOR: Picture a letter D. 748 00:34:11,260 --> 00:34:12,670 AUDIENCE: Well, it depends on how you draw your 4. 749 00:34:12,670 --> 00:34:13,210 PROFESSOR: It does. 750 00:34:13,210 --> 00:34:15,142 AUDIENCE: Yeah, I don't draw my 4s like that. 751 00:34:15,142 --> 00:34:17,615 [INTERPOSING VOICES] 752 00:34:17,615 --> 00:34:19,345 PROFESSOR: A printed 4, yes. 753 00:34:19,345 --> 00:34:20,800 AUDIENCE: Do you do the curly 2s or the regular 2s? 754 00:34:20,800 --> 00:34:21,770 AUDIENCE: Regular 2s. 755 00:34:21,770 --> 00:34:22,740 AUDIENCE: I used to do the curly 2s. 756 00:34:22,740 --> 00:34:24,690 AUDIENCE: I did the curly 2s [INAUDIBLE].. 757 00:34:24,690 --> 00:34:26,939 PROFESSOR: All right, so this is another good example. 758 00:34:26,939 --> 00:34:30,510 This is another task that's using your visuospatial 759 00:34:30,510 --> 00:34:36,949 sketchpad to actively take a simple image-- a D-- 760 00:34:36,949 --> 00:34:39,889 probably from storage-- we all know what a D looks like-- 761 00:34:39,889 --> 00:34:41,364 and then do some operations on it. 762 00:34:41,364 --> 00:34:42,739 So you can take this information. 763 00:34:42,739 --> 00:34:44,540 You can modify it. 764 00:34:44,540 --> 00:34:46,639 This is a really good example of this idea 765 00:34:46,639 --> 00:34:49,580 that Baddeley has that working memory is not just 766 00:34:49,580 --> 00:34:50,960 a storage box. 767 00:34:50,960 --> 00:34:52,978 It's actually an active work space. 768 00:34:57,139 --> 00:35:00,720 Visual encoding seems to be not the preferred means 769 00:35:00,720 --> 00:35:02,940 of encoding for most material. 770 00:35:02,940 --> 00:35:16,860 Most people seem to encode stuff verbally when given the option. 771 00:35:16,860 --> 00:35:21,300 So there was a pretty classic study where they gave subjects 772 00:35:21,300 --> 00:35:24,960 a set of, like, six photos of simple nouns, 773 00:35:24,960 --> 00:35:27,850 like a piece of candy, or a pipe, 774 00:35:27,850 --> 00:35:32,250 or I don't remember what else was in it. 775 00:35:32,250 --> 00:35:34,380 And they were asked to memorize the list. 776 00:35:34,380 --> 00:35:39,870 And when they would prove that they had memorized 777 00:35:39,870 --> 00:35:45,180 the list of objects by taking a set of cards of these objects 778 00:35:45,180 --> 00:35:49,860 and putting them in the correct order, 779 00:35:49,860 --> 00:35:51,510 then the experimenter said, so OK, I 780 00:35:51,510 --> 00:35:55,230 want you to visualize object number four. 781 00:35:55,230 --> 00:35:56,250 So if object-- 782 00:35:56,250 --> 00:35:57,870 I will not knock stuff over. 783 00:35:57,870 --> 00:36:00,629 So if object number four was the piece of candy-- 784 00:36:00,629 --> 00:36:02,295 so we're thinking something that looks-- 785 00:36:06,340 --> 00:36:11,035 all right, now it's a piece of candy, right? 786 00:36:11,035 --> 00:36:12,410 And the experimenter would say, I 787 00:36:12,410 --> 00:36:14,510 want you to visualize object number four. 788 00:36:14,510 --> 00:36:18,045 And I want you to subtract the part of it 789 00:36:18,045 --> 00:36:18,920 that looks like this. 790 00:36:18,920 --> 00:36:20,110 What do you see? 791 00:36:20,110 --> 00:36:21,110 AUDIENCE: Fish. 792 00:36:21,110 --> 00:36:21,980 PROFESSOR: Right. 793 00:36:21,980 --> 00:36:23,684 So what they found was that-- 794 00:36:23,684 --> 00:36:24,632 [INTERPOSING VOICES] 795 00:36:24,632 --> 00:36:26,528 AUDIENCE: I see a [INAUDIBLE]. 796 00:36:26,528 --> 00:36:29,372 AUDIENCE: Where the fish? 797 00:36:29,372 --> 00:36:32,220 If you drew the [INAUDIBLE]. 798 00:36:32,220 --> 00:36:37,900 PROFESSOR: Yeah, I think my candy wrappers are 799 00:36:37,900 --> 00:36:40,460 a little bit too enthusiastic. 800 00:36:40,460 --> 00:36:41,210 What do you think? 801 00:36:41,210 --> 00:36:42,010 Does that look more fish-like? 802 00:36:42,010 --> 00:36:42,760 AUDIENCE: Yeah. 803 00:36:42,760 --> 00:36:43,426 PROFESSOR: Yeah. 804 00:36:43,426 --> 00:36:46,132 AUDIENCE: [INAUDIBLE] 805 00:36:46,132 --> 00:36:47,840 PROFESSOR: So they were asked to do this. 806 00:36:47,840 --> 00:36:51,760 And what they found is that for one group, 807 00:36:51,760 --> 00:36:53,275 they just didn't tell them anything 808 00:36:53,275 --> 00:36:54,400 about how to memorize them. 809 00:36:54,400 --> 00:36:56,500 They just said, go ahead and memorize these things. 810 00:36:56,500 --> 00:36:58,333 For the other group, they said, while you're 811 00:36:58,333 --> 00:37:01,720 memorizing them, sit there and go, la la la la la la la la 812 00:37:01,720 --> 00:37:07,010 la, which suppresses the phonological loop 813 00:37:07,010 --> 00:37:11,650 and keeps the verbal portion of your memorization system 814 00:37:11,650 --> 00:37:13,594 from being able to use this. 815 00:37:13,594 --> 00:37:15,010 And what they found is that people 816 00:37:15,010 --> 00:37:19,180 who'd been able to use their phonological loop to learn 817 00:37:19,180 --> 00:37:23,470 the list, who were really bad at finding 818 00:37:23,470 --> 00:37:26,320 these kind of secondary objects hidden 819 00:37:26,320 --> 00:37:30,020 in by subtracting part of the image, 820 00:37:30,020 --> 00:37:33,520 but people whose phonological loop had been suppressed 821 00:37:33,520 --> 00:37:35,110 managed to see them much better. 822 00:37:35,110 --> 00:37:44,270 And so their hypothesis is that verbal encoding these-- candy, 823 00:37:44,270 --> 00:37:47,320 pipe, house, cat-- 824 00:37:47,320 --> 00:37:52,060 was the preferred way of learning that list. 825 00:37:52,060 --> 00:37:54,910 And it wasn't unless you took that option away and forced 826 00:37:54,910 --> 00:37:57,890 people to learn it using their visual memory 827 00:37:57,890 --> 00:38:00,010 that they would do that. 828 00:38:00,010 --> 00:38:02,860 So visuospatial encoding, for most things, 829 00:38:02,860 --> 00:38:08,080 seems to be not as preferred as verbal encoding. 830 00:38:08,080 --> 00:38:12,560 All right, so we are talking about Alan Baddeley's model 831 00:38:12,560 --> 00:38:18,150 of working memory, which has three key components. 832 00:38:18,150 --> 00:38:21,160 It's got this phonological component 833 00:38:21,160 --> 00:38:24,280 and this visuospatial component. 834 00:38:24,280 --> 00:38:26,860 And Baddeley was one of the first guys who really said, 835 00:38:26,860 --> 00:38:30,010 hey, look, working memory isn't just all one unit. 836 00:38:30,010 --> 00:38:33,767 Different kinds of working-memory tasks 837 00:38:33,767 --> 00:38:36,100 don't interfere with each other as much as you'd expect. 838 00:38:36,100 --> 00:38:40,330 They seem to be being handled by independent processes. 