1 00:00:00,060 --> 00:00:02,430 The following content is provided under a Creative 2 00:00:02,430 --> 00:00:03,820 Commons license. 3 00:00:03,820 --> 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 --> 00:00:17,620 at ocw.mit.edu. 8 00:00:21,239 --> 00:00:22,780 ABBY NOYCE: We're going to talk today 9 00:00:22,780 --> 00:00:27,286 a little bit about what we know about language and the brain. 10 00:00:27,286 --> 00:00:29,160 Particularly, today we're going to talk about 11 00:00:29,160 --> 00:00:32,040 what's involved in understanding language, what 12 00:00:32,040 --> 00:00:34,470 we know about how meaning is represented, 13 00:00:34,470 --> 00:00:36,042 all of that good stuff. 14 00:00:36,042 --> 00:00:37,274 Or some of it. 15 00:00:37,274 --> 00:00:38,940 And then tomorrow, we'll talk more about 16 00:00:38,940 --> 00:00:43,339 what's involved in production and when you actually speak. 17 00:00:43,339 --> 00:00:45,630 We're going to see some patient descriptions today that 18 00:00:45,630 --> 00:00:48,930 have patients who have deficits in comprehension 19 00:00:48,930 --> 00:00:51,690 or deficits in production, or both. 20 00:00:51,690 --> 00:00:53,650 They don't always break out evenly. 21 00:00:53,650 --> 00:00:55,191 But the fact that you'll see patients 22 00:00:55,191 --> 00:00:57,690 who can understand but not produce, or vice versa, 23 00:00:57,690 --> 00:01:00,410 seems to imply that there is at least some distinction there. 24 00:01:11,954 --> 00:01:12,549 Good job. 25 00:01:12,549 --> 00:01:14,090 And of course, not on the one I want. 26 00:01:14,090 --> 00:01:14,589 There we go. 27 00:01:17,387 --> 00:01:18,970 The first thing I want to talk about-- 28 00:01:18,970 --> 00:01:20,600 hey, look, that's where we were-- 29 00:01:20,600 --> 00:01:21,870 is levels of-- 30 00:01:21,870 --> 00:01:22,370 OK. 31 00:01:22,370 --> 00:01:23,786 When we're talking about language, 32 00:01:23,786 --> 00:01:26,420 you can think about language on a lot of different levels. 33 00:01:26,420 --> 00:01:29,480 You guys are listening to me, and you're hopefully parsing 34 00:01:29,480 --> 00:01:31,220 it getting something out of it. 35 00:01:31,220 --> 00:01:34,040 And this level of thinking about the actual meaning 36 00:01:34,040 --> 00:01:38,870 that you're extracting from a group of sentences 37 00:01:38,870 --> 00:01:40,640 is called discourse level. 38 00:01:40,640 --> 00:01:45,260 So the fact that you are hopefully getting ideas 39 00:01:45,260 --> 00:01:47,960 triggered by the sweet sound of my voice. 40 00:01:51,520 --> 00:01:53,220 [INAUDIBLE] 41 00:01:53,220 --> 00:01:58,600 Working further down, if we look at a specific sentence-- 42 00:01:58,600 --> 00:01:59,685 so for example-- 43 00:02:02,664 --> 00:02:03,580 let's pick a sentence. 44 00:02:06,968 --> 00:02:23,760 Peter-- all right. 45 00:02:23,760 --> 00:02:25,740 There's a moderately complex sentence. 46 00:02:25,740 --> 00:02:27,757 It doesn't have any subordinate clauses, 47 00:02:27,757 --> 00:02:30,090 but it's got this lovely prepositional phrase at the end 48 00:02:30,090 --> 00:02:32,730 there. 49 00:02:32,730 --> 00:02:35,070 So how is this thing put together? 50 00:02:37,710 --> 00:02:41,130 When people first started trying to make sense out 51 00:02:41,130 --> 00:02:44,130 of how language structure works-- 52 00:02:44,130 --> 00:02:46,805 for a while, people thought it was all about word order. 53 00:02:46,805 --> 00:02:48,180 If you've got a noun, then you're 54 00:02:48,180 --> 00:02:50,170 most likely to have your next thing be a verb. 55 00:02:50,170 --> 00:02:51,878 If your next thing is a verb, then you'll 56 00:02:51,878 --> 00:02:55,920 most likely to get another noun, or maybe a preposition. 57 00:02:55,920 --> 00:02:59,400 And word order falls apart pretty hard 58 00:02:59,400 --> 00:03:05,475 when you start trying to structure things like, 59 00:03:05,475 --> 00:03:08,160 that boy you saw yesterday put the poodle in the closet, 60 00:03:08,160 --> 00:03:12,142 or that boy you saw yesterday who was wearing a blue shirt 61 00:03:12,142 --> 00:03:13,350 put the poodle in the closet. 62 00:03:13,350 --> 00:03:16,136 And the word order stops being so useful a tool 63 00:03:16,136 --> 00:03:18,510 for figuring out what the different parts of the sentence 64 00:03:18,510 --> 00:03:20,740 are. 65 00:03:20,740 --> 00:03:29,050 Syntactitians break this down into saying that-- let's see. 66 00:03:29,050 --> 00:03:33,650 We've got a sentence here which is 67 00:03:33,650 --> 00:03:39,136 composed of a noun phrase and a big honking verb phrase. 68 00:03:39,136 --> 00:03:41,300 So this is our noun. 69 00:03:41,300 --> 00:03:44,830 All of this is a verb phrase. 70 00:03:44,830 --> 00:03:46,550 Our noun phrase is a really basic one. 71 00:03:46,550 --> 00:03:51,960 It's just got one noun in it, Peter. 72 00:03:51,960 --> 00:03:54,910 And this verb phrase has a couple of parts. 73 00:03:57,535 --> 00:03:58,910 If you were going to break that-- 74 00:03:58,910 --> 00:04:03,860 if you were going to break put the poodle in the closet down 75 00:04:03,860 --> 00:04:06,440 into the important parts, where would you break it up? 76 00:04:06,440 --> 00:04:07,815 You probably wouldn't break it up 77 00:04:07,815 --> 00:04:09,560 between the and poodle, for example. 78 00:04:09,560 --> 00:04:11,090 That would be pretty silly. 79 00:04:11,090 --> 00:04:12,256 Where might you break it up? 80 00:04:15,710 --> 00:04:22,731 AUDIENCE: The verb place, the noun, and the-- 81 00:04:22,731 --> 00:04:23,730 ABBY NOYCE: Preposition. 82 00:04:23,730 --> 00:04:24,450 AUDIENCE: Yeah. 83 00:04:24,450 --> 00:04:25,200 ABBY NOYCE: Yeah. 84 00:04:25,200 --> 00:04:30,600 So you'd say that this phrase has a verb-- 85 00:04:30,600 --> 00:04:37,300 quite right-- and a noun phrase, and this big, long 86 00:04:37,300 --> 00:04:38,190 prepositional phrase. 87 00:04:38,190 --> 00:04:39,530 So put. 88 00:04:39,530 --> 00:04:42,270 This noun phrase has a determiner 89 00:04:42,270 --> 00:04:44,530 and a noun, the poodle. 90 00:04:47,150 --> 00:04:51,180 And then this, in turn, has the preposition in, 91 00:04:51,180 --> 00:04:55,100 and another noun phrase. 92 00:04:55,100 --> 00:04:57,100 In the closet. 93 00:05:00,800 --> 00:05:01,300 OK. 94 00:05:03,990 --> 00:05:06,750 People who study syntax-- syntactitians-- 95 00:05:06,750 --> 00:05:08,880 think that what's going on when you parse 96 00:05:08,880 --> 00:05:12,280 sentences is that you go-- 97 00:05:12,280 --> 00:05:16,380 do this transition from the word order to putting together some 98 00:05:16,380 --> 00:05:17,900 model of the underlying structure-- 99 00:05:17,900 --> 00:05:19,650 the structure of which pieces go together, 100 00:05:19,650 --> 00:05:24,120 which pieces are modifying which other pieces-- 101 00:05:24,120 --> 00:05:25,680 in order to extract meaning from it. 102 00:05:28,200 --> 00:05:29,640 You can also talk-- 103 00:05:29,640 --> 00:05:32,400 on lower levels you can talk about language, 104 00:05:32,400 --> 00:05:34,830 you can talk about the individual words. 105 00:05:34,830 --> 00:05:39,540 You can talk about the morphemes that make up the words. 106 00:05:39,540 --> 00:05:43,250 So in a sentence like-- 107 00:05:43,250 --> 00:05:44,100 oh, let's say-- 108 00:05:51,320 --> 00:05:54,260 Josh. 109 00:05:54,260 --> 00:05:56,910 Picking names at random. 110 00:05:56,910 --> 00:06:00,530 To [INAUDIBLE]. 111 00:06:04,490 --> 00:06:05,702 AUDIENCE: Squeaking. 112 00:06:05,702 --> 00:06:06,410 ABBY NOYCE: Yeah. 113 00:06:06,410 --> 00:06:08,118 There's apparently furniture on the move. 114 00:06:10,606 --> 00:06:11,980 CDs. 115 00:06:11,980 --> 00:06:12,510 All right. 116 00:06:12,510 --> 00:06:15,690 This is a sentence that has five words, 117 00:06:15,690 --> 00:06:17,520 but it's got more morphemes than that. 118 00:06:17,520 --> 00:06:22,290 So Josh is-- a name like that is a single-morpheme word. 119 00:06:22,290 --> 00:06:24,750 It doesn't break down into smaller components. 120 00:06:24,750 --> 00:06:28,530 But for example, this one has a root word and then 121 00:06:28,530 --> 00:06:32,580 this -ed ending that means past tense. 122 00:06:32,580 --> 00:06:34,380 And of course, that's an ending that's 123 00:06:34,380 --> 00:06:36,504 still the same way in all kinds of different words, 124 00:06:36,504 --> 00:06:38,490 but there's a lot of different pronunciations 125 00:06:38,490 --> 00:06:41,850 of that past tense marker, depending on what kind of word. 126 00:06:41,850 --> 00:06:44,159 The past tense marker in washed is 127 00:06:44,159 --> 00:06:46,200 different than the past tense marker in listened, 128 00:06:46,200 --> 00:06:49,140 which is different than the past tense marker in hunted, 129 00:06:49,140 --> 00:06:52,800 in terms of what they actually sound like. 130 00:06:52,800 --> 00:06:55,419 But that's one word that's made up of two morphemes. 131 00:06:55,419 --> 00:06:56,460 There's another one here. 132 00:06:56,460 --> 00:06:58,530 There's Joe, and then there's this apostrophe S 133 00:06:58,530 --> 00:07:00,670 that's marking possessive. 134 00:07:00,670 --> 00:07:03,510 There's CDs, and this S that marks the plural. 135 00:07:03,510 --> 00:07:08,470 Morphemes can be smaller than words, 136 00:07:08,470 --> 00:07:09,790 but they still carry meaning. 137 00:07:09,790 --> 00:07:12,750 Sometimes morphemes can't stand alone. 138 00:07:12,750 --> 00:07:14,880 They can only exist tagged on to something, 139 00:07:14,880 --> 00:07:18,810 like those past-tense markers or those plurality markers. 140 00:07:18,810 --> 00:07:20,730 And then morphemes are, of course, 141 00:07:20,730 --> 00:07:23,700 themselves made up of phonemes, which are the sounds that 142 00:07:23,700 --> 00:07:27,252 make up a language, which you guys talked about with Lisa 143 00:07:27,252 --> 00:07:27,960 yesterday, right? 144 00:07:36,120 --> 00:07:39,000 Different languages have different phonemes. 145 00:07:39,000 --> 00:07:41,580 English has a phoneme set of about 40 sounds. 146 00:07:41,580 --> 00:07:44,070 Hawaiian has a lot less. 147 00:07:44,070 --> 00:07:45,450 Some languages have more. 148 00:07:45,450 --> 00:07:48,510 English is kind of on the high end. 149 00:07:48,510 --> 00:07:49,350 Cool. 150 00:07:49,350 --> 00:07:50,850 These are the things that go into it 151 00:07:50,850 --> 00:07:52,530 that you have to be able to-- 152 00:07:52,530 --> 00:07:55,322 when you are parsing language, when you're hearing something 153 00:07:55,322 --> 00:07:57,030 that people say and make sense out of it, 154 00:07:57,030 --> 00:07:59,257 you've got to be dealing with it on all 155 00:07:59,257 --> 00:08:00,340 of these different levels. 156 00:08:00,340 --> 00:08:02,506 You've got to be trying to make sense out of things. 157 00:08:02,506 --> 00:08:07,260 And if your ability to pull out the phonemes fails, 158 00:08:07,260 --> 00:08:09,270 you're not going to get anything higher up. 159 00:08:09,270 --> 00:08:11,395 It's like listening to somebody with a really thick 160 00:08:11,395 --> 00:08:12,360 foreign accent. 161 00:08:12,360 --> 00:08:13,404 You almost have to-- 162 00:08:13,404 --> 00:08:14,820 when you first start listing them, 163 00:08:14,820 --> 00:08:16,980 you have a really hard time understanding them, 164 00:08:16,980 --> 00:08:18,938 and then if you listen to them talk for a while 165 00:08:18,938 --> 00:08:20,880 you start being better at following it 166 00:08:20,880 --> 00:08:23,615 and better at figuring out what they're saying. 167 00:08:23,615 --> 00:08:24,990 And this is partly because you're 168 00:08:24,990 --> 00:08:26,364 kind of resetting where you think 169 00:08:26,364 --> 00:08:28,860 the phonemes-- what you think the space for the phonemes 170 00:08:28,860 --> 00:08:29,520 can be. 171 00:08:29,520 --> 00:08:34,140 Your idea of what sounds can fit a particular English phoneme 172 00:08:34,140 --> 00:08:36,250 expand to include what this person is saying, 173 00:08:36,250 --> 00:08:37,969 and once you can pull out the phonemes, 174 00:08:37,969 --> 00:08:39,510 then you can build up the rest of it. 175 00:08:42,240 --> 00:08:43,450 All right. 176 00:08:43,450 --> 00:08:47,080 A couple of classic examples of people whose language 177 00:08:47,080 --> 00:08:48,535 processing skills are broken. 178 00:08:51,370 --> 00:08:55,675 Paul Broca, in the 1850s in Europe, 179 00:08:55,675 --> 00:08:58,180 because all the cool stuff here happened in Europe. 180 00:08:58,180 --> 00:09:01,222 We got Phineas Gage, but they got the language kids. 181 00:09:01,222 --> 00:09:02,430 AUDIENCE: Who's Phineas Gage? 182 00:09:02,430 --> 00:09:04,305 ABBY NOYCE: Phineas Gage had a railroad spike 183 00:09:04,305 --> 00:09:06,520 goes through his frontal lobe due to an accident. 184 00:09:06,520 --> 00:09:07,270 AUDIENCE: Oh, wow. 185 00:09:07,270 --> 00:09:08,151 ABBY NOYCE: Yes. 186 00:09:08,151 --> 00:09:11,020 AUDIENCE: And then he became all nasty. 187 00:09:11,020 --> 00:09:13,270 ABBY NOYCE: Yeah, and then he became a really not nice 188 00:09:13,270 --> 00:09:14,620 person. 189 00:09:14,620 --> 00:09:16,000 All right. 190 00:09:16,000 --> 00:09:22,630 Broca presented to the [FRENCH] an account 191 00:09:22,630 --> 00:09:25,210 of a patient at the hospital there in Paris, 192 00:09:25,210 --> 00:09:28,180 who had been there for about 15 years 193 00:09:28,180 --> 00:09:31,490 after having had an attack. 194 00:09:31,490 --> 00:09:36,380 This is before times like stroke were really reasonably common. 195 00:09:36,380 --> 00:09:38,920 And this was a patient, Monsieur Leborgne, 196 00:09:38,920 --> 00:09:42,921 who seemed to display more or less normal intelligence, who 197 00:09:42,921 --> 00:09:43,420 could-- 198 00:09:43,420 --> 00:09:46,480 if you asked him to do things, he could do them. 199 00:09:46,480 --> 00:09:47,520 If you asked him-- 200 00:09:51,369 --> 00:09:52,285 but he couldn't speak. 201 00:09:52,285 --> 00:09:54,094 If you asked him to point to the-- 202 00:09:54,094 --> 00:09:56,260 point to the example of a case in this, for example. 203 00:09:56,260 --> 00:09:58,551 He could tell you which of these pictures was a banana, 204 00:09:58,551 --> 00:10:00,374 or which one was a shoe, or what have you. 205 00:10:00,374 --> 00:10:01,290 But he couldn't speak. 206 00:10:01,290 --> 00:10:06,000 All he would ever say was tom tom tom, which 207 00:10:06,000 --> 00:10:08,010 is just a nonsense syllable. 208 00:10:08,010 --> 00:10:10,260 And what Broca they discovered in 1851, 209 00:10:10,260 --> 00:10:13,214 after Monsieur Leborgne passed on, 210 00:10:13,214 --> 00:10:15,630 is that-- he autopsied him, looked at his brain, and said, 211 00:10:15,630 --> 00:10:18,110 OK, can we point to a spot that's 212 00:10:18,110 --> 00:10:19,860 what's wrong with this guy, and discovered 213 00:10:19,860 --> 00:10:25,170 that he had a lesion in the left frontal lobes. 214 00:10:25,170 --> 00:10:27,420 The neurons in that area had died, so there was this-- 215 00:10:27,420 --> 00:10:32,670 just a fluid-filled space in his brain about chicken egg-sized. 216 00:10:32,670 --> 00:10:36,510 This is a big hole in this guy's frontal lobes. 217 00:10:36,510 --> 00:10:39,750 On the left-hand side, right in here. 218 00:10:39,750 --> 00:10:44,576 You can see Broca's area marked on the diagram there. 219 00:10:44,576 --> 00:10:47,610 And Broca looked at this and said, OK, so we've 220 00:10:47,610 --> 00:10:49,350 got this patient-- 221 00:10:49,350 --> 00:10:50,880 and he started collecting cases. 222 00:10:50,880 --> 00:10:54,090 He pretty rapidly started getting other accounts 223 00:10:54,090 --> 00:10:56,970 of patients who were similarly afflicted, 224 00:10:56,970 --> 00:10:59,844 who are neither paralyzed nor idiots. 225 00:10:59,844 --> 00:11:01,260 And he decided they needed their-- 226 00:11:01,260 --> 00:11:03,570 to have their own term for this. 227 00:11:03,570 --> 00:11:05,680 He coined the term aphasia. 228 00:11:05,680 --> 00:11:07,460 Remember, we've had our agnosias, 229 00:11:07,460 --> 00:11:10,560 and we will talk about alexias and agraphias. 230 00:11:10,560 --> 00:11:13,890 This corner is all about the Greek roots. 231 00:11:13,890 --> 00:11:16,260 So aphasia, this inability to speak, 232 00:11:16,260 --> 00:11:18,150 to handle language-- spoken language 233 00:11:18,150 --> 00:11:19,685 in particular-- correctly. 234 00:11:24,809 --> 00:11:26,850 Broca started collecting cases of these patients, 235 00:11:26,850 --> 00:11:32,260 and one of the things he found was that all of his cases 236 00:11:32,260 --> 00:11:35,374 had were lesions on the left-hand side. 237 00:11:35,374 --> 00:11:37,540 All of these were patients with lesions on the left. 238 00:11:37,540 --> 00:11:39,100 None of them were patients with lesions on the right. 239 00:11:39,100 --> 00:11:40,720 And Broca says, well, everything we 240 00:11:40,720 --> 00:11:42,550 know about the brain so far seems 241 00:11:42,550 --> 00:11:47,199 to point to it being really symmetrical, 242 00:11:47,199 --> 00:11:48,740 so this is probably just coincidence. 243 00:11:48,740 --> 00:11:52,250 I'm not convinced that the left-hand side of this 244 00:11:52,250 --> 00:11:52,970 means anything. 245 00:11:52,970 --> 00:11:55,700 At this point, Broca probably had about eight or 10 246 00:11:55,700 --> 00:11:58,520 of these cases that he was documenting. 247 00:11:58,520 --> 00:12:02,520 Over the next five or so years, Broca 248 00:12:02,520 --> 00:12:05,570 kept collecting cases he got up to about 20. 249 00:12:05,570 --> 00:12:08,540 He found a doc-- 250 00:12:08,540 --> 00:12:15,560 the [FRENCH] acquired a document from a physician working 251 00:12:15,560 --> 00:12:18,560 in southern France, who had also been collecting such cases 252 00:12:18,560 --> 00:12:22,640 and had also documented 40 of them, 253 00:12:22,640 --> 00:12:25,730 also all with lesions on the left hand side. 254 00:12:25,730 --> 00:12:28,640 And Broca finally found his clinching piece of evidence 255 00:12:28,640 --> 00:12:31,500 that this really is a left-hand side, 256 00:12:31,500 --> 00:12:33,920 left-hemisphere phenomenon when he 257 00:12:33,920 --> 00:12:36,476 got a case of a patient with an equivalent lesion 258 00:12:36,476 --> 00:12:37,850 on the right-hand side who didn't 259 00:12:37,850 --> 00:12:40,160 show any linguistic deficit. 260 00:12:40,160 --> 00:12:45,650 So Broca says, OK, clearly this is something in which 261 00:12:45,650 --> 00:12:47,120 the brain is not symmetric. 262 00:12:47,120 --> 00:12:49,520 The left-hand side is more important for language 263 00:12:49,520 --> 00:12:53,420 and for speech than the right-hand side is. 264 00:12:53,420 --> 00:12:56,210 And what Broca didn't really notice 265 00:12:56,210 --> 00:12:58,010 about this particular area-- 266 00:12:58,010 --> 00:13:00,680 nobody really noticed until probably close to 100 years 267 00:13:00,680 --> 00:13:05,120 after Broca was working on it-- is that patient Broca's aphasia 268 00:13:05,120 --> 00:13:07,750 speak slowly and hesitantly. 269 00:13:07,750 --> 00:13:11,210 They'll have a hard time finding words. 270 00:13:11,210 --> 00:13:14,220 But also, the words they use are mostly nouns, 271 00:13:14,220 --> 00:13:16,190 with the occasional verb, and they 272 00:13:16,190 --> 00:13:19,640 don't have any of the syntax markers thrown 273 00:13:19,640 --> 00:13:23,570 in there-- all of the stuff that signposts a sentence, 274 00:13:23,570 --> 00:13:25,790 tells you how the words are related 275 00:13:25,790 --> 00:13:28,050 to each other, what's a subordinate clause, 276 00:13:28,050 --> 00:13:28,550 any of that. 277 00:13:28,550 --> 00:13:31,580 They don't have any of this stuff. 278 00:13:31,580 --> 00:13:35,300 Here's an example of an interview 279 00:13:35,300 --> 00:13:40,970 with a patient with Broca's aphasia, a former Coast Guard 280 00:13:40,970 --> 00:13:46,970 guy who had a stroke that seems to have wiped out Broca's area. 281 00:14:06,010 --> 00:14:10,124 You can see the hesitance, the searching for words, 282 00:14:10,124 --> 00:14:12,040 the difficulty in apparently finding the words 283 00:14:12,040 --> 00:14:16,840 he wants, but also that the sentence 284 00:14:16,840 --> 00:14:20,590 structure you'd expect to see in a fluent English speaker-- 285 00:14:20,590 --> 00:14:21,430 it's just not there. 286 00:14:24,470 --> 00:14:27,790 So you just get things like, were you in the Coast Guard? 287 00:14:27,790 --> 00:14:28,630 No. 288 00:14:28,630 --> 00:14:29,920 Yes, yes. 289 00:14:29,920 --> 00:14:30,550 Ship. 290 00:14:30,550 --> 00:14:32,590 Massachusetts Coast Guard. 291 00:14:32,590 --> 00:14:34,262 Years. 292 00:14:34,262 --> 00:14:35,845 That is not a fluent English sentence. 293 00:14:42,100 --> 00:14:43,600 So Broca was documenting this stuff. 294 00:14:43,600 --> 00:14:47,440 Broca was determining that we speak-- 295 00:14:47,440 --> 00:14:49,810 that speech depends on the left hemisphere, especially 296 00:14:49,810 --> 00:14:52,000 the left frontal lobes. 297 00:14:52,000 --> 00:14:55,960 And about 15 years later, along comes 298 00:14:55,960 --> 00:14:58,780 this other guy, Carl Wernicke. 299 00:14:58,780 --> 00:15:02,860 Wernicke was also documenting aphasic patients, 300 00:15:02,860 --> 00:15:07,030 but he had managed-- he was collecting cases of patients 301 00:15:07,030 --> 00:15:10,120 whose aphasia was very different from Broca's. 302 00:15:10,120 --> 00:15:12,100 These patients spoke very fluently, 303 00:15:12,100 --> 00:15:14,470 unlike Broca's patients, who would stammer 304 00:15:14,470 --> 00:15:16,450 and pause and search for words. 305 00:15:16,450 --> 00:15:18,490 Words flowed freely with these patients. 306 00:15:18,490 --> 00:15:21,700 They could just go on and talk. 307 00:15:21,700 --> 00:15:23,620 But the words were not the right words, 308 00:15:23,620 --> 00:15:28,270 or they weren't even words, and so patients made no sense. 309 00:15:28,270 --> 00:15:30,010 One of the things Wernicke points out 310 00:15:30,010 --> 00:15:33,550 is that these patients often were diagnosed as confused, 311 00:15:33,550 --> 00:15:36,970 and it took some doing to show that they 312 00:15:36,970 --> 00:15:40,090 had this specific deficit in language 313 00:15:40,090 --> 00:15:45,830 versus simply just being out of touch with the world. 314 00:15:45,830 --> 00:15:49,020 Wernicke's patients also have difficulty with comprehension. 315 00:15:49,020 --> 00:15:53,775 Remember that Broca's aphasia patients can comprehend 316 00:15:53,775 --> 00:15:54,775 at least basic language. 317 00:15:57,520 --> 00:16:00,950 Wernicke's aphasia patients can't. 318 00:16:00,950 --> 00:16:06,760 Wernicke showed that these guys, instead of having lesions up 319 00:16:06,760 --> 00:16:10,640 here in Broca's area, in the lower lateral frontal lobe, 320 00:16:10,640 --> 00:16:14,330 Wernicke's area is somewhere-- depending on who you ask-- 321 00:16:14,330 --> 00:16:16,280 down here in the temporal lobe. 322 00:16:16,280 --> 00:16:19,690 Remember, this is kind area that does auditory processing. 323 00:16:19,690 --> 00:16:22,750 And what Wernicke also pointed out 324 00:16:22,750 --> 00:16:26,680 is that Broca's area up here is right in with this strip 325 00:16:26,680 --> 00:16:27,820 of motor cortex. 326 00:16:27,820 --> 00:16:31,360 The primary motor cortex is what, of course, controls, 327 00:16:31,360 --> 00:16:32,545 plans, structures actions. 328 00:16:35,890 --> 00:16:39,760 So Wernicke theorized that what's happening 329 00:16:39,760 --> 00:16:43,570 in Broca's patients is that the motor representations 330 00:16:43,570 --> 00:16:46,420 of words-- the kind of connection between your idea 331 00:16:46,420 --> 00:16:50,620 of a word and the lip and tongue and mouth movements that are 332 00:16:50,620 --> 00:16:53,500 necessary to produce it-- 333 00:16:53,500 --> 00:16:55,870 that that second motor representation 334 00:16:55,870 --> 00:16:57,580 was broken in some way, so patients 335 00:16:57,580 --> 00:16:59,830 had a hard time producing words. 336 00:16:59,830 --> 00:17:03,150 And he said that for patients with Wernicke's aphasia 337 00:17:03,150 --> 00:17:07,520 the auditory representation of words was broken, 338 00:17:07,520 --> 00:17:10,150 and so they weren't really connected to meaning anymore. 339 00:17:10,150 --> 00:17:13,396 It was just the sounds. 340 00:17:13,396 --> 00:17:14,000 All right. 341 00:17:14,000 --> 00:17:15,250 Want to see a patient with it? 342 00:17:15,250 --> 00:17:18,440 Here's another example. 343 00:17:18,440 --> 00:17:20,190 This is a patient with Wernicke's aphasia. 344 00:17:20,190 --> 00:17:23,319 Both these are from the middle of the 20th century. 345 00:17:23,319 --> 00:17:26,470 Howard Gardner collected-- wrote a lovely book 346 00:17:26,470 --> 00:17:28,650 on language processing and collected examples. 347 00:17:34,160 --> 00:17:37,040 You can see this is a patient who just kind of starts 348 00:17:37,040 --> 00:17:37,572 talking. 349 00:17:37,572 --> 00:17:39,530 And at first-- what brings you to the hospital? 350 00:17:39,530 --> 00:17:40,280 Boy, I'm sweating. 351 00:17:40,280 --> 00:17:42,710 I'm awful nervous. 352 00:17:42,710 --> 00:17:45,590 At first, this sounds like it might be leading 353 00:17:45,590 --> 00:17:48,350 into a reasonable answer. 354 00:17:48,350 --> 00:17:50,544 And then it just goes. 355 00:17:50,544 --> 00:17:51,710 I'm awful nervous, you know. 356 00:17:51,710 --> 00:17:53,100 Once in a while, I get caught up. 357 00:17:53,100 --> 00:17:54,391 I can't mention the terror poy. 358 00:17:54,391 --> 00:17:55,640 A month ago, quite a little. 