839 00:38:40,330 --> 00:38:44,050 So he said that there's at least two kind of storage buffers-- 840 00:38:44,050 --> 00:38:46,430 this phonological auditory storage buffer, 841 00:38:46,430 --> 00:38:50,830 and this visuospatial storage buffer, and that these 842 00:38:50,830 --> 00:38:55,630 are controlled and managed by a central executive function. 843 00:38:55,630 --> 00:38:59,530 So this central executive has a lot 844 00:38:59,530 --> 00:39:05,800 of really important and really sophisticated roles. 845 00:39:05,800 --> 00:39:09,310 It determines how information is moved 846 00:39:09,310 --> 00:39:13,150 into and out of these phonological and visuospatial 847 00:39:13,150 --> 00:39:15,370 storage buffers. 848 00:39:15,370 --> 00:39:17,560 It assigns it to which one. 849 00:39:17,560 --> 00:39:20,560 So this task with the memorizing the list of shapes 850 00:39:20,560 --> 00:39:24,096 that we were talking about right before the break, 851 00:39:24,096 --> 00:39:25,720 you can think of your central executive 852 00:39:25,720 --> 00:39:28,900 as looking at that saying, OK, names of nouns-- 853 00:39:28,900 --> 00:39:31,210 nouns are easier to store as verbal information. 854 00:39:31,210 --> 00:39:33,424 I will put it in the phonological buffer. 855 00:39:33,424 --> 00:39:34,840 But in the case where subjects had 856 00:39:34,840 --> 00:39:37,720 to sit there and go, la la la la la la la la 857 00:39:37,720 --> 00:39:39,880 la, the central executive says, hey look, 858 00:39:39,880 --> 00:39:44,797 there isn't a phonological buffer available. 859 00:39:44,797 --> 00:39:46,130 All right, I feel pretty flimsy. 860 00:39:46,130 --> 00:39:47,838 Someone want to give me a hand with this? 861 00:39:47,838 --> 00:39:49,360 AUDIENCE: I'll do it. 862 00:39:49,360 --> 00:39:50,680 PROFESSOR: You want to get it? 863 00:39:50,680 --> 00:39:52,499 Rah. 864 00:39:52,499 --> 00:39:54,040 It doesn't have to be closed, closed. 865 00:39:54,040 --> 00:39:56,090 It's just loud. 866 00:39:56,090 --> 00:39:59,090 I wanted to get it down to about 6 inches open. 867 00:39:59,090 --> 00:39:59,830 I'm pathetic. 868 00:39:59,830 --> 00:40:03,340 OK, so they had this idea that this central executive 869 00:40:03,340 --> 00:40:07,780 is what's putting things into and out of these storage 870 00:40:07,780 --> 00:40:12,730 buffers and is making that judgment of, hey look, 871 00:40:12,730 --> 00:40:15,670 the phonological buffer that I would like to use-- 872 00:40:15,670 --> 00:40:16,720 [INTERPOSING VOICES] 873 00:40:16,720 --> 00:40:17,547 PROFESSOR: --is-- 874 00:40:17,547 --> 00:40:18,680 AUDIENCE: [INAUDIBLE] 875 00:40:18,680 --> 00:40:21,880 PROFESSOR: --full, is busy, I can't have it. 876 00:40:21,880 --> 00:40:25,670 I'm going to store this information visually instead. 877 00:40:25,670 --> 00:40:27,970 So it's taking all of this information 878 00:40:27,970 --> 00:40:30,430 from both those short-term buffers that 879 00:40:30,430 --> 00:40:32,070 are part of working memory. 880 00:40:32,070 --> 00:40:33,960 It's taking information from sensory input. 881 00:40:33,960 --> 00:40:36,400 It's taking information from long-term memory, 882 00:40:36,400 --> 00:40:39,430 integrating it all, coordinating it, 883 00:40:39,430 --> 00:40:41,350 and it's also got this really important role 884 00:40:41,350 --> 00:40:45,130 of suppressing irrelevant information. 885 00:40:45,130 --> 00:40:47,560 That should sound familiar. 886 00:40:47,560 --> 00:40:49,960 That's basically what your attentional system 887 00:40:49,960 --> 00:40:53,680 does, like we've been talking about for a few days. 888 00:40:53,680 --> 00:40:58,240 Attention is all about taking the flood of information that's 889 00:40:58,240 --> 00:41:00,580 available to you at any point in time, 890 00:41:00,580 --> 00:41:03,160 deciding which piece of it is most important, 891 00:41:03,160 --> 00:41:05,260 and getting the rest of it out of the way 892 00:41:05,260 --> 00:41:06,710 so it doesn't distract you, so it 893 00:41:06,710 --> 00:41:08,835 doesn't get in the way of what you're trying to do. 894 00:41:11,340 --> 00:41:15,380 So the central executive seems to be closely related 895 00:41:15,380 --> 00:41:17,120 to this attention stuff, possibly even 896 00:41:17,120 --> 00:41:21,890 the thing that does attention, although our nice competitive 897 00:41:21,890 --> 00:41:25,460 model for how attention works doesn't fit quite so neatly 898 00:41:25,460 --> 00:41:27,860 with other things that the central executive does. 899 00:41:27,860 --> 00:41:30,570 It's involved in planning, not planning. 900 00:41:30,570 --> 00:41:34,050 It's involved in coordinating behavior. 901 00:41:34,050 --> 00:41:37,490 So the central executive is this really sophisticated piece 902 00:41:37,490 --> 00:41:38,212 of the model. 903 00:41:38,212 --> 00:41:40,670 And a lot of people really don't like the central executive 904 00:41:40,670 --> 00:41:45,140 part of this, except that something needs to be doing it. 905 00:41:45,140 --> 00:41:48,710 The classic problem with thinking 906 00:41:48,710 --> 00:41:54,270 about how your brain handles any complicated processing task is 907 00:41:54,270 --> 00:41:56,650 it's tempting to believe that there's 908 00:41:56,650 --> 00:42:00,380 a mini Abby inside my head that's doing 909 00:42:00,380 --> 00:42:02,690 all of this decision-making. 910 00:42:02,690 --> 00:42:05,759 But that doesn't really solve the problem. 911 00:42:05,759 --> 00:42:07,550 It just nudges it down a level, because how 912 00:42:07,550 --> 00:42:09,949 does that mini Abby make any kind of decision? 913 00:42:09,949 --> 00:42:11,240 How does it handle information? 914 00:42:11,240 --> 00:42:12,980 Well, maybe there's a mini-mini Abby 915 00:42:12,980 --> 00:42:15,710 inside that mini Abby's head. 916 00:42:15,710 --> 00:42:17,800 And this is a really dangerous idea. 917 00:42:17,800 --> 00:42:20,030 It's not a very functional idea. 918 00:42:20,030 --> 00:42:22,310 And it's something that's really hard to keep yourself 919 00:42:22,310 --> 00:42:24,800 from falling into when you're trying to solve 920 00:42:24,800 --> 00:42:27,060 these sorts of problems. 921 00:42:27,060 --> 00:42:29,390 But nonetheless, somehow, our brains 922 00:42:29,390 --> 00:42:31,550 do need to be able to do all of these what 923 00:42:31,550 --> 00:42:33,399 are called executive functions. 924 00:42:33,399 --> 00:42:34,940 And so just saying, hey look, there's 925 00:42:34,940 --> 00:42:40,790 some portion of this that does that is important. 926 00:42:40,790 --> 00:42:41,690 AUDIENCE: Oh, OK. 927 00:42:41,690 --> 00:42:41,970 Are we going? 928 00:42:41,970 --> 00:42:42,950 PROFESSOR: Are you the introduction? 929 00:42:42,950 --> 00:42:43,610 AUDIENCE: We are the introduction. 930 00:42:43,610 --> 00:42:44,540 PROFESSOR: Then you are going first. 931 00:42:44,540 --> 00:42:44,913 AUDIENCE: OK. 932 00:42:44,913 --> 00:42:46,620 So, well, the introduction was, like, 933 00:42:46,620 --> 00:42:48,726 talking about the entire study and stuff. 934 00:42:48,726 --> 00:42:50,845 And it said that there were, like, 935 00:42:50,845 --> 00:42:53,470 two ways that spatial attention is directed to [? peripheral ?] 936 00:42:53,470 --> 00:42:54,640 [? vision ?] events. 937 00:42:54,640 --> 00:42:56,760 So there's, like, overt shifts of attention 938 00:42:56,760 --> 00:42:58,240 and covert shifts of attention. 939 00:42:58,240 --> 00:43:00,180 And overt shifts of attention are like [? where you ?] 940 00:43:00,180 --> 00:43:01,370 [? head and ?] [? eye movements ?] and stuff. 941 00:43:01,370 --> 00:43:04,140 So, like, you're deliberately moving your head and stuff 942 00:43:04,140 --> 00:43:05,580 to focus on something else. 943 00:43:05,580 --> 00:43:06,289 PROFESSOR: Mm-hm. 944 00:43:06,289 --> 00:43:08,163 AUDIENCE: And then covert shifts of attention 945 00:43:08,163 --> 00:43:10,270 are like when you're still looking at something, 946 00:43:10,270 --> 00:43:14,610 but you're focusing on something else, right? 947 00:43:14,610 --> 00:43:15,170 Yeah. 948 00:43:15,170 --> 00:43:17,461 OK, and then there are also, like, two different shifts 949 00:43:17,461 --> 00:43:18,500 of attention stimuli. 950 00:43:18,500 --> 00:43:21,429 So there's exogenous-- is that how you say it? 951 00:43:21,429 --> 00:43:22,304 PROFESSOR: Exogenous. 952 00:43:22,304 --> 00:43:24,490 AUDIENCE: OK, exogenous, which is we 953 00:43:24,490 --> 00:43:26,574 have, like, externally driven shifts of attention. 954 00:43:26,574 --> 00:43:29,031 So that means, like, if you see a crash or something, like, 955 00:43:29,031 --> 00:43:31,370 something can be outside that makes you pay attention 956 00:43:31,370 --> 00:43:32,220 to that. 957 00:43:32,220 --> 00:43:34,985 And then endogenous shifts of attention 958 00:43:34,985 --> 00:43:36,610 are like strategic shifts of attention. 959 00:43:36,610 --> 00:43:39,110 So it's like when you, like, purposely try and make yourself 960 00:43:39,110 --> 00:43:40,494 pay attention to something. 961 00:43:40,494 --> 00:43:41,327 AUDIENCE: [CHUCKLES] 962 00:43:41,327 --> 00:43:42,410 AUDIENCE: What? 963 00:43:42,410 --> 00:43:43,877 PROFESSOR: Yep, no, that's great. 964 00:43:43,877 --> 00:43:44,354 AUDIENCE: [CHUCKLES] 965 00:43:44,354 --> 00:43:45,604 AUDIENCE: Stop laughing at me. 966 00:43:45,604 --> 00:43:47,216 AUDIENCE: I'm not. 967 00:43:47,216 --> 00:43:50,053 AUDIENCE: OK, yeah, and then both covert and overt shifts 968 00:43:50,053 --> 00:43:52,814 of attention can be either exogenous or endogenous. 969 00:43:56,190 --> 00:43:58,990 And in previous studies, there's been, like, 970 00:43:58,990 --> 00:44:02,066 three possible relationship theory things between covert 971 00:44:02,066 --> 00:44:03,850 and over shifts of attention. 972 00:44:03,850 --> 00:44:06,950 So one is that, like, covert and over shifts of attention 973 00:44:06,950 --> 00:44:09,110 are completely independent of one another, which 974 00:44:09,110 --> 00:44:11,105 means that they happen in, like, different parts of the brain 975 00:44:11,105 --> 00:44:11,605 and stuff. 976 00:44:11,605 --> 00:44:13,642 And they just aren't even connected at all. 977 00:44:13,642 --> 00:44:14,350 PROFESSOR: Mm-hm. 978 00:44:14,350 --> 00:44:15,724 AUDIENCE: And then another one is 979 00:44:15,724 --> 00:44:17,500 covert and over shifts of attention 980 00:44:17,500 --> 00:44:19,865 are completely interdependent, which means, 981 00:44:19,865 --> 00:44:22,948 like, they use the same part of the brain, basically. 982 00:44:28,195 --> 00:44:30,103 [INAUDIBLE] 983 00:44:30,103 --> 00:44:31,120 AUDIENCE: It's three. 984 00:44:31,120 --> 00:44:32,355 One is they're independent. 985 00:44:32,355 --> 00:44:33,780 [INTERPOSING VOICES] 986 00:44:33,780 --> 00:44:34,730 AUDIENCE: Oh, OK. 987 00:44:34,730 --> 00:44:36,271 [INAUDIBLE],, they talk about, like-- 988 00:44:36,271 --> 00:44:38,418 AUDIENCE: One is [INAUDIBLE]. 989 00:44:38,418 --> 00:44:41,415 AUDIENCE: Like, it talked about the middle one [INAUDIBLE].. 990 00:44:41,415 --> 00:44:42,290 AUDIENCE: [INAUDIBLE] 991 00:44:42,290 --> 00:44:43,260 AUDIENCE: OK, whatever. 992 00:44:43,260 --> 00:44:44,760 PROFESSOR: Order doesn't matter. 993 00:44:44,760 --> 00:44:45,465 There are three. 994 00:44:45,465 --> 00:44:46,090 I've heard two. 995 00:44:46,090 --> 00:44:48,084 They were both ones that were listed. 996 00:44:48,084 --> 00:44:49,000 What's the third case? 997 00:44:49,000 --> 00:44:51,375 AUDIENCE: OK, so one is like when they're interdependent, 998 00:44:51,375 --> 00:44:54,050 which means they, like, use the exact same parts of the brain 999 00:44:54,050 --> 00:44:56,021 and stuff and like, yeah. 1000 00:44:56,021 --> 00:44:57,812 There's a name for that, but it's not here. 1001 00:44:57,812 --> 00:45:00,400 OK, so anyways, and then there's another one where-- 1002 00:45:00,400 --> 00:45:01,750 AUDIENCE: That's premotive. 1003 00:45:01,750 --> 00:45:03,910 AUDIENCE: OK, that one was premotive. 1004 00:45:03,910 --> 00:45:08,460 Yeah, OK, and then there's another one where it's like, 1005 00:45:08,460 --> 00:45:09,160 it's in between. 1006 00:45:09,160 --> 00:45:10,875 So, like, some parts are like the same, 1007 00:45:10,875 --> 00:45:12,083 and some parts are different. 1008 00:45:12,083 --> 00:45:13,030 PROFESSOR: Do they kind of have-- 1009 00:45:13,030 --> 00:45:14,905 AUDIENCE: So they use some of the same things 1010 00:45:14,905 --> 00:45:16,336 but not all of the same things. 1011 00:45:16,336 --> 00:45:17,044 PROFESSOR: Mm-hm. 1012 00:45:17,044 --> 00:45:19,738 AUDIENCE: [INAUDIBLE] 1013 00:45:21,690 --> 00:45:25,110 AUDIENCE: Obviously, if the whole purpose of the study was 1014 00:45:25,110 --> 00:45:28,660 just to check all the other studies' results, 1015 00:45:28,660 --> 00:45:32,740 they wouldn't get much funding, because who really goes around 1016 00:45:32,740 --> 00:45:35,550 [INAUDIBLE] looking to see if you can replicate on somebody 1017 00:45:35,550 --> 00:45:36,284 else's work? 1018 00:45:36,284 --> 00:45:38,754 It's kind of silly. 1019 00:45:38,754 --> 00:45:39,490 It is. 1020 00:45:39,490 --> 00:45:44,031 PROFESSOR: It is important but unglamorous, like many things. 1021 00:45:44,031 --> 00:45:46,560 AUDIENCE: Yeah, you usually want to be the person who 1022 00:45:46,560 --> 00:45:48,030 discovers a new thing. 