359 00:17:55,640 --> 00:17:56,540 I've done a lot well. 360 00:17:56,540 --> 00:17:57,754 I impose a lot well. 361 00:17:57,754 --> 00:17:59,420 On the other hand, you know what I mean? 362 00:17:59,420 --> 00:18:01,461 I have to run around, look at over Trevin and all 363 00:18:01,461 --> 00:18:02,392 that sort of stuff. 364 00:18:02,392 --> 00:18:04,100 And what Gardner says is-- at this point, 365 00:18:04,100 --> 00:18:06,200 he physically interrupted the patient, 366 00:18:06,200 --> 00:18:08,909 put a hand on his shoulder and said, thank you, 367 00:18:08,909 --> 00:18:11,450 because he seems to have had the impression that this patient 368 00:18:11,450 --> 00:18:14,861 would have just kept talking if he hadn't interfered. 369 00:18:19,656 --> 00:18:21,030 Again, there's some stuff in here 370 00:18:21,030 --> 00:18:23,700 that I'm pretty sure is not actually real words. 371 00:18:23,700 --> 00:18:26,860 The whole thing doesn't really make sense. 372 00:18:26,860 --> 00:18:30,810 But clearly, production or syntax, stringing words 373 00:18:30,810 --> 00:18:32,640 together into sophisticated sentences, 374 00:18:32,640 --> 00:18:36,660 don't seem to be difficult for these patients in the way 375 00:18:36,660 --> 00:18:39,040 that they are for Broca's patients. 376 00:18:39,040 --> 00:18:40,977 AUDIENCE: [INAUDIBLE] recorded this? 377 00:18:40,977 --> 00:18:42,560 ABBY NOYCE: I assume he recorded this. 378 00:18:42,560 --> 00:18:46,997 AUDIENCE: That would be [INAUDIBLE] shorthand. 379 00:18:46,997 --> 00:18:48,997 ABBY NOYCE: Some extraordinarily good shorthand. 380 00:18:48,997 --> 00:18:50,816 AUDIENCE: [INAUDIBLE] 381 00:18:50,816 --> 00:18:51,482 ABBY NOYCE: Yes. 382 00:19:23,787 --> 00:19:24,781 All right. 383 00:19:29,250 --> 00:19:32,070 This was in the middle of the 1800s, 384 00:19:32,070 --> 00:19:35,470 and this all went out of fashion in the early 20th century. 385 00:19:35,470 --> 00:19:37,030 This was behaviorism, and you just 386 00:19:37,030 --> 00:19:39,180 stalled on all of these guys. 387 00:19:39,180 --> 00:19:43,620 And there was some reasonable, some not reasonable criticism 388 00:19:43,620 --> 00:19:48,150 of what the diagram makers of the 19th century had done, 389 00:19:48,150 --> 00:19:50,610 the biggest of which was that Wernicke's-- 390 00:19:50,610 --> 00:19:53,850 the most valid of which was Wernicke's descriptions 391 00:19:53,850 --> 00:19:55,930 of this idea that there were two speech centers, 392 00:19:55,930 --> 00:19:57,690 and disrupting one led to one aphasia 393 00:19:57,690 --> 00:20:00,960 and disrupting the other led to the other aphasia, 394 00:20:00,960 --> 00:20:05,520 was that it didn't fit well with the symptoms 395 00:20:05,520 --> 00:20:09,037 that patients were actually suffering. 396 00:20:09,037 --> 00:20:11,370 For example, that thing-- remember that Broca's patients 397 00:20:11,370 --> 00:20:13,044 have difficulty with syntax. 398 00:20:13,044 --> 00:20:14,460 They don't have syntactical words. 399 00:20:14,460 --> 00:20:16,410 And what people showed in the '70s 400 00:20:16,410 --> 00:20:19,650 is that this also means that they have difficulty with oddly 401 00:20:19,650 --> 00:20:21,150 constructed sentences. 402 00:20:21,150 --> 00:20:23,275 So their of word comprehension is good. 403 00:20:23,275 --> 00:20:26,440 Their syntactical comprehension is not so good. 404 00:20:26,440 --> 00:20:29,450 So if you gave-- 405 00:20:29,450 --> 00:20:36,300 if you said to a Broca's aphasia patient-- 406 00:20:36,300 --> 00:20:42,420 oh, I don't know, the difference between the cat 407 00:20:42,420 --> 00:20:45,870 caught the mouse and the cat was caught by the mouse-- 408 00:20:45,870 --> 00:20:48,000 these are two very different situations, 409 00:20:48,000 --> 00:20:50,570 but Broca's aphasia patients won't parse that. 410 00:20:50,570 --> 00:20:52,470 They'll be able to pull out-- 411 00:20:52,470 --> 00:20:54,300 it looks like-- the two nouns and the verb, 412 00:20:54,300 --> 00:20:55,800 and they know what order they're in, 413 00:20:55,800 --> 00:21:00,169 and they'll guess what the most likely construct is. 414 00:21:00,169 --> 00:21:01,710 So they here cat and mouse and catch. 415 00:21:01,710 --> 00:21:03,876 OK, clearly this should be the cat caught the mouse. 416 00:21:03,876 --> 00:21:06,530 It is much less likely that a mouse caught a cat 417 00:21:06,530 --> 00:21:09,280 in any kind of useful way. 418 00:21:09,280 --> 00:21:12,120 So they get tripped up by things like passive voice, things 419 00:21:12,120 --> 00:21:14,484 like complicated sentence structures. 420 00:21:18,280 --> 00:21:21,070 Another criticism of these diagram makers 421 00:21:21,070 --> 00:21:24,604 was that very few cases fit neatly into categories 422 00:21:24,604 --> 00:21:26,020 where patients would have just one 423 00:21:26,020 --> 00:21:28,480 set of deficits and no others. 424 00:21:28,480 --> 00:21:30,730 On the other hand, since most of these cases come 425 00:21:30,730 --> 00:21:36,207 from trauma of some sort, or stroke, or abcesses, 426 00:21:36,207 --> 00:21:38,290 very rarely are you going to have one little brain 427 00:21:38,290 --> 00:21:39,081 center knocked out. 428 00:21:39,081 --> 00:21:42,340 You're likely to see damage spreading into adjacent areas. 429 00:21:42,340 --> 00:21:43,840 And there was a lot of arguing about 430 00:21:43,840 --> 00:21:47,295 whether the anatomical evidence was as good as the diagram 431 00:21:47,295 --> 00:21:49,420 makers-- as Broca and Wernicke and their colleagues 432 00:21:49,420 --> 00:21:50,610 thought it was. 433 00:21:50,610 --> 00:21:54,130 So for a while this was out of fashion. 434 00:21:54,130 --> 00:21:56,530 And then, in the mid-20th century-- 435 00:21:56,530 --> 00:21:57,030 come on. 436 00:22:02,010 --> 00:22:02,510 No? 437 00:22:02,510 --> 00:22:04,490 Did my PowerPoint just die? 438 00:22:04,490 --> 00:22:05,030 Apparently. 439 00:22:05,030 --> 00:22:08,780 So then in the mid-20th century, this-- 440 00:22:11,449 --> 00:22:13,490 in the mid 20th century, this started coming back 441 00:22:13,490 --> 00:22:15,860 in for a number of reasons. 442 00:22:15,860 --> 00:22:18,320 This is the information-- 443 00:22:18,320 --> 00:22:18,980 processing. 444 00:22:18,980 --> 00:22:19,970 OK, good. 445 00:22:19,970 --> 00:22:21,470 We are responding to keyboard again. 446 00:22:21,470 --> 00:22:24,901 Maybe, sort of, I think. 447 00:22:24,901 --> 00:22:25,400 Yes. 448 00:22:25,400 --> 00:22:27,880 Now we've gone back to having a window. 449 00:22:27,880 --> 00:22:31,017 No, and it's not where I can reach it. 450 00:22:31,017 --> 00:22:31,991 Is it F7? 451 00:22:36,380 --> 00:22:37,910 No, that was not what I wanted. 452 00:22:37,910 --> 00:22:39,368 What's the keyboard command for it? 453 00:22:39,368 --> 00:22:40,171 Anyone know it? 454 00:22:40,171 --> 00:22:40,670 No. 455 00:22:44,420 --> 00:22:45,460 F12 is doing something. 456 00:22:45,460 --> 00:22:47,088 Nope. 457 00:22:47,088 --> 00:22:48,600 AUDIENCE: Just click on the bottom-- 458 00:22:48,600 --> 00:22:50,100 ABBY NOYCE: The problem is that it's 459 00:22:50,100 --> 00:22:52,430 a piece that's off the screen, so now it's like waaah. 460 00:22:52,430 --> 00:22:52,930 Come on. 461 00:22:55,760 --> 00:22:57,090 Oh, you fail. 462 00:22:57,090 --> 00:22:58,670 Fail utterly. 463 00:22:58,670 --> 00:23:00,460 We'll go with that. 464 00:23:00,460 --> 00:23:01,120 F5. 465 00:23:01,120 --> 00:23:01,620 OK. 466 00:23:01,620 --> 00:23:03,390 I can remember. 467 00:23:03,390 --> 00:23:06,360 Skip, skip, skip. 468 00:23:06,360 --> 00:23:06,970 There we go. 469 00:23:06,970 --> 00:23:08,760 So this whole modularity idea comes back 470 00:23:08,760 --> 00:23:11,990 in in the 20th century. 471 00:23:11,990 --> 00:23:15,780 This is partly because of this information processing 472 00:23:15,780 --> 00:23:20,250 approach being respectable again, and it's also-- 473 00:23:20,250 --> 00:23:22,950 as people got started thinking about this again, 474 00:23:22,950 --> 00:23:26,580 they started looking for these cases, finding these cases, 475 00:23:26,580 --> 00:23:28,590 documenting patients with specific deficits. 476 00:23:28,590 --> 00:23:30,840 And one of the things they started finding in the '50s 477 00:23:30,840 --> 00:23:35,260 and '60s is, you'd see patients who had difficulty, 478 00:23:35,260 --> 00:23:38,370 for example, reading words for abstract nouns 479 00:23:38,370 --> 00:23:40,260 but not for concrete nouns, or defining 480 00:23:40,260 --> 00:23:44,670 words for abstract nouns versus concrete nouns. 481 00:23:44,670 --> 00:23:46,950 But even finer than that, you'd see 482 00:23:46,950 --> 00:23:50,250 patients who could not identify fruits and vegetables, 483 00:23:50,250 --> 00:23:52,470 but could identify other kinds of things, 484 00:23:52,470 --> 00:23:54,210 or patients who could not name colors 485 00:23:54,210 --> 00:23:55,751 but could name other kinds of things. 486 00:23:58,360 --> 00:24:02,160 And this is getting us into this idea of, OK, 487 00:24:02,160 --> 00:24:06,660 how does this work? 488 00:24:06,660 --> 00:24:09,150 One possibility is that in different-- you've 489 00:24:09,150 --> 00:24:11,024 got different tiny regions of your brain. 490 00:24:11,024 --> 00:24:12,690 And you've got one portion of your brain 491 00:24:12,690 --> 00:24:15,420 that stores color words, and one portion of your brain 492 00:24:15,420 --> 00:24:17,730 that stores shape words, and one portion of your brain 493 00:24:17,730 --> 00:24:20,040 that that stores names, and one portion of your brain 494 00:24:20,040 --> 00:24:23,640 that stores animals, and so on and so forth. 495 00:24:26,200 --> 00:24:31,200 That's possible, but as the fineness of categories goes up, 496 00:24:31,200 --> 00:24:34,019 it seems unlikely, especially given what we already know 497 00:24:34,019 --> 00:24:36,310 and what we've already talked about how the brain seems 498 00:24:36,310 --> 00:24:39,560 to store information like memory information-- 499 00:24:39,560 --> 00:24:42,250 as this distributed pattern of activation 500 00:24:42,250 --> 00:24:45,370 throughout different parts of the cortex. 501 00:24:45,370 --> 00:24:51,215 So how is lexical information stored in the brain? 502 00:24:51,215 --> 00:24:51,840 Jumping around. 503 00:24:54,750 --> 00:24:59,310 Linguists think of lexicons as having 504 00:24:59,310 --> 00:25:00,570 three important features. 505 00:25:00,570 --> 00:25:05,020 They say that you've got the phonetic information about 506 00:25:05,020 --> 00:25:10,722 the word and the orthographic information about a word-- 507 00:25:10,722 --> 00:25:12,430 that's how to write it, how to spell it-- 508 00:25:16,610 --> 00:25:18,040 and you've got the meaning. 509 00:25:25,040 --> 00:25:27,770 It's pretty clear there's got to be-- 510 00:25:27,770 --> 00:25:31,112 you've got to be able to go from a word's meaning 511 00:25:31,112 --> 00:25:32,570 and come up with its pronunciation, 512 00:25:32,570 --> 00:25:35,450 or from a word's pronunciation and come up with its meaning, 513 00:25:35,450 --> 00:25:36,590 fairly easily. 514 00:25:36,590 --> 00:25:37,940 Spoken language works. 515 00:25:37,940 --> 00:25:39,950 In normal functional human beings, 516 00:25:39,950 --> 00:25:44,315 there is a two-way connection between 517 00:25:44,315 --> 00:25:47,002 the phonetic information and the meaning information. 518 00:25:53,325 --> 00:25:55,930 There's some evidence showing there's also 519 00:25:55,930 --> 00:25:59,980 a direct connection between the orthographic information 520 00:25:59,980 --> 00:26:01,630 and the phonetic information. 521 00:26:01,630 --> 00:26:04,450 If I give you a word, you can write it down. 522 00:26:04,450 --> 00:26:08,386 If you see a word, you can read it out loud. 523 00:26:08,386 --> 00:26:10,510 Interestingly-- you'll see there's been a few cases 524 00:26:10,510 --> 00:26:15,160 of patients who, for example, can still read words, 525 00:26:15,160 --> 00:26:18,520 including words that are nonstandardly pronounced-- 526 00:26:18,520 --> 00:26:20,980 words with nonstandard spelling, because English 527 00:26:20,980 --> 00:26:23,160 has some of those-- 528 00:26:23,160 --> 00:26:25,680 but wouldn't be able to tell you what they meant, 529 00:26:25,680 --> 00:26:27,280 so that the link between the phonetic 530 00:26:27,280 --> 00:26:28,780 in the orthographic representation 531 00:26:28,780 --> 00:26:31,552 is still good but the link between either of them 532 00:26:31,552 --> 00:26:32,635 and the meaning is broken. 533 00:26:36,047 --> 00:26:37,630 There have been cases of patients who, 534 00:26:37,630 --> 00:26:40,930 if you ask them to define a word by speaking to them-- 535 00:26:40,930 --> 00:26:42,070 tell them the word-- 536 00:26:42,070 --> 00:26:44,350 and they'll say, I don't know what it means, 537 00:26:44,350 --> 00:26:47,050 but if I write it down and then look at it, 538 00:26:47,050 --> 00:26:49,540 I can tell you what it means. 