1023 00:45:48,030 --> 00:45:52,650 So what they added was ways to improve the study 1024 00:45:52,650 --> 00:45:54,340 from previous ones. 1025 00:45:54,340 --> 00:45:59,160 For example, they mentioned that in previous studies, 1026 00:45:59,160 --> 00:46:03,870 the tests that aims to produce covert or covert shifts 1027 00:46:03,870 --> 00:46:08,580 of attention had different demands. 1028 00:46:08,580 --> 00:46:12,090 Like, I believe there were different points 1029 00:46:12,090 --> 00:46:15,210 in the vision, where there were different distances. 1030 00:46:15,210 --> 00:46:20,550 And that could cause the different levels of activation 1031 00:46:20,550 --> 00:46:21,410 in the brain. 1032 00:46:21,410 --> 00:46:22,700 PROFESSOR: Mm-hm. 1033 00:46:22,700 --> 00:46:24,790 AUDIENCE: Also, the previous studies 1034 00:46:24,790 --> 00:46:29,943 disagreed on whether greater neuroactivity is cause 1035 00:46:29,943 --> 00:46:32,880 by covert or overt shifts. 1036 00:46:32,880 --> 00:46:34,770 Some argue that it's one. 1037 00:46:34,770 --> 00:46:38,264 Some argue that it's the other. 1038 00:46:38,264 --> 00:46:41,086 So they want to clarify that. 1039 00:46:41,086 --> 00:46:42,310 PROFESSOR: Mm-hm. 1040 00:46:42,310 --> 00:46:45,510 AUDIENCE: Thirdly, the studies differed on 1041 00:46:45,510 --> 00:46:49,776 whether they were exogenously or endogenously driven. 1042 00:46:49,776 --> 00:46:52,080 Like, for the majority of the previous studies, 1043 00:46:52,080 --> 00:46:55,140 they were endogenous. 1044 00:46:55,140 --> 00:46:56,415 AUDIENCE: Yeah, [INAUDIBLE]. 1045 00:46:56,415 --> 00:46:57,822 AUDIENCE: Endogenous tests. 1046 00:46:57,822 --> 00:46:58,530 PROFESSOR: Mm-hm. 1047 00:46:58,530 --> 00:47:01,420 AUDIENCE: But then there was this one study 1048 00:47:01,420 --> 00:47:06,335 that found that covert and overt shifts in attention 1049 00:47:06,335 --> 00:47:08,590 produced similar results. 1050 00:47:08,590 --> 00:47:10,600 And that was with exogenous [INAUDIBLE].. 1051 00:47:10,600 --> 00:47:15,120 So they wanted to figure out if it's possible 1052 00:47:15,120 --> 00:47:19,130 that this is a confounding variable, that just 1053 00:47:19,130 --> 00:47:21,540 the difference between endogenous and exogenous 1054 00:47:21,540 --> 00:47:23,382 were resulting in different [INAUDIBLE].. 1055 00:47:23,382 --> 00:47:23,965 PROFESSOR: OK. 1056 00:47:27,850 --> 00:47:29,890 AUDIENCE: The study itself-- 1057 00:47:29,890 --> 00:47:32,950 each subject was presented with two [INAUDIBLE] tasks. 1058 00:47:32,950 --> 00:47:37,860 And the MRI [INAUDIBLE] measured. 1059 00:47:37,860 --> 00:47:41,234 One task was to perform an overt shift of attention. 1060 00:47:41,234 --> 00:47:42,900 And the other was to [INAUDIBLE] covert. 1061 00:47:42,900 --> 00:47:46,660 And both of the tasks had one peripheral shifting 1062 00:47:46,660 --> 00:47:49,237 of attention and one, like, maintaining central attention, 1063 00:47:49,237 --> 00:47:50,820 that you're looking at the same thing. 1064 00:47:50,820 --> 00:47:52,610 PROFESSOR: Right, which would be a not shift. 1065 00:47:52,610 --> 00:47:54,540 So they're comparing the case where you shift your attention 1066 00:47:54,540 --> 00:47:56,172 from central to periphery-- to one 1067 00:47:56,172 --> 00:47:57,380 where you're not shifting it. 1068 00:47:57,380 --> 00:47:58,290 AUDIENCE: Mm-hm. 1069 00:47:58,290 --> 00:47:59,873 PROFESSOR: So which of those would you 1070 00:47:59,873 --> 00:48:01,927 say is the control condition, Danny? 1071 00:48:01,927 --> 00:48:03,010 AUDIENCE: The [INAUDIBLE]? 1072 00:48:03,010 --> 00:48:05,930 PROFESSOR: Right, yeah. 1073 00:48:05,930 --> 00:48:09,750 AUDIENCE: Yeah, OK, and then [INAUDIBLE] 1074 00:48:09,750 --> 00:48:11,670 was that, like, the other ones, they only 1075 00:48:11,670 --> 00:48:14,800 took in within subject variability and not 1076 00:48:14,800 --> 00:48:16,260 between subject variability. 1077 00:48:16,260 --> 00:48:21,130 So the results couldn't be applied to larger populations. 1078 00:48:21,130 --> 00:48:24,611 So in this one, the statistical analysis 1079 00:48:24,611 --> 00:48:27,557 controlled for both the thing and between subject 1080 00:48:27,557 --> 00:48:29,060 variability. 1081 00:48:29,060 --> 00:48:32,880 And there was identical visual stimuli in both the attention 1082 00:48:32,880 --> 00:48:33,780 shift tasks. 1083 00:48:33,780 --> 00:48:35,474 So the task demands were very similar. 1084 00:48:35,474 --> 00:48:36,890 And all of the shifts of attention 1085 00:48:36,890 --> 00:48:38,742 were endogenous [INAUDIBLE]. 1086 00:48:42,920 --> 00:48:43,587 PROFESSOR: Good. 1087 00:48:43,587 --> 00:48:45,795 Anyone have any questions about any of the background 1088 00:48:45,795 --> 00:48:46,780 information for this? 1089 00:48:51,810 --> 00:48:54,110 If you do have questions, holler out. 1090 00:48:54,110 --> 00:48:57,970 All right, who's talking about the experimental procedures? 1091 00:49:00,955 --> 00:49:02,010 That you guys? 1092 00:49:02,010 --> 00:49:03,070 AUDIENCE: Yeah. 1093 00:49:03,070 --> 00:49:05,779 PROFESSOR: OK, experimental procedures, 1094 00:49:05,779 --> 00:49:07,570 which means we're reading this out of order 1095 00:49:07,570 --> 00:49:10,111 because I think it makes more sense to read about the methods 1096 00:49:10,111 --> 00:49:12,660 before you read about what they found out, because otherwise 1097 00:49:12,660 --> 00:49:13,360 you don't know what they did. 1098 00:49:13,360 --> 00:49:15,063 And it doesn't make any sense. 1099 00:49:15,063 --> 00:49:16,030 AUDIENCE: So-- 1100 00:49:16,030 --> 00:49:17,910 PROFESSOR: So, Natasha, start us off. 1101 00:49:17,910 --> 00:49:21,550 AUDIENCE: Yeah, so there were 12 subjects-- 1102 00:49:21,550 --> 00:49:25,040 9 guys and 3 girls. 1103 00:49:25,040 --> 00:49:27,840 And they were from this [INAUDIBLE] population. 1104 00:49:27,840 --> 00:49:31,550 And they were all healthy, with no neurological problems. 1105 00:49:31,550 --> 00:49:35,310 And they had normal or corrected normal vision. 1106 00:49:35,310 --> 00:49:37,710 They had to sign a form to consent, 1107 00:49:37,710 --> 00:49:41,000 and they got money for their participation. 1108 00:49:41,000 --> 00:49:43,770 So for the task design procedure, 1109 00:49:43,770 --> 00:49:47,646 there were two tasks that they did while lying 1110 00:49:47,646 --> 00:49:48,770 in the bore of the scanner. 1111 00:49:48,770 --> 00:49:49,790 What's a bore of the-- 1112 00:49:49,790 --> 00:49:51,748 PROFESSOR: It's like the hole down the middle-- 1113 00:49:51,748 --> 00:49:53,550 so while they're in the fMRI scanner. 1114 00:49:53,550 --> 00:49:56,110 AUDIENCE: And they were the covert shifts of attentions 1115 00:49:56,110 --> 00:49:58,410 and the overt shifts of attention. 1116 00:49:58,410 --> 00:50:02,130 So they had an LCD projector illuminate the projected 1117 00:50:02,130 --> 00:50:03,260 screen. 1118 00:50:03,260 --> 00:50:06,480 And the order of the tasks was [? counter-balanced. ?] 1119 00:50:06,480 --> 00:50:09,990 The stimuli was identical except for injections. 1120 00:50:09,990 --> 00:50:13,410 Both had blocked fMRI design with two 1121 00:50:13,410 --> 00:50:15,758 experimental conditions. 1122 00:50:15,758 --> 00:50:17,694 One was peripheral attention condition, 1123 00:50:17,694 --> 00:50:20,216 and the other was central attention condition. 1124 00:50:20,216 --> 00:50:25,400 And the blocks were 12.8 seconds with a 12.8-second rest between 1125 00:50:25,400 --> 00:50:26,410 them. 1126 00:50:26,410 --> 00:50:28,360 And they had a fixed A, B, A, B pattern. 1127 00:50:28,360 --> 00:50:30,318 So there was like one was peripheral attention, 1128 00:50:30,318 --> 00:50:34,037 and the next was central attention and [INAUDIBLE].. 1129 00:50:34,037 --> 00:50:36,720 And there were 49 blocks, including the rests. 1130 00:50:36,720 --> 00:50:41,440 And the peripheral attention condition 1131 00:50:41,440 --> 00:50:45,800 and central attention condition were each presented 12 times. 1132 00:50:45,800 --> 00:50:49,216 And the covert and overt shifts were run in separate sessions. 1133 00:50:51,776 --> 00:50:53,150 PROFESSOR: Anyone have questions? 1134 00:50:53,150 --> 00:50:54,500 AUDIENCE: Oh, there's more. 1135 00:50:54,500 --> 00:50:55,720 PROFESSOR: Oh. 1136 00:50:55,720 --> 00:50:57,500 [INAUDIBLE] at a time. 1137 00:50:57,500 --> 00:51:00,125 AUDIENCE: In both the covert and the overt shifts of attention, 1138 00:51:00,125 --> 00:51:02,755 test subjects were presented with pictures displaying 1139 00:51:02,755 --> 00:51:06,170 nice intros step-wise, rotating [INAUDIBLE] 1140 00:51:06,170 --> 00:51:11,170 overlaid with 8 white circles surrounded by 8 white circles. 1141 00:51:11,170 --> 00:51:24,466 You can see the picture on page 109. 1142 00:51:24,466 --> 00:51:28,770 Stimuli were shown blocks of 12.8 [INAUDIBLE],, 1143 00:51:28,770 --> 00:51:32,120 with each block composed of 16 trials, 1144 00:51:32,120 --> 00:51:36,742 and each trial lasted for 0.8 second. 1145 00:51:36,742 --> 00:51:42,835 The 9 white circles were briefly displayed for each trial. 1146 00:51:42,835 --> 00:51:45,980 And single one of the 9 white circles 1147 00:51:45,980 --> 00:51:49,670 were randomly chosen to be slightly smaller than normal. 1148 00:51:49,670 --> 00:51:54,110 During blocks, the central cross was rotated 45 degrees 1149 00:51:54,110 --> 00:51:57,070 every 3.2 seconds. 1150 00:51:57,070 --> 00:51:59,420 During the task, the only difference 1151 00:51:59,420 --> 00:52:03,110 between the peripheral and central attention condition 1152 00:52:03,110 --> 00:52:06,950 blocks was that one arm of the central cross 1153 00:52:06,950 --> 00:52:11,240 was [INAUDIBLE] during the peripheral attention. 1154 00:52:11,240 --> 00:52:13,900 In the peripheral overt attention tasks, 1155 00:52:13,900 --> 00:52:17,695 subject were told to glance at the peripheral circle 1156 00:52:17,695 --> 00:52:23,126 that the red arm of the central cross was pointing toward. 1157 00:52:23,126 --> 00:52:26,140 During the peripheral covert attention task, 1158 00:52:26,140 --> 00:52:29,920 subjects were required to press a button with their right index 1159 00:52:29,920 --> 00:52:34,380 finger if the red arm point to the smaller circle. 1160 00:52:34,380 --> 00:52:37,920 In the central covert attention task, 1161 00:52:37,920 --> 00:52:42,440 subject were similarly staring at a central white circle. 1162 00:52:42,440 --> 00:52:46,080 During the central covert attention task, 1163 00:52:46,080 --> 00:52:49,460 subject were asked to press a button with their right index 1164 00:52:49,460 --> 00:52:53,410 finger if the centrally presented circle was smaller 1165 00:52:53,410 --> 00:52:57,058 than the normal ones. 1166 00:52:57,058 --> 00:52:57,724 PROFESSOR: Good. 1167 00:52:57,724 --> 00:53:00,340 AUDIENCE: And then the researchers 1168 00:53:00,340 --> 00:53:06,070 then tried to measure covert and overt with just the attention 1169 00:53:06,070 --> 00:53:07,510 results. 1170 00:53:07,510 --> 00:53:09,140 I'm not sure how they measured it, 1171 00:53:09,140 --> 00:53:12,550 but it was probably in seconds. 1172 00:53:12,550 --> 00:53:14,500 And they measured the covert shifts attention 1173 00:53:14,500 --> 00:53:17,240 by subtracting their covert peripheral attention 1174 00:53:17,240 --> 00:53:20,190 condition from the covert central attention condition. 1175 00:53:20,190 --> 00:53:23,632 And all this referring to the neuroactivation [INAUDIBLE].. 1176 00:53:23,632 --> 00:53:24,340 PROFESSOR: Right. 1177 00:53:24,340 --> 00:53:27,040 AUDIENCE: And they did the same with the overt condition. 1178 00:53:27,040 --> 00:53:29,770 And they hoped to perform this experiment 1179 00:53:29,770 --> 00:53:32,742 on pretty advanced imaging and information technology 1180 00:53:32,742 --> 00:53:37,842 [INAUDIBLE],, including a [INAUDIBLE] imaging 1181 00:53:37,842 --> 00:53:41,286 and a fMRI imaging system. 1182 00:53:41,286 --> 00:53:42,270 And that's it. 1183 00:53:42,270 --> 00:53:44,292 PROFESSOR: Yeah, OK, so the subtraction paradigm 1184 00:53:44,292 --> 00:53:46,250 they're using is pretty typical for these kinds 1185 00:53:46,250 --> 00:53:47,210 of imaging studies. 1186 00:53:47,210 --> 00:53:49,700 If you want to know which portion of the brain 1187 00:53:49,700 --> 00:53:52,610 is activate, because let's face it, most of the time 1188 00:53:52,610 --> 00:53:54,530 most of your brain is busy. 1189 00:53:54,530 --> 00:53:56,050 It's doing something or other. 1190 00:53:56,050 --> 00:53:57,860 Your visual portion of your brain is busy. 1191 00:53:57,860 --> 00:54:00,110 And if you want to pin down what portion of your brain 1192 00:54:00,110 --> 00:54:03,960 is busy on some very small task, the model that's used 1193 00:54:03,960 --> 00:54:07,250 is to give you two tasks that are reasonably similar 1194 00:54:07,250 --> 00:54:09,464 but check the one difference you're interested in. 1195 00:54:09,464 --> 00:54:10,880 So in this case, they're comparing 1196 00:54:10,880 --> 00:54:15,860 an attention-maintenance task to an attention-shifting task, 1197 00:54:15,860 --> 00:54:17,300 look at what portions of the brain 1198 00:54:17,300 --> 00:54:20,810 have the blood flow to the brain for each of these, 1199 00:54:20,810 --> 00:54:24,050 and subtract the maintenance-attention blood 1200 00:54:24,050 --> 00:54:25,910 flow from the shifting blood flow. 1201 00:54:25,910 --> 00:54:28,670 So only the places where blood flow in the brain 1202 00:54:28,670 --> 00:54:31,310 is different in the shifting task 1203 00:54:31,310 --> 00:54:34,690 from in the maintaining-attention task 1204 00:54:34,690 --> 00:54:37,650 will show up once you do that subtraction. 