539 00:26:49,540 --> 00:26:51,730 All of these cases seem to show that there's 540 00:26:51,730 --> 00:26:54,010 links to all three directions between these three 541 00:26:54,010 --> 00:26:58,030 pieces of the representation of a word. 542 00:26:58,030 --> 00:27:00,831 And this is a model that's built from the linguistics point 543 00:27:00,831 --> 00:27:01,330 of view. 544 00:27:01,330 --> 00:27:05,110 They're really interested in the things that 545 00:27:05,110 --> 00:27:07,030 make it a word, as something that 546 00:27:07,030 --> 00:27:11,829 can be combined into a sentence, versus what a knowledge 547 00:27:11,829 --> 00:27:13,870 and memory person might say, which is, how do you 548 00:27:13,870 --> 00:27:15,100 connect that with all the other things 549 00:27:15,100 --> 00:27:16,180 you know about a concept? 550 00:27:20,240 --> 00:27:22,282 Thinking a little bit more abstractly, 551 00:27:22,282 --> 00:27:23,990 and thinking a little bit more about what 552 00:27:23,990 --> 00:27:26,810 we know about how the brain stores 553 00:27:26,810 --> 00:27:32,000 other kinds of information, it's probable 554 00:27:32,000 --> 00:27:34,310 that when you think of a word-- 555 00:27:34,310 --> 00:27:37,100 think of the word hyena. 556 00:27:37,100 --> 00:27:41,480 You've got the phonetic component of it. 557 00:27:41,480 --> 00:27:43,416 You probably know how to spell it. 558 00:27:43,416 --> 00:27:45,290 And when you think about hyenas, you probably 559 00:27:45,290 --> 00:27:49,100 bring up all sorts of other associations, like Africa 560 00:27:49,100 --> 00:27:54,050 or savannas or The Lion King, or something 561 00:27:54,050 --> 00:27:56,690 that connects to your ideas of what hyenas are like. 562 00:27:59,600 --> 00:28:01,760 If you think about apples, you probably 563 00:28:01,760 --> 00:28:04,004 come up with the really salient things. 564 00:28:04,004 --> 00:28:05,795 How do you know when you're looking at an-- 565 00:28:05,795 --> 00:28:07,520 if you think about apples in the abstract sense, 566 00:28:07,520 --> 00:28:08,510 you probably get all sorts of stuff. 567 00:28:08,510 --> 00:28:09,230 You get color. 568 00:28:09,230 --> 00:28:10,010 You get the shape. 569 00:28:10,010 --> 00:28:12,230 You get the texture of biting into an apple. 570 00:28:12,230 --> 00:28:14,090 You get the flavor of an apple. 571 00:28:14,090 --> 00:28:17,650 All of these are associations with the pronunciation apple 572 00:28:17,650 --> 00:28:18,830 that you can bring up. 573 00:28:23,430 --> 00:28:26,225 There's a theory that says that for some of these categories-- 574 00:28:26,225 --> 00:28:28,400 and particularly for categories that you tend to see 575 00:28:28,400 --> 00:28:30,470 as specific deficits-- 576 00:28:30,470 --> 00:28:34,190 what kind of information is stored 577 00:28:34,190 --> 00:28:38,600 that is relevant to identifying things in that category 578 00:28:38,600 --> 00:28:39,540 is very specific. 579 00:28:39,540 --> 00:28:41,450 So for identifying fruits and vegetables, 580 00:28:41,450 --> 00:28:45,680 color and shape are probably really salient. 581 00:28:45,680 --> 00:28:47,320 Information about how this object moves 582 00:28:47,320 --> 00:28:49,200 is probably not so relevant. 583 00:28:49,200 --> 00:28:53,150 If we're identifying man-made objects, 584 00:28:53,150 --> 00:28:55,430 usually function is really-- it's 585 00:28:55,430 --> 00:28:57,050 this idea that function is important, 586 00:28:57,050 --> 00:28:58,917 but color is not so much so. 587 00:28:58,917 --> 00:29:01,250 For example, you can probably identify a chair no matter 588 00:29:01,250 --> 00:29:03,590 what color it is, but you might have difficulty 589 00:29:03,590 --> 00:29:06,930 identifying a purple banana. 590 00:29:06,930 --> 00:29:08,450 And so there's this theory that says 591 00:29:08,450 --> 00:29:12,620 that patients whose ability to represent color in their brains 592 00:29:12,620 --> 00:29:16,820 is decreased are then going to have difficulty making 593 00:29:16,820 --> 00:29:24,170 the connection between, for example, the word banana 594 00:29:24,170 --> 00:29:26,840 and a picture of a banana, because one of the pieces 595 00:29:26,840 --> 00:29:29,360 is not strong enough to bring up the other ones, 596 00:29:29,360 --> 00:29:31,850 because such a big piece of the meaning is missing. 597 00:29:34,430 --> 00:29:37,310 So I show one of these patients a picture of a banana, 598 00:29:37,310 --> 00:29:39,110 but their are yellow representation maybe 599 00:29:39,110 --> 00:29:40,070 isn't very strong. 600 00:29:40,070 --> 00:29:43,340 Their color module is damaged in some way. 601 00:29:43,340 --> 00:29:45,530 Then the pattern of activation isn't 602 00:29:45,530 --> 00:29:49,040 going to be able to bring up the word banana the way it would 603 00:29:49,040 --> 00:29:51,260 if the color representation was working. 604 00:29:51,260 --> 00:29:53,330 The leftover bits aren't strong enough 605 00:29:53,330 --> 00:29:54,560 to bring in that activation. 606 00:29:58,691 --> 00:30:01,180 Does that make sense? 607 00:30:01,180 --> 00:30:03,320 Do you want me to back up and go over it again? 608 00:30:03,320 --> 00:30:04,280 AUDIENCE: Yes. 609 00:30:04,280 --> 00:30:06,465 ABBY NOYCE: OK. 610 00:30:06,465 --> 00:30:08,090 I'm still wrapping my head around this. 611 00:30:08,090 --> 00:30:12,380 This was a new theory for me, so I'm not sure how clear it is. 612 00:30:12,380 --> 00:30:13,340 This theory says that-- 613 00:30:16,070 --> 00:30:20,330 we know that there are patients who have difficulty in naming, 614 00:30:20,330 --> 00:30:22,940 or difficulty identifying, particular categories 615 00:30:22,940 --> 00:30:23,440 of things. 616 00:30:23,440 --> 00:30:25,760 So if I have a stack of cards with pictures on them, 617 00:30:25,760 --> 00:30:27,700 and I just show them to patient and I say, what is this? 618 00:30:27,700 --> 00:30:29,240 They'll say-- I'll show them a house. 619 00:30:29,240 --> 00:30:30,020 They'll say, it's a house. 620 00:30:30,020 --> 00:30:30,980 I'll show them an alarm clock. 621 00:30:30,980 --> 00:30:31,790 They'll say, it's a clock. 622 00:30:31,790 --> 00:30:32,771 I'll show them a cat. 623 00:30:32,771 --> 00:30:33,770 They'll say, it's a cat. 624 00:30:33,770 --> 00:30:34,770 I'll show them an apple. 625 00:30:34,770 --> 00:30:37,640 They'll say, I've got no idea. 626 00:30:37,640 --> 00:30:39,420 So they'll be good in most categories 627 00:30:39,420 --> 00:30:41,630 and fail at one specific category. 628 00:30:41,630 --> 00:30:44,100 You'll see the same thing for animals. 629 00:30:44,100 --> 00:30:47,084 You'll see the same thing for man-made objects. 630 00:30:47,084 --> 00:30:49,250 You'll see the same thing for shapes or color names. 631 00:30:49,250 --> 00:30:52,470 There's very specific places in which this breaks. 632 00:30:52,470 --> 00:30:53,930 People will be good at most things, 633 00:30:53,930 --> 00:30:57,770 and unable to deal with one very specific category. 634 00:30:57,770 --> 00:31:01,070 What we know about how you hold information 635 00:31:01,070 --> 00:31:04,790 is that information is stored as a pattern of synapses 636 00:31:04,790 --> 00:31:07,670 in the brain of some sort, where one piece of a representation 637 00:31:07,670 --> 00:31:09,691 can bring up others. 638 00:31:09,691 --> 00:31:10,190 Yes, Sarah? 639 00:31:10,190 --> 00:31:13,046 AUDIENCE: Can they still see the color? 640 00:31:13,046 --> 00:31:15,782 Or were they just not able to identify it at all? 641 00:31:15,782 --> 00:31:17,740 ABBY NOYCE: For patients who can't name colors? 642 00:31:17,740 --> 00:31:18,365 AUDIENCE: Yeah. 643 00:31:18,365 --> 00:31:19,760 Would they still actually see it? 644 00:31:19,760 --> 00:31:22,679 ABBY NOYCE: Usually-- how do you test it? 645 00:31:22,679 --> 00:31:24,470 I think that people have done it by testing 646 00:31:24,470 --> 00:31:26,240 with matching things. 647 00:31:26,240 --> 00:31:29,325 And I know that at least some of these patients 648 00:31:29,325 --> 00:31:30,950 can definitely still do color matching. 649 00:31:30,950 --> 00:31:32,870 They can say, this color is the same as that color, 650 00:31:32,870 --> 00:31:34,661 but I have no idea what the name for it is. 651 00:31:37,100 --> 00:31:38,630 Or the same thing with a banana. 652 00:31:38,630 --> 00:31:41,130 They can say that this picture matches that picture, and not 653 00:31:41,130 --> 00:31:42,590 this piece of broccoli over here, 654 00:31:42,590 --> 00:31:44,465 but I don't know what the names for them are. 655 00:31:47,600 --> 00:31:49,322 So they're still seeing them. 656 00:31:49,322 --> 00:31:50,780 And you'll see things like patients 657 00:31:50,780 --> 00:31:55,130 who can, for example, not identify simple objects 658 00:31:55,130 --> 00:31:58,030 but can mime what they can do with them. 659 00:31:58,030 --> 00:32:01,500 So if you give them a key, they say, I don't know what this is, 660 00:32:01,500 --> 00:32:03,920 but this is what you do with it. 661 00:32:03,920 --> 00:32:04,610 Or a hairbrush. 662 00:32:04,610 --> 00:32:07,224 I don't know what this is, but this is what you do with it. 663 00:32:07,224 --> 00:32:09,140 And they might not be able to say what you do, 664 00:32:09,140 --> 00:32:09,800 but they can show it. 665 00:32:09,800 --> 00:32:11,360 So these patients aren't necessarily 666 00:32:11,360 --> 00:32:12,984 losing their knowledge about the world. 667 00:32:12,984 --> 00:32:15,380 They're losing, in some way, the connection 668 00:32:15,380 --> 00:32:20,840 between the linguistic, the phonetic information 669 00:32:20,840 --> 00:32:23,317 about a concept with everything else they 670 00:32:23,317 --> 00:32:24,275 know about the concept. 671 00:32:28,090 --> 00:32:30,800 The theory says that these categories 672 00:32:30,800 --> 00:32:32,300 where you'll see these very specific 673 00:32:32,300 --> 00:32:34,700 inabilities to name things-- 674 00:32:34,700 --> 00:32:39,950 usually that category is defined by some property or group 675 00:32:39,950 --> 00:32:40,620 of properties. 676 00:32:40,620 --> 00:32:42,950 So for fruit color, and shape are important properties 677 00:32:42,950 --> 00:32:44,750 for identifying fruit. 678 00:32:44,750 --> 00:32:48,800 For animals, probably motion patterns and some amount 679 00:32:48,800 --> 00:32:51,317 of structural detail. 680 00:32:51,317 --> 00:32:53,150 Think about what makes a deer look different 681 00:32:53,150 --> 00:32:57,590 from an antelope. it's mostly that the antlers are different. 682 00:32:57,590 --> 00:33:00,410 Functioning adult English speakers do this all the time. 683 00:33:00,410 --> 00:33:01,800 But a lot of-- 684 00:33:01,800 --> 00:33:04,250 but you'll see patients who can't make those distinctions 685 00:33:04,250 --> 00:33:05,480 anymore. 686 00:33:05,480 --> 00:33:08,630 And these guys' theory about why this happens 687 00:33:08,630 --> 00:33:11,480 says that what you would-- 688 00:33:11,480 --> 00:33:13,110 that if you can't-- 689 00:33:13,110 --> 00:33:15,680 I'm going to stick with fruit, because I keep jumping 690 00:33:15,680 --> 00:33:17,270 categories I'll get confused. 691 00:33:17,270 --> 00:33:20,630 If you can't identify fruit, then it's 692 00:33:20,630 --> 00:33:23,300 likely that the portions of your brain that 693 00:33:23,300 --> 00:33:26,660 hold shape information and color information 694 00:33:26,660 --> 00:33:30,380 are damaged in some ways, so that the shape-- 695 00:33:30,380 --> 00:33:32,660 if you're looking at the picture of the apple, 696 00:33:32,660 --> 00:33:36,620 what should happen is that the round shape and the red color 697 00:33:36,620 --> 00:33:39,180 and the stem on the top should-- 698 00:33:39,180 --> 00:33:42,620 that your representation of these in your visual cortex 699 00:33:42,620 --> 00:33:45,650 should then be able to trigger other information-- 700 00:33:45,650 --> 00:33:47,600 your representation of the sound apple, 701 00:33:47,600 --> 00:33:50,850 your representation of the spelling A-P-P-L-E, 702 00:33:50,850 --> 00:33:54,410 or representation of your memory of going apple picking in first 703 00:33:54,410 --> 00:33:56,000 grade, whatever. 