1205 00:54:37,650 --> 00:54:40,790 So the goal here is that, like, because your visual system is 1206 00:54:40,790 --> 00:54:42,290 going to be doing all of this stuff, 1207 00:54:42,290 --> 00:54:44,636 just processing this white cross and white circles. 1208 00:54:44,636 --> 00:54:46,010 But hopefully, that's going to be 1209 00:54:46,010 --> 00:54:48,980 the same on both the maintaining-attention 1210 00:54:48,980 --> 00:54:50,960 and the shifting-attention conditions. 1211 00:54:50,960 --> 00:54:52,820 And so you can just subtract it out. 1212 00:54:52,820 --> 00:54:54,320 And it'll go away because that's not 1213 00:54:54,320 --> 00:54:57,380 really relevant to the shifting attention 1214 00:54:57,380 --> 00:54:59,940 that they're interested in. 1215 00:54:59,940 --> 00:55:02,750 So they get a good map for both types of tasks 1216 00:55:02,750 --> 00:55:05,140 and subtract one from the other to see what's left over 1217 00:55:05,140 --> 00:55:08,985 or what's different between them, if that makes sense. 1218 00:55:08,985 --> 00:55:09,860 Does that make sense? 1219 00:55:09,860 --> 00:55:12,580 AUDIENCE: A little. 1220 00:55:12,580 --> 00:55:19,920 PROFESSOR: So it's like if I am an fMRI researcher. 1221 00:55:19,920 --> 00:55:24,540 And here's your brain. 1222 00:55:24,540 --> 00:55:26,355 And I ask you to do something, and maybe 1223 00:55:26,355 --> 00:55:28,630 if I ask you to maintain attention, 1224 00:55:28,630 --> 00:55:30,150 and maybe there's lighting up here 1225 00:55:30,150 --> 00:55:33,540 and some down here and maybe some over here, 1226 00:55:33,540 --> 00:55:36,810 this is the maintaining attention, 1227 00:55:36,810 --> 00:55:39,240 that central condition. 1228 00:55:39,240 --> 00:55:43,388 And then I ask you to do an attentional shift condition. 1229 00:55:45,995 --> 00:55:48,120 And we find that, hey look, all of these same areas 1230 00:55:48,120 --> 00:55:51,960 still light up, but this one gets even brighter right up 1231 00:55:51,960 --> 00:55:54,450 here on the top. 1232 00:55:54,450 --> 00:55:57,620 And oh hey, a little bit down here gets extra bright, too. 1233 00:55:57,620 --> 00:56:00,839 And then if I subtract them, what's left over 1234 00:56:00,839 --> 00:56:02,880 are the things that are different between the two 1235 00:56:02,880 --> 00:56:04,088 cases, which is this little-- 1236 00:56:04,088 --> 00:56:05,580 AUDIENCE: [INAUDIBLE] 1237 00:56:05,580 --> 00:56:06,950 PROFESSOR: I don't care-- 1238 00:56:06,950 --> 00:56:08,430 extra bit. 1239 00:56:08,430 --> 00:56:09,657 You guys can figure it out. 1240 00:56:09,657 --> 00:56:12,240 There's this little extra bit up here and this little bit down 1241 00:56:12,240 --> 00:56:13,470 here. 1242 00:56:13,470 --> 00:56:16,680 And therefore, you conclude that these 1243 00:56:16,680 --> 00:56:22,005 are regions that are used in shifting but not 1244 00:56:22,005 --> 00:56:22,860 in maintenance. 1245 00:56:22,860 --> 00:56:24,764 AUDIENCE: You have to take the opposite [? of the whole ?] 1246 00:56:24,764 --> 00:56:25,716 [? back ?] [? side. ?] 1247 00:56:25,716 --> 00:56:26,670 AUDIENCE: Shh. 1248 00:56:26,670 --> 00:56:29,040 PROFESSOR: Yeah, yeah, yeah, it's the wrong direction. 1249 00:56:29,040 --> 00:56:30,332 You get the point, right? 1250 00:56:30,332 --> 00:56:31,540 So that's what they're doing. 1251 00:56:31,540 --> 00:56:34,320 They're comparing the level of activation in these two 1252 00:56:34,320 --> 00:56:36,750 similar tasks and looking at where 1253 00:56:36,750 --> 00:56:38,910 it's different in the case that they're 1254 00:56:38,910 --> 00:56:41,645 most interested in, which in this case is the shift. 1255 00:56:41,645 --> 00:56:43,728 AUDIENCE: Or you can take the absolute [? value ?] 1256 00:56:43,728 --> 00:56:45,580 [INAUDIBLE]. 1257 00:56:45,580 --> 00:56:48,090 PROFESSOR: Yes, except not because really 1258 00:56:48,090 --> 00:56:53,010 what we're interested in is both increases and decreases. 1259 00:56:53,010 --> 00:56:54,090 OK, good, yes. 1260 00:56:54,090 --> 00:56:55,770 And then they have a lot of babble-- 1261 00:56:55,770 --> 00:57:00,240 they have a lot of information about the fMRI methodology 1262 00:57:00,240 --> 00:57:01,180 that they use. 1263 00:57:01,180 --> 00:57:03,630 So they used a good fMRI. 1264 00:57:03,630 --> 00:57:10,200 And they're scanning at about a 3 millimeter cubed resolution. 1265 00:57:10,200 --> 00:57:14,640 And they're doing that neuroactivation subtraction. 1266 00:57:14,640 --> 00:57:19,410 And they're looking for what areas of the activation 1267 00:57:19,410 --> 00:57:25,770 is significantly different between the covert and overt 1268 00:57:25,770 --> 00:57:27,165 attention shifting tasks. 1269 00:57:34,860 --> 00:57:36,740 Good. 1270 00:57:36,740 --> 00:57:39,174 Anyone have questions about the methodology? 1271 00:57:39,174 --> 00:57:40,590 Everyone feel like they understand 1272 00:57:40,590 --> 00:57:43,410 what the tasks they used were and how they worked? 1273 00:57:48,710 --> 00:57:51,300 If you don't, this would be the time to say something. 1274 00:57:58,030 --> 00:57:59,870 All right, moving along then. 1275 00:57:59,870 --> 00:58:02,370 Results-- what did they find out? 1276 00:58:02,370 --> 00:58:05,070 AUDIENCE: OK, so for the results, 1277 00:58:05,070 --> 00:58:11,380 after they did the differences in the highlighted areas, 1278 00:58:11,380 --> 00:58:14,525 they wanted to find the areas of the brain that 1279 00:58:14,525 --> 00:58:18,485 were activated during covert or overt shifts of attention. 1280 00:58:18,485 --> 00:58:22,493 And they did that using two one-sample t-tests. 1281 00:58:22,493 --> 00:58:23,966 I don't know [INAUDIBLE]. 1282 00:58:23,966 --> 00:58:28,080 PROFESSOR: A t-test is a test of significance 1283 00:58:28,080 --> 00:58:28,890 between two groups. 1284 00:58:28,890 --> 00:58:31,306 AUDIENCE: It's when you don't know the population standard 1285 00:58:31,306 --> 00:58:33,534 deviation, and so you take the [INAUDIBLE].. 1286 00:58:33,534 --> 00:58:34,200 PROFESSOR: Yeah. 1287 00:58:37,240 --> 00:58:37,950 AUDIENCE: And-- 1288 00:58:37,950 --> 00:58:39,700 PROFESSOR: You've taken this more recently 1289 00:58:39,700 --> 00:58:40,920 than me, haven't you? 1290 00:58:40,920 --> 00:58:42,879 AUDIENCE: [INAUDIBLE] 1291 00:58:42,879 --> 00:58:43,920 PROFESSOR: Stats is good. 1292 00:58:43,920 --> 00:58:45,750 Everybody should take stats. 