704 00:33:58,970 --> 00:34:06,260 And that if your ability to represent color 705 00:34:06,260 --> 00:34:09,020 is flawed in some way, then-- 706 00:34:09,020 --> 00:34:11,270 maybe if you can still see the roundness and the stem, 707 00:34:11,270 --> 00:34:14,810 but you can't represent the red color the way 708 00:34:14,810 --> 00:34:16,489 that you used to be able to, maybe you 709 00:34:16,489 --> 00:34:19,230 had a stroke or something-- 710 00:34:19,230 --> 00:34:21,170 then the representation of the roundness 711 00:34:21,170 --> 00:34:23,480 and the stem without the color isn't 712 00:34:23,480 --> 00:34:25,880 enough to bring up the representations 713 00:34:25,880 --> 00:34:28,370 of the phonetic information and the spelling, 714 00:34:28,370 --> 00:34:34,227 and all these other aspects of the word apple, from the shape. 715 00:34:34,227 --> 00:34:37,340 Is that better? 716 00:34:37,340 --> 00:34:39,300 OK. 717 00:34:39,300 --> 00:34:40,800 This is really tying into this idea 718 00:34:40,800 --> 00:34:43,710 that we've been looking at for a couple of weeks, where 719 00:34:43,710 --> 00:34:46,500 information in the brain is a pattern of activation 720 00:34:46,500 --> 00:34:50,922 of some sort, a pattern of neurons for different concepts. 721 00:34:50,922 --> 00:34:52,630 And this gets really hard to think about, 722 00:34:52,630 --> 00:34:54,338 especially when you start trying to think 723 00:34:54,338 --> 00:34:56,639 about multiple things in mind at the same time, 724 00:34:56,639 --> 00:34:58,715 and it's all on top of each other. 725 00:34:58,715 --> 00:35:00,840 And nobody's really come up with a good explanation 726 00:35:00,840 --> 00:35:02,100 for how that works. 727 00:35:02,100 --> 00:35:03,360 Clearly it does work. 728 00:35:03,360 --> 00:35:04,980 We know we can do this. 729 00:35:04,980 --> 00:35:09,400 You can think about apples and bananas at the same time, 730 00:35:09,400 --> 00:35:09,900 right? 731 00:35:09,900 --> 00:35:10,400 I hope. 732 00:35:20,175 --> 00:35:22,675 So you get patients with these startlingly precise deficits. 733 00:35:30,920 --> 00:35:34,910 This is another patient from-- a 20th-century patient who had 734 00:35:34,910 --> 00:35:37,790 a small tumor in his left frontal lobe-- 735 00:35:37,790 --> 00:35:39,190 not quite in Broca's area. 736 00:35:39,190 --> 00:35:43,100 A little bit further and further forwards, kind of prefrontal. 737 00:35:43,100 --> 00:35:45,800 And he had surgery to have it removed. 738 00:35:45,800 --> 00:35:48,830 And afterwards, what they found is 739 00:35:48,830 --> 00:35:53,030 that he had a little bit of-- and they 740 00:35:53,030 --> 00:35:54,410 were testing him afterwards. 741 00:35:54,410 --> 00:35:55,910 If you do surgery on your brain, you 742 00:35:55,910 --> 00:35:57,500 run a bunch of neurological tests on people 743 00:35:57,500 --> 00:35:58,940 to make sure that came out of it OK. 744 00:35:58,940 --> 00:36:00,740 If they have deficits, you want to know about this 745 00:36:00,740 --> 00:36:02,823 before you let them get behind the wheel of a car, 746 00:36:02,823 --> 00:36:03,890 for example. 747 00:36:03,890 --> 00:36:06,680 And so they were testing him, and they found that in general, 748 00:36:06,680 --> 00:36:08,870 if you asked him to do activities with his right 749 00:36:08,870 --> 00:36:11,690 hand-- remember, right hand controlled by left frontal 750 00:36:11,690 --> 00:36:12,920 lobe-- 751 00:36:12,920 --> 00:36:16,089 that they were-- the right side of his body 752 00:36:16,089 --> 00:36:17,880 was a little weak, which is to be expected. 753 00:36:17,880 --> 00:36:21,174 You were futzing around in the left frontal lobe. 754 00:36:21,174 --> 00:36:22,715 And they were asking him to compare-- 755 00:36:25,259 --> 00:36:27,050 writing with both hands, and he's a righty. 756 00:36:27,050 --> 00:36:28,820 So he writes with his right hand, 757 00:36:28,820 --> 00:36:30,819 and they had him write stuff with his left hand. 758 00:36:30,819 --> 00:36:32,060 And his left hand was-- 759 00:36:32,060 --> 00:36:33,976 you'd expect a right who's writing left-handed 760 00:36:33,976 --> 00:36:35,990 to be sloppy, but they also found that he 761 00:36:35,990 --> 00:36:38,120 was making strange mistakes. 762 00:36:38,120 --> 00:36:40,640 He wrote-- they asked him to write yesterday, 763 00:36:40,640 --> 00:36:48,140 and he wrote Y-O-N-T-I, and very weird substitutions like that. 764 00:36:48,140 --> 00:36:50,510 And they started testing this further, 765 00:36:50,510 --> 00:36:53,150 and what they ended up finding-- and this is weird. 766 00:36:53,150 --> 00:36:54,680 Left hand controlled by right brain. 767 00:36:54,680 --> 00:36:56,900 We didn't do anything to the right hemisphere 768 00:36:56,900 --> 00:36:57,800 of this guy's brain. 769 00:36:57,800 --> 00:36:59,850 We didn't go there. 770 00:36:59,850 --> 00:37:03,590 And they started testing him further, and what they found 771 00:37:03,590 --> 00:37:06,830 was that if he could see items, he 772 00:37:06,830 --> 00:37:08,591 could identify them just fine. 773 00:37:08,591 --> 00:37:11,090 But if you had him put his hand into a box where he couldn't 774 00:37:11,090 --> 00:37:15,980 see it, and had him identify an item by touch, 775 00:37:15,980 --> 00:37:18,240 he could do it with his right hand just fine, 776 00:37:18,240 --> 00:37:20,610 but if you asked him to do it with his left hand, 777 00:37:20,610 --> 00:37:23,606 he couldn't name what these objects were. 778 00:37:23,606 --> 00:37:24,980 One possibility is, he's just not 779 00:37:24,980 --> 00:37:28,220 recognizing them when they come in with the left hand. 780 00:37:28,220 --> 00:37:30,830 But if they watched what his hands were doing, 781 00:37:30,830 --> 00:37:35,404 then he would be looking at would be the salient features 782 00:37:35,404 --> 00:37:37,820 of the object, the things that would tell you what it was. 783 00:37:37,820 --> 00:37:39,278 So if he was looking at a key, he'd 784 00:37:39,278 --> 00:37:42,780 run his fingers over the little bumpy edges of the key. 785 00:37:42,780 --> 00:37:45,152 If it was a toothbrush, he'd feel the long handle 786 00:37:45,152 --> 00:37:45,860 and the bristles. 787 00:37:45,860 --> 00:37:48,320 He knew what the important parts of this were, 788 00:37:48,320 --> 00:37:51,250 but he couldn't name it. 789 00:37:51,250 --> 00:37:55,710 And they said, OK, put it back down. 790 00:37:55,710 --> 00:37:57,200 Take your hand out. 791 00:37:57,200 --> 00:37:59,740 Can you draw it? 792 00:37:59,740 --> 00:38:01,610 Couldn't draw it with the right hand. 793 00:38:01,610 --> 00:38:02,985 Could draw it with the left hand, 794 00:38:02,985 --> 00:38:05,150 the hand that had felt it. 795 00:38:05,150 --> 00:38:09,280 They said, OK, can you mime what you would do with it? 796 00:38:09,280 --> 00:38:10,190 And he could. 797 00:38:10,190 --> 00:38:12,980 He could mime turning a key or combing his hair 798 00:38:12,980 --> 00:38:14,532 or brushing his teeth. 799 00:38:14,532 --> 00:38:16,490 And if they asked him to say what he was doing, 800 00:38:16,490 --> 00:38:20,810 he'd mime turning a key and say, I'm 801 00:38:20,810 --> 00:38:23,360 erasing a chalkboard with this. 802 00:38:23,360 --> 00:38:26,720 Or he'd brush his teeth and say, I'm combing my hair with this. 803 00:38:26,720 --> 00:38:29,060 So clearly the action is right. 804 00:38:29,060 --> 00:38:30,332 He knows what this object is. 805 00:38:30,332 --> 00:38:31,790 He knows what you would do with it. 806 00:38:31,790 --> 00:38:32,790 But he can't say it. 807 00:38:38,125 --> 00:38:41,164 AUDIENCE: Is this for his left hand? 808 00:38:41,164 --> 00:38:41,830 ABBY NOYCE: Yes. 809 00:38:41,830 --> 00:38:44,680 This is tasks with his left hand, which 810 00:38:44,680 --> 00:38:47,878 is controlled by his right brain, 811 00:38:47,878 --> 00:38:51,710 or right motor cortex, which they didn't do an operation to. 812 00:38:51,710 --> 00:38:54,409 And the hypothesis that these researchers came up with, 813 00:38:54,409 --> 00:38:56,950 and which was eventually proven to be more or less right when 814 00:38:56,950 --> 00:39:01,230 they did an autopsy on this patient many years later-- 815 00:39:01,230 --> 00:39:03,124 AUDIENCE: [INAUDIBLE] 816 00:39:03,124 --> 00:39:03,790 ABBY NOYCE: Yes. 817 00:39:03,790 --> 00:39:05,789 Just because you have a hypothesis does not mean 818 00:39:05,789 --> 00:39:07,630 you get to do autopsies on people-- 819 00:39:07,630 --> 00:39:10,870 was that during the operation, the patient's corpus callosum 820 00:39:10,870 --> 00:39:12,141 had been damaged. 821 00:39:12,141 --> 00:39:14,140 You guys know-- remember, the corpus callosum is 822 00:39:14,140 --> 00:39:16,760 that big tract of nerve fibers that goes between the two 823 00:39:16,760 --> 00:39:18,820 halves of your brain, so that the left brain 824 00:39:18,820 --> 00:39:21,600 and the right brain can talk to each other and be in sync 825 00:39:21,600 --> 00:39:25,060 and figure out what they're doing. 826 00:39:25,060 --> 00:39:27,450 And not all of his corpus callosum had been damaged. 827 00:39:27,450 --> 00:39:29,080 But what it looks like is that the part 828 00:39:29,080 --> 00:39:31,660 that was carrying verbal and linguistic information 829 00:39:31,660 --> 00:39:33,370 was damaged. 830 00:39:33,370 --> 00:39:35,870 If his left hand is feeling this object, 831 00:39:35,870 --> 00:39:40,540 then it goes into right sensory cortex, right? 832 00:39:40,540 --> 00:39:45,520 Right cortex handles language for most of the population. 833 00:39:45,520 --> 00:39:47,140 But it couldn't get over-- 834 00:39:47,140 --> 00:39:49,720 right cortex does not handle language. 835 00:39:49,720 --> 00:39:52,870 So it goes to the right cortex, but in order to say what it is, 836 00:39:52,870 --> 00:39:55,600 you've got to get the left hemisphere involved. 837 00:39:55,600 --> 00:39:57,040 The corpus callosum is broken. 838 00:39:57,040 --> 00:40:03,940 The information can't get to the parts of your brain 839 00:40:03,940 --> 00:40:05,720 that produce language. 840 00:40:05,720 --> 00:40:07,839 So this patient-- and the other thing 841 00:40:07,839 --> 00:40:09,880 you'll see that happens with patients like this-- 842 00:40:09,880 --> 00:40:11,440 if you say, what are you doing? 843 00:40:11,440 --> 00:40:12,930 Rather than saying, I don't know, 844 00:40:12,930 --> 00:40:15,340 they'll come up with some story about what they're doing. 845 00:40:15,340 --> 00:40:16,437 They'll say, I'm doing-- 846 00:40:16,437 --> 00:40:17,020 you heard him. 847 00:40:17,020 --> 00:40:20,470 He'd say I'm erasing a chalkboard. 848 00:40:20,470 --> 00:40:21,980 With my key-turning. 849 00:40:21,980 --> 00:40:22,480 No. 850 00:40:22,480 --> 00:40:23,230 But he would say-- 851 00:40:23,230 --> 00:40:29,080 because the brain will fill in something if it doesn't have 852 00:40:29,080 --> 00:40:30,531 good information to work with. 853 00:40:30,531 --> 00:40:32,030 You'll see that in other cases, too. 854 00:40:32,030 --> 00:40:34,575 Who was talking about macular degeneration at some point? 855 00:40:34,575 --> 00:40:36,700 So you'll see that people with macular degeneration 856 00:40:36,700 --> 00:40:38,658 will often hallucinate, because they don't have 857 00:40:38,658 --> 00:40:40,750 any information coming into that central part 858 00:40:40,750 --> 00:40:43,720 of their visual field, and the brain fills it in. 859 00:40:43,720 --> 00:40:45,920 You see the same thing happening here. 860 00:40:45,920 --> 00:40:50,800 The brain likes to have something that it can do. 861 00:40:50,800 --> 00:40:53,255 So this is another piece of information-- 862 00:40:53,255 --> 00:40:54,880 another piece of evidence for this idea 863 00:40:54,880 --> 00:40:58,174 that linguistic capabilities are located, really, 864 00:40:58,174 --> 00:40:59,840 in the left hemisphere, for most people. 865 00:40:59,840 --> 00:41:04,390 For 97% of righties, linguistic information 866 00:41:04,390 --> 00:41:07,342 is controlled by the left hemisphere, linguistic ability. 867 00:41:07,342 --> 00:41:08,800 It's not quite so high for lefties. 868 00:41:08,800 --> 00:41:09,770 How many lefties? 869 00:41:09,770 --> 00:41:10,780 Anyone in this room? 870 00:41:10,780 --> 00:41:11,680 Two? 871 00:41:11,680 --> 00:41:13,660 Not bad. 872 00:41:13,660 --> 00:41:16,070 So for lefties, there was-- originally, 873 00:41:16,070 --> 00:41:18,050 when this hemispheric dominance thing came out, 874 00:41:18,050 --> 00:41:20,900 everybody said, well, maybe it's just hand dominance. 875 00:41:20,900 --> 00:41:22,952 So if you're a righty, then you would 876 00:41:22,952 --> 00:41:24,660 expect to have left hemisphere dominance, 877 00:41:24,660 --> 00:41:26,354 so left hemisphere does language. 878 00:41:26,354 --> 00:41:27,770 For lefties, therefore, you should 879 00:41:27,770 --> 00:41:29,870 see that the right hemisphere does language. 880 00:41:29,870 --> 00:41:31,640 And people started actually studying this 881 00:41:31,640 --> 00:41:34,269 once you started getting like good brain imagery, where 882 00:41:34,269 --> 00:41:35,810 you ask people to do linguistic stuff 883 00:41:35,810 --> 00:41:38,210 and see which part of their brain is more active. 884 00:41:38,210 --> 00:41:39,950 And what they found is that in lefties-- 885 00:41:39,950 --> 00:41:42,230 70% of left-handed people, it's still 886 00:41:42,230 --> 00:41:44,390 the left hemisphere that controls language 887 00:41:44,390 --> 00:41:46,130 and linguistic ability. 888 00:41:46,130 --> 00:41:48,200 And the other 30% is about evenly 889 00:41:48,200 --> 00:41:53,217 split between right-hemisphere controlled and language being 890 00:41:53,217 --> 00:41:55,550 really split between the hemispheres, where both of them 891 00:41:55,550 --> 00:41:56,091 are involved. 892 00:41:59,000 --> 00:42:00,470 We will probably keep saying, this 893 00:42:00,470 --> 00:42:01,970 happens in the left hemisphere, but there's 894 00:42:01,970 --> 00:42:04,220 this big caveat here that says it's not necessarily 895 00:42:04,220 --> 00:42:05,220 like that for everybody. 896 00:42:05,220 --> 00:42:08,430 It's just like that for a big majority of the population, 897 00:42:08,430 --> 00:42:10,204 and we get sloppy in our terminology. 898 00:42:10,204 --> 00:42:15,400 AUDIENCE: So if you're ambidextrous, [INAUDIBLE]?? 899 00:42:15,400 --> 00:42:16,420 ABBY NOYCE: It might be. 900 00:42:16,420 --> 00:42:17,380 You'll see often if-- 901 00:42:17,380 --> 00:42:18,250 I don't know if you guys have noticed. 902 00:42:18,250 --> 00:42:20,770 There's usually study flyers posted along the hallways here, 903 00:42:20,770 --> 00:42:23,620 because it's summer, and college students are looking for cash, 904 00:42:23,620 --> 00:42:27,830 and being a study participant is usually easy money. 905 00:42:27,830 --> 00:42:30,390 They're like, 28-day sleep trial, $5,000. 906 00:42:30,390 --> 00:42:34,251 And college students are like, OK, I'll go for that. 907 00:42:34,251 --> 00:42:35,750 But anyway, one of the things you'll 908 00:42:35,750 --> 00:42:37,870 see with people who are studying-- doing imaging 909 00:42:37,870 --> 00:42:38,890 stuff with language, where they want 910 00:42:38,890 --> 00:42:40,180 to look at what parts of your brain 911 00:42:40,180 --> 00:42:41,554 are involved in linguistic tasks, 912 00:42:41,554 --> 00:42:43,616 will ask for native English speakers, which 913 00:42:43,616 --> 00:42:44,740 makes some amount of sense. 914 00:42:44,740 --> 00:42:47,740 You want to take this native language confound out of there. 915 00:42:47,740 --> 00:42:50,110 They'll also ask for people who are right-handed. 916 00:42:50,110 --> 00:42:51,730 They don't want to throw lefties in there because it 917 00:42:51,730 --> 00:42:52,729 will mess up their data. 918 00:42:52,729 --> 00:42:55,330 They want it to be as close as possible to a really clean data 919 00:42:55,330 --> 00:42:55,830 set. 920 00:42:58,690 --> 00:43:00,609 If I'm trying to average over 10 people, 921 00:43:00,609 --> 00:43:02,650 I want them all to be coming in with more or less 922 00:43:02,650 --> 00:43:04,188 the same configuration. 923 00:43:07,534 --> 00:43:09,281 Did someone else have a hand up? 924 00:43:09,281 --> 00:43:09,780 Yeah. 925 00:43:09,780 --> 00:43:10,321 AUDIENCE: OK. 926 00:43:10,321 --> 00:43:13,631 Well, this isn't really [INAUDIBLE],, but in some way 927 00:43:13,631 --> 00:43:15,247 it kind of is. 928 00:43:15,247 --> 00:43:17,235 I was at my optometrist's office, 929 00:43:17,235 --> 00:43:20,217 and my right eye is worse than my left eye. 930 00:43:20,217 --> 00:43:22,196 And I was asking my optometrist why that is. 931 00:43:22,196 --> 00:43:23,696 And he's like, are you right-handed? 932 00:43:23,696 --> 00:43:24,690 I'm like, yeah. 933 00:43:24,690 --> 00:43:26,181 He goes, that's why. 934 00:43:26,181 --> 00:43:28,587 But do you know how that connects? 935 00:43:28,587 --> 00:43:30,920 ABBY NOYCE: Your left eye is better than your right eye, 936 00:43:30,920 --> 00:43:31,800 or the other way around? 937 00:43:31,800 --> 00:43:32,425 AUDIENCE: Yeah. 938 00:43:32,425 --> 00:43:33,710 ABBY NOYCE: Left eye's better. 939 00:43:33,710 --> 00:43:34,980 I don't know. 940 00:43:34,980 --> 00:43:39,470 I don't have a good explanation for that, because-- 941 00:43:39,470 --> 00:43:42,524 also because, remember, that each eye's input gets split, 942 00:43:42,524 --> 00:43:44,940 because it gets split down the middle of the visual field. 943 00:43:44,940 --> 00:43:47,870 So each eye's input goes half to one side and half to the other. 944 00:43:47,870 --> 00:43:49,494 But the muscles that control it are all 945 00:43:49,494 --> 00:43:50,960 controlled from the opposite side. 946 00:43:50,960 --> 00:43:52,430 So your right eye is controlled from the left side 947 00:43:52,430 --> 00:43:53,680 of your brain, and vice versa. 948 00:43:56,630 --> 00:43:57,552 I don't know-- 949 00:43:57,552 --> 00:43:59,510 I'm one of those lucky kids with perfect vision 950 00:43:59,510 --> 00:44:01,370 who doesn't have glasses and stuff, 951 00:44:01,370 --> 00:44:05,880 so I don't know a lot about what effects it or makes it worse. 952 00:44:05,880 --> 00:44:09,870 But-- I was going somewhere with this. 953 00:44:09,870 --> 00:44:14,160 Something about-- there may be a connection 954 00:44:14,160 --> 00:44:15,660 between like how good your control 955 00:44:15,660 --> 00:44:17,490 of the muscles in each eye are versus how 956 00:44:17,490 --> 00:44:18,870 bad your vision in that eye is. 957 00:44:18,870 --> 00:44:19,980 I don't know. 958 00:44:19,980 --> 00:44:23,760 This is-- because it also has new how the eyeballs are shaped 959 00:44:23,760 --> 00:44:25,440 and stuff, and how good-- 960 00:44:25,440 --> 00:44:29,160 but it's also-- the reason people get far-sighted as they 961 00:44:29,160 --> 00:44:31,380 get older is that-- 962 00:44:31,380 --> 00:44:33,540 you've got a lens, and you basically 963 00:44:33,540 --> 00:44:35,720 change the thickness of it in order to focus. 964 00:44:35,720 --> 00:44:38,790 So there's muscles that contract it and that stretch out. 965 00:44:38,790 --> 00:44:44,430 And over time, those muscles stop having as much range 966 00:44:44,430 --> 00:44:46,720 as they do in younger people. 967 00:44:46,720 --> 00:44:48,090 So this is why-- 968 00:44:48,090 --> 00:44:53,790 hit 40 or 50, and everybody starts needing reading glasses. 969 00:44:53,790 --> 00:44:57,715 Almost all middle-aged on up adults do. 970 00:44:57,715 --> 00:44:59,340 I remember when my dad had to get them. 971 00:44:59,340 --> 00:45:01,560 I get my perfect vision from my father, who was always-- 972 00:45:01,560 --> 00:45:03,300 for his entire life had been very smug about, oh, 973 00:45:03,300 --> 00:45:04,175 I don't need glasses. 974 00:45:04,175 --> 00:45:05,670 My mom is nearsighted. 975 00:45:05,670 --> 00:45:07,560 And of course, he just refused to admit 976 00:45:07,560 --> 00:45:09,240 that he needed reading glasses, until it got to the point 977 00:45:09,240 --> 00:45:11,240 where he was like-- we were going out to dinner, 978 00:45:11,240 --> 00:45:13,970 and he's trying to read a menu way out there, 979 00:45:13,970 --> 00:45:16,960 because he couldn't focus on it any closer to his face. 980 00:45:16,960 --> 00:45:20,280 Nothing like being laughed at by your children to get you to go 981 00:45:20,280 --> 00:45:21,240 get reading glasses. 982 00:45:21,240 --> 00:45:22,140 OK. 983 00:45:22,140 --> 00:45:23,100 Back on topic. 984 00:45:26,264 --> 00:45:27,930 Looking at what kind of deficit patients 985 00:45:27,930 --> 00:45:30,480 have tells us something, gives us 986 00:45:30,480 --> 00:45:33,390 some clue about how language is stored in the brain. 987 00:45:33,390 --> 00:45:35,010 You can think of this as-- 988 00:45:35,010 --> 00:45:37,830 we were thinking already about episodic memory, 989 00:45:37,830 --> 00:45:39,000 about memory for events. 990 00:45:39,000 --> 00:45:42,570 You can think similarly about memory for word meanings, 991 00:45:42,570 --> 00:45:44,820 especially for concrete word meanings-- 992 00:45:44,820 --> 00:45:48,570 words that refer to categories of things or names, 993 00:45:48,570 --> 00:45:54,330 being able to in turn bring up representations of the sensory 994 00:45:54,330 --> 00:45:56,940 pieces that go together with that, 995 00:45:56,940 --> 00:45:59,460 but also to be able to bring up these more abstract things 996 00:45:59,460 --> 00:46:02,670 like, the pronunciation of a word or other words that mean 997 00:46:02,670 --> 00:46:04,980 roughly the same thing-- synonyms-- 998 00:46:04,980 --> 00:46:07,040 or the spelling of a word. 999 00:46:07,040 --> 00:46:09,810 And all of this is still hinging on this idea 1000 00:46:09,810 --> 00:46:12,790 for one part of a representation to activate others and bring 1001 00:46:12,790 --> 00:46:13,290 them online. 1002 00:46:17,601 --> 00:46:18,500 All right. 1003 00:46:18,500 --> 00:46:19,545 Oh, hey. 1004 00:46:19,545 --> 00:46:20,420 Let's see if we get-- 1005 00:46:23,336 --> 00:46:25,280 Do I have internet yet? 1006 00:46:25,280 --> 00:46:26,738 Go, Firefox, go. 1007 00:46:31,231 --> 00:46:33,480 This is what happens when I build this on one computer 1008 00:46:33,480 --> 00:46:34,610 and play it on another. 1009 00:46:44,610 --> 00:46:46,810 Have you guys seen this before, the McGurk effect? 1010 00:46:46,810 --> 00:46:47,893 Has anyone here seen this? 1011 00:46:47,893 --> 00:46:49,630 This is a classic-- 1012 00:46:49,630 --> 00:46:50,440 oh, shut up. 1013 00:46:58,216 --> 00:47:00,084 That's what I wanted. 1014 00:47:00,084 --> 00:47:01,050 Go. 1015 00:47:01,050 --> 00:47:03,700 Can we go? 1016 00:47:03,700 --> 00:47:04,750 Yeah. 1017 00:47:04,750 --> 00:47:05,500 OK. 1018 00:47:05,500 --> 00:47:12,370 This is a classic effect of how different things that 1019 00:47:12,370 --> 00:47:15,130 are happening when you're perceiving language 1020 00:47:15,130 --> 00:47:16,129 have an effect on it. 1021 00:47:16,129 --> 00:47:17,920 What I want you guys to do is-- we're going 1022 00:47:17,920 --> 00:47:21,250 to play with a bunch of times. 1023 00:47:21,250 --> 00:47:23,020 We're going to pause. 1024 00:47:23,020 --> 00:47:23,520 Shh. 1025 00:47:23,520 --> 00:47:24,869 Are we done? 1026 00:47:24,869 --> 00:47:26,160 Let me get you some sound here. 1027 00:47:26,160 --> 00:47:26,690 And you're only going to, I think, 1028 00:47:26,690 --> 00:47:28,150 get the sound off of this. 1029 00:47:28,150 --> 00:47:32,240 But-- oh, come on. 1030 00:47:32,240 --> 00:47:32,794 Unmute. 1031 00:47:32,794 --> 00:47:33,960 AUDIENCE: Is there a volume? 1032 00:47:33,960 --> 00:47:35,060 ABBY NOYCE: There is a volume thing over there, 1033 00:47:35,060 --> 00:47:36,685 but my computer is not plugged into it. 1034 00:47:36,685 --> 00:47:39,830 Let me see if this is good enough to work. 1035 00:47:39,830 --> 00:47:40,990 Sound. 1036 00:47:40,990 --> 00:47:41,490 All right. 1037 00:47:44,424 --> 00:47:45,840 I want you guys to close your eyes 1038 00:47:45,840 --> 00:47:47,548 and just listen to the soundtrack of this 1039 00:47:47,548 --> 00:47:49,160 for a minute. 1040 00:47:49,160 --> 00:47:50,160 [AUDIO PLAYBACK] 1041 00:47:50,160 --> 00:47:54,047 - Ba ba, ba ba, ba ba. 1042 00:47:54,047 --> 00:47:54,630 [END PLAYBACK] 1043 00:47:54,630 --> 00:47:55,546 ABBY NOYCE: All right. 1044 00:47:55,546 --> 00:47:57,300 What did he say? 1045 00:47:57,300 --> 00:47:58,230 AUDIENCE: [INAUDIBLE] 1046 00:47:58,230 --> 00:48:00,270 ABBY NOYCE: Not loud enough to hear it? 1047 00:48:00,270 --> 00:48:02,460 Do I have an audio cable thing? 1048 00:48:02,460 --> 00:48:04,026 I do. 1049 00:48:04,026 --> 00:48:06,000 AUDIENCE: I thought he said ba ba. 1050 00:48:06,000 --> 00:48:07,910 ABBY NOYCE: Yeah? 1051 00:48:07,910 --> 00:48:10,400 Where's the sound out on this baby? 1052 00:48:10,400 --> 00:48:12,395 Come on, baby. 1053 00:48:12,395 --> 00:48:15,125 Where is it? 1054 00:48:15,125 --> 00:48:21,680 [INAUDIBLE] No. 1055 00:48:21,680 --> 00:48:24,140 I know you have a sound out. 1056 00:48:24,140 --> 00:48:24,891 It's in the front. 1057 00:48:24,891 --> 00:48:26,765 AUDIENCE: His hair is definitely really nice. 1058 00:48:26,765 --> 00:48:28,826 ABBY NOYCE: Was his hair better than yours? 1059 00:48:28,826 --> 00:48:30,634 AUDIENCE: [INAUDIBLE] 1060 00:48:30,634 --> 00:48:31,550 ABBY NOYCE: All right. 1061 00:48:31,550 --> 00:48:32,341 Listen to it again. 1062 00:48:35,729 --> 00:48:37,020 Now we need the volume control. 1063 00:48:40,104 --> 00:48:41,020 Not into the red zone. 1064 00:48:49,060 --> 00:48:49,790 Nope. 1065 00:48:49,790 --> 00:48:50,630 I don't know. 1066 00:48:50,630 --> 00:48:51,963 We're not going to go with that. 1067 00:48:51,963 --> 00:48:54,990 OK, we'll go back to doing it on the laptop. 1068 00:48:54,990 --> 00:49:00,206 AUDIENCE: [INAUDIBLE] bus yesterday when-- 1069 00:49:00,206 --> 00:49:01,164 ABBY NOYCE: [? Nisa? ?] 