1293 00:58:45,750 --> 00:58:47,645 I need to take more stats. 1294 00:58:47,645 --> 00:58:48,270 AUDIENCE: And-- 1295 00:58:48,270 --> 00:58:49,897 AUDIENCE: Is that like math? 1296 00:58:49,897 --> 00:58:50,480 AUDIENCE: Yes. 1297 00:58:50,480 --> 00:58:54,342 AUDIENCE: In a very practical manner. 1298 00:58:54,342 --> 00:58:55,980 AUDIENCE: OK, bless you. 1299 00:58:55,980 --> 00:58:59,070 And these one-sample t-tests showed 1300 00:58:59,070 --> 00:59:02,640 that because of a large amount of activation 1301 00:59:02,640 --> 00:59:06,700 in the frontal parietal system in the brain, that 1302 00:59:06,700 --> 00:59:11,220 resulted in both covert and overt shifts of attention. 1303 00:59:11,220 --> 00:59:13,052 AUDIENCE: Right. 1304 00:59:13,052 --> 00:59:18,210 And then they also took [INAUDIBLE].. 1305 00:59:18,210 --> 00:59:22,270 So it looks like that, except they [? add the ?] intersection 1306 00:59:22,270 --> 00:59:23,770 instead of the subtraction. 1307 00:59:23,770 --> 00:59:26,180 And-- 1308 00:59:26,180 --> 00:59:28,600 PROFESSOR: That's this diagram on the bottom of page 104, 1309 00:59:28,600 --> 00:59:29,100 right? 1310 00:59:29,100 --> 00:59:29,794 AUDIENCE: Yeah. 1311 00:59:29,794 --> 00:59:35,760 And then they found that the overlap was very extensive 1312 00:59:35,760 --> 00:59:36,500 for that. 1313 00:59:36,500 --> 00:59:40,890 And that meant that the covert and overt shifts 1314 00:59:40,890 --> 00:59:45,314 of of attention used a lot of the same areas of the brain. 1315 00:59:50,792 --> 00:59:56,270 AUDIENCE: [INAUDIBLE] inspection of [INAUDIBLE] 1316 00:59:56,270 --> 01:00:01,748 suggests that [INAUDIBLE] to be in areas [INAUDIBLE] 1317 01:00:01,748 --> 01:00:03,574 shifts of attention tests is more 1318 01:00:03,574 --> 01:00:06,230 widespread during the overt shifts of attention test 1319 01:00:06,230 --> 01:00:08,720 than that covert shifts of attention test. 1320 01:00:08,720 --> 01:00:12,704 And the results of these statistical tests 1321 01:00:12,704 --> 01:00:14,198 showed no significant differences 1322 01:00:14,198 --> 01:00:17,186 between the amount of [? neutroactivation ?] 1323 01:00:17,186 --> 01:00:20,423 of central conditions or the overt [INAUDIBLE] shifts 1324 01:00:20,423 --> 01:00:21,668 of attention. 1325 01:00:27,230 --> 01:00:28,940 PROFESSOR: OK. 1326 01:00:28,940 --> 01:00:30,860 So, yeah, look at the pretty picture. 1327 01:00:30,860 --> 01:00:33,260 So tell me about this pretty picture, 1328 01:00:33,260 --> 01:00:36,710 somebody in that group. 1329 01:00:36,710 --> 01:00:38,142 What do the different colors mean? 1330 01:00:38,142 --> 01:00:39,290 AUDIENCE: Wait, which one 1331 01:00:39,290 --> 01:00:44,130 PROFESSOR: Page 104, figure 1. 1332 01:00:44,130 --> 01:00:46,506 AUDIENCE: OK, so covert is red. 1333 01:00:46,506 --> 01:00:49,464 And overt is green. 1334 01:00:49,464 --> 01:00:51,929 And both is yellow. 1335 01:00:51,929 --> 01:00:56,859 AUDIENCE: Because red and green [INAUDIBLE].. 1336 01:00:56,859 --> 01:00:58,425 AUDIENCE: [INAUDIBLE] 1337 01:00:58,425 --> 01:01:00,050 PROFESSOR: So yellow is everything that 1338 01:01:00,050 --> 01:01:02,044 was involved in both tasks. 1339 01:01:02,044 --> 01:01:04,460 Green is everything that was involved in the overt but not 1340 01:01:04,460 --> 01:01:05,480 the covert task. 1341 01:01:05,480 --> 01:01:08,450 And red is vice versa, the covert but not the overt task. 1342 01:01:08,450 --> 01:01:10,387 AUDIENCE: Yes. 1343 01:01:10,387 --> 01:01:10,970 PROFESSOR: OK. 1344 01:01:10,970 --> 01:01:12,762 AUDIENCE: And there's more yellow than red. 1345 01:01:12,762 --> 01:01:14,428 PROFESSOR: There's more yellow than red. 1346 01:01:14,428 --> 01:01:16,230 There's more green than red, too, 1347 01:01:16,230 --> 01:01:17,730 which I think is the thing that they 1348 01:01:17,730 --> 01:01:20,260 think is more relevant here. 1349 01:01:20,260 --> 01:01:24,330 So what do these results mean? 1350 01:01:27,770 --> 01:01:30,704 What do they find out in this study? 1351 01:01:30,704 --> 01:01:32,370 So the Discussion section is the section 1352 01:01:32,370 --> 01:01:34,770 in which the researchers basically are saying, 1353 01:01:34,770 --> 01:01:35,970 so this is what we found. 1354 01:01:35,970 --> 01:01:39,390 And this is why it's relevant and what we think it means. 1355 01:01:39,390 --> 01:01:42,300 So any interesting things that people 1356 01:01:42,300 --> 01:01:44,820 found reading through it? 1357 01:01:44,820 --> 01:01:47,700 AUDIENCE: [INAUDIBLE] 1358 01:01:49,344 --> 01:01:50,010 PROFESSOR: Yeah. 1359 01:01:50,010 --> 01:01:53,169 So what is this premotor theory of attention? 1360 01:01:53,169 --> 01:01:55,419 AUDIENCE: It's the fact that, like, [INAUDIBLE] shifts 1361 01:01:55,419 --> 01:01:58,578 of attention activate [INAUDIBLE] 1362 01:01:58,578 --> 01:02:05,868 the brain and that, like, [INAUDIBLE] 1363 01:02:05,868 --> 01:02:08,298 overt shift is the fact that, like, the eyes don't move, 1364 01:02:08,298 --> 01:02:09,280 but it's [INAUDIBLE]. 1365 01:02:09,280 --> 01:02:10,900 PROFESSOR: Right, which is how it gets that premotor 1366 01:02:10,900 --> 01:02:12,441 name, that all of the important stuff 1367 01:02:12,441 --> 01:02:14,770 is happening earlier in processing 1368 01:02:14,770 --> 01:02:18,586 than the actual muscle command to move your eyes. 1369 01:02:18,586 --> 01:02:19,960 AUDIENCE: Didn't they want a test 1370 01:02:19,960 --> 01:02:22,969 on monkeys to not support it because they tested 1371 01:02:22,969 --> 01:02:25,010 the individual neurons, and they found that there 1372 01:02:25,010 --> 01:02:26,446 are separate neurons-- 1373 01:02:26,446 --> 01:02:29,760 the different neuron group-- for-- 1374 01:02:29,760 --> 01:02:33,270 PROFESSOR: That's the study cited in the intro, right? 1375 01:02:33,270 --> 01:02:34,580 No, or was that in here? 1376 01:02:34,580 --> 01:02:35,379 Where was it? 1377 01:02:35,379 --> 01:02:36,670 I'm getting my pieces mixed up. 1378 01:02:36,670 --> 01:02:37,941 No, there it is. 1379 01:02:37,941 --> 01:02:38,940 AUDIENCE: It's on page-- 1380 01:02:38,940 --> 01:02:39,773 PROFESSOR: Page 106. 1381 01:02:39,773 --> 01:02:44,702 AUDIENCE: 106, yeah, which it says it undermines the premotor 1382 01:02:44,702 --> 01:02:46,817 theory of [INAUDIBLE]. 1383 01:02:46,817 --> 01:02:47,525 PROFESSOR: Right. 1384 01:02:47,525 --> 01:02:49,870 So that was a study looking in particularly at neurons 1385 01:02:49,870 --> 01:02:52,594 in the frontal eye fields. 