1070 00:49:01,164 --> 00:49:01,788 AUDIENCE: Yeah. 1071 00:49:01,788 --> 00:49:05,059 When she [INAUDIBLE] we had to put it on full blast, too. 1072 00:49:05,059 --> 00:49:06,975 ABBY NOYCE: You had to pull it all the way up? 1073 00:49:06,975 --> 00:49:07,690 All right. 1074 00:49:07,690 --> 00:49:09,590 We'll try it. 1075 00:49:09,590 --> 00:49:10,409 AUDIENCE: Oh, yeah. 1076 00:49:10,409 --> 00:49:12,950 ABBY NOYCE: My life as a techie makes me not want to do that. 1077 00:49:12,950 --> 00:49:13,450 All right. 1078 00:49:13,450 --> 00:49:14,830 That's all the way up. 1079 00:49:14,830 --> 00:49:15,330 Way up. 1080 00:49:18,400 --> 00:49:21,355 That's bad. 1081 00:49:21,355 --> 00:49:22,345 [AUDIO PLAYBACK] 1082 00:49:22,345 --> 00:49:26,263 - Ba ba, ba ba, ba ba. 1083 00:49:26,263 --> 00:49:26,970 [END PLAYBACK] 1084 00:49:26,970 --> 00:49:27,520 ABBY NOYCE: All right. 1085 00:49:27,520 --> 00:49:29,350 What syllable did-- what did it sound like he was saying? 1086 00:49:29,350 --> 00:49:30,396 What syllable? 1087 00:49:30,396 --> 00:49:33,132 AUDIENCE: Ba. 1088 00:49:33,132 --> 00:49:34,605 AUDIENCE: It sounded like bye. 1089 00:49:34,605 --> 00:49:35,980 ABBY NOYCE: How many people think 1090 00:49:35,980 --> 00:49:39,217 they heard a B as in balloon as the first consonant? 1091 00:49:39,217 --> 00:49:41,050 How many would heard they thought they heard 1092 00:49:41,050 --> 00:49:43,700 a D as in donkey? 1093 00:49:43,700 --> 00:49:45,950 AUDIENCE: Wait, does [INAUDIBLE]?? 1094 00:49:45,950 --> 00:49:47,705 Because I could [INAUDIBLE]. 1095 00:49:47,705 --> 00:49:49,330 ABBY NOYCE: It's not as well-- it's not 1096 00:49:49,330 --> 00:49:50,890 the best dubbed version of this. 1097 00:49:50,890 --> 00:49:52,940 Listen to it with your eyes closed one time, 1098 00:49:52,940 --> 00:49:54,040 and it should be pretty straightforward 1099 00:49:54,040 --> 00:49:54,880 what he's saying. 1100 00:49:54,880 --> 00:49:55,510 And then it's different-- 1101 00:49:55,510 --> 00:49:56,340 [AUDIO PLAYBACK] 1102 00:49:56,340 --> 00:49:59,777 - Ba ba, ba ba, ba ba. 1103 00:49:59,777 --> 00:50:00,360 [END PLAYBACK] 1104 00:50:00,360 --> 00:50:00,700 ABBY NOYCE: All right. 1105 00:50:00,700 --> 00:50:01,667 What was that? 1106 00:50:01,667 --> 00:50:02,621 AUDIENCE: Ba ba. 1107 00:50:06,200 --> 00:50:07,572 ABBY NOYCE: Ba ba? 1108 00:50:07,572 --> 00:50:08,544 All right. 1109 00:50:08,544 --> 00:50:09,960 What most people hear with this is 1110 00:50:09,960 --> 00:50:11,460 that it if they listed to it with their eyes closed, 1111 00:50:11,460 --> 00:50:12,918 if they just get the soundtrack, it 1112 00:50:12,918 --> 00:50:17,220 sounds like he's saying ba ba, ba ba, ba ba. 1113 00:50:17,220 --> 00:50:18,599 If you watch it-- 1114 00:50:18,599 --> 00:50:20,265 and it's better on the second two rounds 1115 00:50:20,265 --> 00:50:21,181 than on the first one. 1116 00:50:21,181 --> 00:50:22,983 The first one is dubbed a little weird. 1117 00:50:22,983 --> 00:50:23,930 [AUDIO PLAYBACK] 1118 00:50:23,930 --> 00:50:27,677 - Ba ba, ba ba, ba ba. 1119 00:50:27,677 --> 00:50:28,260 [END PLAYBACK] 1120 00:50:28,260 --> 00:50:29,968 ABBY NOYCE: Then what people tend to hear 1121 00:50:29,968 --> 00:50:34,940 is da da, da da, da da. 1122 00:50:34,940 --> 00:50:36,000 You heard L's? 1123 00:50:36,000 --> 00:50:37,350 AUDIENCE: I heard G's. 1124 00:50:37,350 --> 00:50:38,550 ABBY NOYCE: You heard G? 1125 00:50:38,550 --> 00:50:40,320 What this is is-- 1126 00:50:40,320 --> 00:50:41,040 here's a secret. 1127 00:50:41,040 --> 00:50:44,340 The soundtrack and the video, not recorded at the same time. 1128 00:50:44,340 --> 00:50:49,410 The soundtrack is of this guy saying ba ba, with B's. 1129 00:50:49,410 --> 00:50:53,267 The video of him saying ga ga, with G's. 1130 00:50:53,267 --> 00:50:54,548 AUDIENCE: I like the D's. 1131 00:50:54,548 --> 00:50:55,214 ABBY NOYCE: Huh? 1132 00:50:55,214 --> 00:50:56,910 AUDIENCE: I like the D's-- 1133 00:50:56,910 --> 00:50:59,070 ABBY NOYCE: Because if you mix-- 1134 00:50:59,070 --> 00:51:03,390 if you combine these two, what you most people hear 1135 00:51:03,390 --> 00:51:07,550 is a sound that's midway in between the guh and the buh, 1136 00:51:07,550 --> 00:51:14,130 which is the da da, da da, da da sound. 1137 00:51:14,130 --> 00:51:15,540 This is called the McGurk effect. 1138 00:51:15,540 --> 00:51:17,450 It was documented, oh, geez, I don't know. 1139 00:51:17,450 --> 00:51:20,700 20th century some time. 1140 00:51:20,700 --> 00:51:24,120 And this is an example of-- 1141 00:51:24,120 --> 00:51:26,561 what's affecting what you hear is not just the sound 1142 00:51:26,561 --> 00:51:27,060 coming in. 1143 00:51:27,060 --> 00:51:29,640 Just watching this guy's lips moving 1144 00:51:29,640 --> 00:51:32,760 with something that's inconsistent with the sound 1145 00:51:32,760 --> 00:51:34,933 causes you to perceive something different. 1146 00:51:40,490 --> 00:51:43,140 Back around to our big theme of what you perceive 1147 00:51:43,140 --> 00:51:44,940 is not necessarily what's actually 1148 00:51:44,940 --> 00:51:46,170 out there in the real world. 1149 00:51:51,406 --> 00:51:52,360 [AUDIO PLAYBACK] 1150 00:51:52,360 --> 00:51:56,307 - Ba ba, ba ba, ba ba. 1151 00:51:56,307 --> 00:51:56,890 [END PLAYBACK] 1152 00:51:56,890 --> 00:51:59,431 ABBY NOYCE: I think the middle one is the strongest D effect. 1153 00:51:59,431 --> 00:52:01,786 I hear the G on the first one. 1154 00:52:01,786 --> 00:52:04,490 AUDIENCE: If I just look to the corner of the picture instead, 1155 00:52:04,490 --> 00:52:05,637 I hear the ba. 1156 00:52:05,637 --> 00:52:07,220 ABBY NOYCE: Yeah, because you're not-- 1157 00:52:07,220 --> 00:52:08,410 AUDIENCE: If you don't focus-- 1158 00:52:08,410 --> 00:52:09,576 ABBY NOYCE: Right, because-- 1159 00:52:09,576 --> 00:52:11,800 remember that you've only got good visual perception 1160 00:52:11,800 --> 00:52:14,110 right in the very center of your visual field. 1161 00:52:14,110 --> 00:52:17,740 If you off see the center, you can't see it so much. 1162 00:52:17,740 --> 00:52:20,640 Anyway, that's a cool effect. 1163 00:52:20,640 --> 00:52:23,080 It's one of the classics. 1164 00:52:23,080 --> 00:52:25,230 I just wanted to show it to you guys. 1165 00:52:25,230 --> 00:52:27,555 Oh! 1166 00:52:27,555 --> 00:52:29,890 Skip, skip, skip. 1167 00:52:29,890 --> 00:52:31,300 Come on. 1168 00:52:31,300 --> 00:52:32,180 Skip one more. 1169 00:52:32,180 --> 00:52:32,680 OK. 1170 00:52:36,510 --> 00:52:41,740 Shifting gears a little bit to talking about-- 1171 00:52:41,740 --> 00:52:43,422 AUDIENCE: You can also say outside 1172 00:52:43,422 --> 00:52:45,237 of a dog, a book is a man's best friend-- 1173 00:52:45,237 --> 00:52:46,570 ABBY NOYCE: Friend, and inside-- 1174 00:52:46,570 --> 00:52:47,670 AUDIENCE: A dog, it's too dark to read. 1175 00:52:47,670 --> 00:52:49,503 ABBY NOYCE: That's either him or Mark Twain, 1176 00:52:49,503 --> 00:52:53,680 depending on whose book of quotations you look at. 1177 00:52:53,680 --> 00:52:55,655 This is what's called a garden path sentence. 1178 00:52:55,655 --> 00:52:57,280 And this one is actually two sentences, 1179 00:52:57,280 --> 00:53:03,661 where it leads you down one interpretation of what 1180 00:53:03,661 --> 00:53:04,410 the sentence says. 1181 00:53:04,410 --> 00:53:06,760 If you just hear, I shot an elephant in my pajamas, 1182 00:53:06,760 --> 00:53:10,505 you envision Monsieur Marks there in his pajamas 1183 00:53:10,505 --> 00:53:11,380 shooting an elephant. 1184 00:53:11,380 --> 00:53:14,140 You think the proposition is modifying the subject, not 1185 00:53:14,140 --> 00:53:15,940 the object. 1186 00:53:15,940 --> 00:53:19,620 And the second sentence causes you to re-evaluate that. 1187 00:53:22,548 --> 00:53:24,012 AUDIENCE: I think of the sentence-- 1188 00:53:24,012 --> 00:53:27,996 I kept picturing an elephant actually in his pajamas. 1189 00:53:27,996 --> 00:53:29,620 ABBY NOYCE: For some reason or another, 1190 00:53:29,620 --> 00:53:34,770 you bring up the different interpretation of it, 1191 00:53:34,770 --> 00:53:36,580 which isn't like a problem or anything. 1192 00:53:36,580 --> 00:53:38,829 It's just perhaps a minority case, which 1193 00:53:38,829 --> 00:53:40,620 is one of the things that makes this funny, 1194 00:53:40,620 --> 00:53:43,470 is it causes you to do that double take. 1195 00:53:43,470 --> 00:53:44,970 Garden path sentences are an example 1196 00:53:44,970 --> 00:53:46,770 of what's called structural ambiguity. 1197 00:53:46,770 --> 00:53:48,870 There's two kinds of ambiguity you can 1198 00:53:48,870 --> 00:53:51,240 encounter in parsing language. 1199 00:53:51,240 --> 00:53:54,780 There's word ambiguity-- like, a word like watch 1200 00:53:54,780 --> 00:53:58,320 can be either a verb meaning to look at something, 1201 00:53:58,320 --> 00:54:01,200 or a noun meaning a little clock that you carry around 1202 00:54:01,200 --> 00:54:03,470 with you in some way. 1203 00:54:03,470 --> 00:54:05,520 Word ambiguity is only moderately interesting, 1204 00:54:05,520 --> 00:54:05,790 I think. 1205 00:54:05,790 --> 00:54:07,414 I think structural ambiguity is cooler. 1206 00:54:10,130 --> 00:54:12,710 Sentences with structural ambiguity-- 1207 00:54:12,710 --> 00:54:16,990 or, for example, if I start out a sentence that says-- 1208 00:54:16,990 --> 00:54:17,769 let's see. 1209 00:54:17,769 --> 00:54:19,060 Oh, I know-- I have a headline. 1210 00:54:19,060 --> 00:54:19,559 What is it? 1211 00:54:28,757 --> 00:54:31,715 AUDIENCE: [INAUDIBLE] 1212 00:54:46,530 --> 00:54:49,710 ABBY NOYCE: Where do you think this sentence is going? 1213 00:54:49,710 --> 00:54:53,260 This is a newspaper headline that my Linguistics 101 teacher 1214 00:54:53,260 --> 00:54:55,950 brought us, which is, "85-year-old Fed secretary 1215 00:54:55,950 --> 00:55:02,850 to--" where it actually goes is, retire. 1216 00:55:02,850 --> 00:55:04,610 Pause. 1217 00:55:04,610 --> 00:55:05,700 Re-analyze. 1218 00:55:05,700 --> 00:55:08,810 That's Fed as in short for Federal something or other. 1219 00:55:08,810 --> 00:55:13,140 "85-year-old Fed secretary to retire." 1220 00:55:13,140 --> 00:55:16,135 That's another garden path sentence there, 1221 00:55:16,135 --> 00:55:19,440 where it sets you up to parse a bunch of earlier stuff one 1222 00:55:19,440 --> 00:55:20,970 way-- 1223 00:55:20,970 --> 00:55:25,680 at first you're thinking of this as subject, verb, object, 1224 00:55:25,680 --> 00:55:27,960 direct object. 1225 00:55:27,960 --> 00:55:30,770 And then you stop, and you have to go back and re-parse 1226 00:55:30,770 --> 00:55:32,220 all of the earlier stuff. 1227 00:55:32,220 --> 00:55:37,320 This is now adjective, noun, verb. 1228 00:55:37,320 --> 00:55:39,624 And it takes a minute to do it. 1229 00:55:39,624 --> 00:55:41,346 AUDIENCE: Basically, those sentences 1230 00:55:41,346 --> 00:55:44,172 have multiple syntactic maps. 1231 00:55:44,172 --> 00:55:44,880 ABBY NOYCE: Yeah. 1232 00:55:44,880 --> 00:55:46,590 Exactly right. 1233 00:55:46,590 --> 00:55:48,900 If you were breaking it down in a syntax tree, 1234 00:55:48,900 --> 00:55:51,362 like we were doing with the poodle in the closet before, 1235 00:55:51,362 --> 00:55:53,320 then there's two possible ways you could do it, 1236 00:55:53,320 --> 00:55:57,430 and which of those you follow affects the outcome. 1237 00:55:57,430 --> 00:56:00,690 So there's two-- and clearly, you 1238 00:56:00,690 --> 00:56:03,600 don't hold the whole thing in your mind 1239 00:56:03,600 --> 00:56:05,580 until you've gotten the whole sentence 1240 00:56:05,580 --> 00:56:07,560 and then analyze the structure. 1241 00:56:07,560 --> 00:56:09,810 Otherwise these things wouldn't be so jarring 1242 00:56:09,810 --> 00:56:13,530 when they throw you the other direction. 1243 00:56:13,530 --> 00:56:17,610 There's two hypotheses for how this works. 1244 00:56:17,610 --> 00:56:20,280 The parser hypothesis, the simpler hypothesis, 1245 00:56:20,280 --> 00:56:23,130 says that you go with what you've got, 1246 00:56:23,130 --> 00:56:28,470 and you pick whichever syntactic structure would be simpler-- 1247 00:56:28,470 --> 00:56:31,020 or, more common, you've got weighted preferences 1248 00:56:31,020 --> 00:56:34,230 for different syntactic structures-- 1249 00:56:34,230 --> 00:56:37,062 and then that structure is checked 1250 00:56:37,062 --> 00:56:39,270 against meaning as the rest of the sentence comes in. 1251 00:56:39,270 --> 00:56:41,880 And when it's wrong, you go back and re-analyze. 1252 00:56:41,880 --> 00:56:46,140 But this parser hypothesis says that as each new word comes in, 1253 00:56:46,140 --> 00:56:48,780 you pick the best available syntactic structure at the time 1254 00:56:48,780 --> 00:56:51,970 and just go with it. 1255 00:56:51,970 --> 00:56:55,290 And then there's this slightly more sophisticated 1256 00:56:55,290 --> 00:56:58,050 ambiguity resolution hypothesis, which most people are 1257 00:56:58,050 --> 00:57:00,720 leaning in favor of nowadays. 