1386 01:02:52,594 --> 01:02:54,010 And they said that, yeah, it looks 1387 01:02:54,010 --> 01:02:56,800 like this area does have one set of neurons that 1388 01:02:56,800 --> 01:02:58,480 responds to overt and one that responds 1389 01:02:58,480 --> 01:03:01,804 to covert shifts of attention. 1390 01:03:01,804 --> 01:03:03,970 They also point out that the superior colliculus has 1391 01:03:03,970 --> 01:03:05,802 neurons that respond to both. 1392 01:03:05,802 --> 01:03:07,510 So there's probably at least a little bit 1393 01:03:07,510 --> 01:03:08,650 of both going on here. 1394 01:03:08,650 --> 01:03:10,540 But the evidence for premotor seems 1395 01:03:10,540 --> 01:03:14,950 to be stronger than the evidence otherwise, 1396 01:03:14,950 --> 01:03:18,070 that if not everything, then at least most of what's 1397 01:03:18,070 --> 01:03:23,510 involved in an intentional shift is the same between the two. 1398 01:03:23,510 --> 01:03:24,490 What else? 1399 01:03:40,180 --> 01:03:42,360 Also, on page 106 they talk about the study 1400 01:03:42,360 --> 01:03:45,820 by Beauchamp, et al., in 2001 who were looking, 1401 01:03:45,820 --> 01:03:53,890 comparing whether overt or covert shifts in attention 1402 01:03:53,890 --> 01:03:56,930 gets more neural activity going on. 1403 01:03:56,930 --> 01:04:00,064 And if you look at the table on page 105, 1404 01:04:00,064 --> 01:04:01,480 you can get a pretty good guess as 1405 01:04:01,480 --> 01:04:05,427 to which one seems to be activating more regions. 1406 01:04:05,427 --> 01:04:07,135 What would you guys say, overt or covert? 1407 01:04:14,232 --> 01:04:15,232 What do you think, Sara? 1408 01:04:15,232 --> 01:04:16,440 AUDIENCE: Overt. 1409 01:04:16,440 --> 01:04:18,981 PROFESSOR: Yeah, so they found that overt shifts of attention 1410 01:04:18,981 --> 01:04:21,180 result in more activation than in covert shifts. 1411 01:04:21,180 --> 01:04:24,050 And there's one other earlier study that had found that. 1412 01:04:24,050 --> 01:04:26,360 But most of the studies so far had actually 1413 01:04:26,360 --> 01:04:28,360 found the opposite, that you get more activation 1414 01:04:28,360 --> 01:04:31,444 in covert shifts than in overt shifts. 1415 01:04:31,444 --> 01:04:33,360 And until they did this one, one of the things 1416 01:04:33,360 --> 01:04:36,150 that was different is that the Beauchamp study in 2001 1417 01:04:36,150 --> 01:04:38,100 that they're comparing themselves to 1418 01:04:38,100 --> 01:04:40,840 had used exogenously driven shifts of attention. 1419 01:04:40,840 --> 01:04:44,916 So they'd actually blinked some kind 1420 01:04:44,916 --> 01:04:46,290 of stimulus out where they wanted 1421 01:04:46,290 --> 01:04:48,450 you to shift your attention to, rather than just directing it 1422 01:04:48,450 --> 01:04:49,950 from where you were already fixated. 1423 01:04:52,400 --> 01:04:56,490 And it was possible that when you're doing endogenous shifts, 1424 01:04:56,490 --> 01:04:58,586 then covert gets more activation, 1425 01:04:58,586 --> 01:05:00,210 and when you're doing exogenous shifts, 1426 01:05:00,210 --> 01:05:01,980 overt gets more activation. 1427 01:05:01,980 --> 01:05:06,750 But then these guys were using endogenous shifts, right? 1428 01:05:06,750 --> 01:05:08,010 It was centrally controlled. 1429 01:05:08,010 --> 01:05:09,960 It was where the arm of the cross 1430 01:05:09,960 --> 01:05:12,450 is pointing, direct your attention in that direction. 1431 01:05:15,600 --> 01:05:19,230 And they still found that overt used more neural activation 1432 01:05:19,230 --> 01:05:21,240 than covert. 1433 01:05:21,240 --> 01:05:23,850 And this is interesting, and they 1434 01:05:23,850 --> 01:05:28,514 don't have a really good answer for it that I could find. 1435 01:05:28,514 --> 01:05:30,680 AUDIENCE: Well, wouldn't it just disprove the theory 1436 01:05:30,680 --> 01:05:35,090 that it has nothing to do with either endogenous or exogenous? 1437 01:05:35,090 --> 01:05:37,320 Because before, weren't they doubting [INAUDIBLE]?? 1438 01:05:37,320 --> 01:05:40,278 Maybe it's because they found opposite results 1439 01:05:40,278 --> 01:05:42,250 for the [INAUDIBLE] exogenous and endogenous. 1440 01:05:42,250 --> 01:05:45,510 They thought it might have been the mechanism that 1441 01:05:45,510 --> 01:05:49,755 caused each stimulus, that [? they think ?] that's why 1442 01:05:49,755 --> 01:05:51,552 the results are [INAUDIBLE]. 1443 01:05:51,552 --> 01:05:54,010 PROFESSOR: Or at least it's a piece of evidence against it. 1444 01:05:54,010 --> 01:05:55,635 Very rarely will you see people saying, 1445 01:05:55,635 --> 01:05:58,240 this one study proves or disprove something. 1446 01:06:02,600 --> 01:06:04,450 So it makes it a lot less likely that that 1447 01:06:04,450 --> 01:06:05,700 is what causes the difference. 1448 01:06:05,700 --> 01:06:07,491 Yeah, but they're still trying to reconcile 1449 01:06:07,491 --> 01:06:11,160 their study in which they're using an endogenously shifted-- 1450 01:06:11,160 --> 01:06:13,936 an endogenous means of shifting attention. 1451 01:06:13,936 --> 01:06:16,060 And they're finding more activation in overt shifts 1452 01:06:16,060 --> 01:06:18,610 with all of these other studies that were also 1453 01:06:18,610 --> 01:06:20,290 using endogenous shifts of attention 1454 01:06:20,290 --> 01:06:24,520 and were finding more activation in covert shifts. 1455 01:06:24,520 --> 01:06:27,310 They consider the idea that it could 1456 01:06:27,310 --> 01:06:31,040 be because they were doing them on two different fMRI runs. 1457 01:06:31,040 --> 01:06:34,210 And for each one, they were doing that activation 1458 01:06:34,210 --> 01:06:38,610 subtraction with the central condition from that task 1459 01:06:38,610 --> 01:06:41,200 so that there may have been some fundamental difference 1460 01:06:41,200 --> 01:06:43,060 in the central attention condition 1461 01:06:43,060 --> 01:06:45,106 for the covert versus the overt. 1462 01:06:45,106 --> 01:06:46,480 And so they would have subtracted 1463 01:06:46,480 --> 01:06:49,430 more in the covert condition than in the overt, 1464 01:06:49,430 --> 01:06:52,840 and thus were left with less activation in total. 1465 01:06:52,840 --> 01:06:56,860 And they consider this, and decide 1466 01:06:56,860 --> 01:07:00,040 that it is unlikely and improbable for a lot 1467 01:07:00,040 --> 01:07:04,415 of reasons, and just shrug and move on. 1468 01:07:04,415 --> 01:07:06,040 They don't have a good answer for this. 1469 01:07:06,040 --> 01:07:08,581 This is one of the things that's interesting about this study 1470 01:07:08,581 --> 01:07:12,360 that they don't really have a good explanation for.