1258 00:57:00,720 --> 00:57:04,540 And that might just be that it has this currently trendy idea 1259 00:57:04,540 --> 00:57:07,390 of bottom-up and top-down processes in competition. 1260 00:57:07,390 --> 00:57:09,324 And this is how-- 1261 00:57:09,324 --> 00:57:11,490 a lot of our current understanding of how this works 1262 00:57:11,490 --> 00:57:16,290 seems to be similar across different areas of knowledge, 1263 00:57:16,290 --> 00:57:19,240 which is a good thing. 1264 00:57:19,240 --> 00:57:21,466 It's nice-- it's reassuring to see 1265 00:57:21,466 --> 00:57:23,590 the same processes in different parts of the brain. 1266 00:57:23,590 --> 00:57:26,250 That seems like it should be what we should see, 1267 00:57:26,250 --> 00:57:28,650 but it also leads to-- someone would be concerned 1268 00:57:28,650 --> 00:57:30,150 that we're getting caught up in what's currently 1269 00:57:30,150 --> 00:57:31,155 fashionable in science. 1270 00:57:33,830 --> 00:57:35,520 The ambiguity resolution hypothesis 1271 00:57:35,520 --> 00:57:43,950 says that all the way through the parsing process, 1272 00:57:43,950 --> 00:57:45,800 all the way through hearing a sentence, 1273 00:57:45,800 --> 00:57:48,060 that the bottom-up information that's coming in 1274 00:57:48,060 --> 00:57:52,560 can activate these different structural representations 1275 00:57:52,560 --> 00:57:55,170 and top-down processes about context. 1276 00:57:55,170 --> 00:57:57,360 If you already know that we're talking about-- 1277 00:57:57,360 --> 00:57:59,680 if you read this on the business page, for example, 1278 00:57:59,680 --> 00:58:02,580 you might be already biased away from the fed secretary 1279 00:58:02,580 --> 00:58:05,460 to the lions interpretation of it, 1280 00:58:05,460 --> 00:58:07,620 because you've got the context. 1281 00:58:07,620 --> 00:58:09,120 So all of these different things can 1282 00:58:09,120 --> 00:58:12,410 activate different representations, 1283 00:58:12,410 --> 00:58:14,166 and whichever representation is activated 1284 00:58:14,166 --> 00:58:15,540 the strongest is the one that you 1285 00:58:15,540 --> 00:58:18,828 use at any given point in time. 1286 00:58:18,828 --> 00:58:21,790 AUDIENCE: Who said dogs? 1287 00:58:21,790 --> 00:58:23,362 ABBY NOYCE: Someone down there. 1288 00:58:23,362 --> 00:58:24,838 AUDIENCE: What? 1289 00:58:24,838 --> 00:58:28,282 AUDIENCE: [INAUDIBLE] 1290 00:58:28,282 --> 00:58:30,742 AUDIENCE: Oh, I thought it said dots. 1291 00:58:30,742 --> 00:58:31,718 AUDIENCE: I like dots. 1292 00:58:31,718 --> 00:58:32,218 [INAUDIBLE] 1293 00:58:32,218 --> 00:58:34,560 ABBY NOYCE: I've never met any ferocious, meat-eating dots. 1294 00:58:34,560 --> 00:58:36,101 I don't know about the rest of y'all. 1295 00:58:40,620 --> 00:58:43,600 All right. 1296 00:58:43,600 --> 00:58:45,510 So the other thing I wanted to talk about 1297 00:58:45,510 --> 00:58:47,040 was a little bit about what we know 1298 00:58:47,040 --> 00:58:49,710 about how kids learn language. 1299 00:58:49,710 --> 00:58:53,205 Because clearly, just about everybody does learn language. 1300 00:58:56,460 --> 00:58:59,730 And up until probably the middle of the 20th century, 1301 00:58:59,730 --> 00:59:02,820 just how difficult this was wasn't quite appreciated. 1302 00:59:02,820 --> 00:59:04,860 So language is clearly learned, right? 1303 00:59:04,860 --> 00:59:08,220 Kids learn whatever language they're exposed to. 1304 00:59:08,220 --> 00:59:13,936 If you are-- if you grow up in a household with people who speak 1305 00:59:13,936 --> 00:59:15,060 English, you speak English. 1306 00:59:15,060 --> 00:59:17,560 If you grow up in a household with people who speak Spanish, 1307 00:59:17,560 --> 00:59:19,490 you speak Spanish. 1308 00:59:19,490 --> 00:59:21,210 If your parents spoke Cherokee, and you 1309 00:59:21,210 --> 00:59:23,626 are adopted into a household with people who speak French, 1310 00:59:23,626 --> 00:59:25,604 you will grow up speaking French. 1311 00:59:25,604 --> 00:59:28,020 It's not like there's this genetic component-- there's not 1312 00:59:28,020 --> 00:59:32,056 like, there's a genetic restriction on what language 1313 00:59:32,056 --> 00:59:32,680 you will speak. 1314 00:59:32,680 --> 00:59:37,140 It's totally dependent on what you grow up hearing. 1315 00:59:37,140 --> 00:59:38,920 There are lots of languages out there. 1316 00:59:38,920 --> 00:59:40,975 They've got a wide variety of vocabulary. 1317 00:59:40,975 --> 00:59:46,560 They've got a wide variety of grammatical structures. 1318 00:59:46,560 --> 00:59:48,540 And kids who are not exposed to language 1319 00:59:48,540 --> 00:59:51,030 during this kind of critical development period 1320 00:59:51,030 --> 00:59:52,410 don't learn it. 1321 00:59:52,410 --> 00:59:54,510 Did I tell you guys about Jeanie on Monday? 1322 00:59:54,510 --> 00:59:55,540 California 70s? 1323 00:59:55,540 --> 00:59:56,260 Yeah. 1324 00:59:56,260 --> 00:59:57,700 AUDIENCE: [INAUDIBLE] 1325 00:59:57,700 --> 00:59:58,493 ABBY NOYCE: Yeah. 1326 00:59:58,493 --> 01:00:00,435 AUDIENCE: [INAUDIBLE] 1327 01:00:00,435 --> 01:00:02,310 ABBY NOYCE: So there's been a couple of cases 1328 01:00:02,310 --> 01:00:07,320 of kids who were raised in more or less abusive situations 1329 01:00:07,320 --> 01:00:11,559 and were not exposed to language for some period of time. 1330 01:00:11,559 --> 01:00:13,350 So the case-- one of the better known cases 1331 01:00:13,350 --> 01:00:18,330 is a girl named Genie who was from California 1332 01:00:18,330 --> 01:00:21,367 whose parents basically locked her in a closet for 12 years, 1333 01:00:21,367 --> 01:00:23,700 didn't talk to her, brought her food and water from time 1334 01:00:23,700 --> 01:00:25,890 to time. 1335 01:00:25,890 --> 01:00:27,480 Very abusive situation. 1336 01:00:27,480 --> 01:00:28,710 Bad situation. 1337 01:00:28,710 --> 01:00:30,720 Was eventually taken out of that home 1338 01:00:30,720 --> 01:00:33,670 and moved into foster care. 1339 01:00:33,670 --> 01:00:35,970 But never learned language in the way 1340 01:00:35,970 --> 01:00:38,550 that any kind of normal person does. 1341 01:00:38,550 --> 01:00:42,510 Like, I think she developed a vocabulary of like 150 words. 1342 01:00:42,510 --> 01:00:47,440 Most adults have somewhere between 50 and 100,000 words. 1343 01:00:47,440 --> 01:00:50,790 Never developed syntax or complicated sentence structure 1344 01:00:50,790 --> 01:00:51,874 or any of that. 1345 01:00:51,874 --> 01:00:53,540 So there's a critical development window 1346 01:00:53,540 --> 01:00:54,870 for being exposed to language. 1347 01:00:58,026 --> 01:00:59,470 No. 1348 01:00:59,470 --> 01:01:00,250 She's too old. 1349 01:01:00,250 --> 01:01:03,154 AUDIENCE: Is that why like, foreigners from America 1350 01:01:03,154 --> 01:01:07,026 like, are like, in like their teens or younger, their accent 1351 01:01:07,026 --> 01:01:08,612 usually [INAUDIBLE]? 1352 01:01:08,612 --> 01:01:11,070 ABBY NOYCE: Yeah, or people who come to America and they're 1353 01:01:11,070 --> 01:01:13,590 like, in elementary school, will just learn English 1354 01:01:13,590 --> 01:01:16,220 really well and really fast. 1355 01:01:16,220 --> 01:01:19,460 Yeah there's a critical window for it. 1356 01:01:19,460 --> 01:01:20,215 Yes? 1357 01:01:20,215 --> 01:01:22,950 AUDIENCE: I was reading about language evolution. 1358 01:01:22,950 --> 01:01:23,646 And-- 1359 01:01:23,646 --> 01:01:24,312 ABBY NOYCE: Ooh! 1360 01:01:24,312 --> 01:01:25,515 A fun topic! 1361 01:01:25,515 --> 01:01:26,140 AUDIENCE: Yeah. 1362 01:01:26,140 --> 01:01:27,995 And there was this thing about, like, 1363 01:01:27,995 --> 01:01:33,876 about [INAUDIBLE] I think that we have in like, ancient times, 1364 01:01:33,876 --> 01:01:35,816 I think this [INAUDIBLE]. 1365 01:01:35,816 --> 01:01:39,050 But there was like some kind of ruler 1366 01:01:39,050 --> 01:01:44,390 who thought that all languages were like, evolved from like, 1367 01:01:44,390 --> 01:01:46,030 one original language. 1368 01:01:46,030 --> 01:01:49,075 So you [INAUDIBLE] say a similar thing, like, 1369 01:01:49,075 --> 01:01:53,130 made someone stay like, away from anyone, like, 1370 01:01:53,130 --> 01:01:55,830 an d to see if they would develop-- 1371 01:01:55,830 --> 01:02:00,540 ABBY NOYCE: Some kind of like proto, super original language. 1372 01:02:00,540 --> 01:02:01,510 Yeah. 1373 01:02:01,510 --> 01:02:02,139 Did it work? 1374 01:02:02,139 --> 01:02:02,680 AUDIENCE: No. 1375 01:02:02,680 --> 01:02:03,670 ABBY NOYCE: No. 1376 01:02:03,670 --> 01:02:04,790 Yeah. 1377 01:02:04,790 --> 01:02:07,110 So there's been a couple of-- 1378 01:02:07,110 --> 01:02:08,714 yeah, so evidence like that shows-- 1379 01:02:08,714 --> 01:02:11,130 was what made people say, OK, clearly language is learned. 1380 01:02:11,130 --> 01:02:11,880 And for a long time, people thought 1381 01:02:11,880 --> 01:02:13,588 it was just like, learning anything else. 1382 01:02:13,588 --> 01:02:17,490 Learning to cook or be a farmer or any of the things 1383 01:02:17,490 --> 01:02:22,740 that young people learn to do pre the Industrial Revolution 1384 01:02:22,740 --> 01:02:25,740 for the vast majority of human history. 1385 01:02:25,740 --> 01:02:26,767 OK. 1386 01:02:26,767 --> 01:02:29,100 The things that young people learn to do post Industrial 1387 01:02:29,100 --> 01:02:31,200 Revolution are very different than things 1388 01:02:31,200 --> 01:02:34,530 the young people learned to do 200, 300 years ago. 1389 01:02:34,530 --> 01:02:36,960 Right? 1390 01:02:36,960 --> 01:02:37,655 OK. 1391 01:02:37,655 --> 01:02:39,930 AUDIENCE: [INAUDIBLE] 1392 01:02:39,930 --> 01:02:42,750 ABBY NOYCE: So in the '50s, Chomsky-- 1393 01:02:42,750 --> 01:02:46,200 this guy Chomsky, you all might've heard of him-- 1394 01:02:46,200 --> 01:02:48,580 pointed out, said, looked at this and said, 1395 01:02:48,580 --> 01:02:50,110 that can't be right. 1396 01:02:50,110 --> 01:02:51,360 And he pointed out two things. 1397 01:02:51,360 --> 01:02:54,120 He said that, OK, so if you speak a language, then 1398 01:02:54,120 --> 01:02:55,010 you can produce-- 1399 01:02:55,010 --> 01:02:56,884 basically, you can produce an infinite number 1400 01:02:56,884 --> 01:02:59,540 of different sentences using the vocabulary in that language, 1401 01:02:59,540 --> 01:03:01,362 recombining it in different ways. 1402 01:03:01,362 --> 01:03:03,070 Every day, you probably produce sentences 1403 01:03:03,070 --> 01:03:04,940 that you've never heard before in your life. 1404 01:03:04,940 --> 01:03:06,900 Every day, you hear sentences that you've never 1405 01:03:06,900 --> 01:03:08,350 heard before in your life. 1406 01:03:08,350 --> 01:03:10,260 Every day, you probably produce sentences 1407 01:03:10,260 --> 01:03:12,810 that it's reasonably likely nobody has ever 1408 01:03:12,810 --> 01:03:14,370 produced before. 1409 01:03:14,370 --> 01:03:18,900 So there's this-- so he said, OK, so this is like, infinite. 1410 01:03:18,900 --> 01:03:22,470 And he said, and little kids, like language 1411 01:03:22,470 --> 01:03:27,126 learning age kids, two or three-year-olds, can do this. 1412 01:03:27,126 --> 01:03:27,750 Learn to do it. 1413 01:03:27,750 --> 01:03:30,890 And they want to do it really, really, really fast. 1414 01:03:30,890 --> 01:03:36,480 Like, so most kids start talking between a year 1415 01:03:36,480 --> 01:03:37,550 and 18 months old. 1416 01:03:37,550 --> 01:03:38,604 Somewhere in there. 1417 01:03:38,604 --> 01:03:40,770 And language development is one of these things that 1418 01:03:40,770 --> 01:03:43,020 ranges really widely when it first starts showing up 1419 01:03:43,020 --> 01:03:45,404 and everybody pretty much evens out by first grade. 1420 01:03:45,404 --> 01:03:47,820 So if you got a two-year-old cousin who isn't talking yet, 1421 01:03:47,820 --> 01:03:49,596 don't freak out. 1422 01:03:49,596 --> 01:03:50,970 But most people start-- most kids 1423 01:03:50,970 --> 01:03:52,920 start talking between about a year and 18 months of age. 1424 01:03:52,920 --> 01:03:55,350 And they'll have like, single word utterances, right? 1425 01:03:55,350 --> 01:03:57,420 Anyone here hang out with like, just 1426 01:03:57,420 --> 01:04:00,750 starting to talk kids ever? 1427 01:04:00,750 --> 01:04:01,350 At some point? 1428 01:04:01,350 --> 01:04:03,210 Siblings, cousins, neighbors. 1429 01:04:03,210 --> 01:04:07,840 And they'll say things like, mama, and bottle, and up. 1430 01:04:07,840 --> 01:04:08,340 Right? 1431 01:04:08,340 --> 01:04:11,790 They'll use these-- and mine. 1432 01:04:11,790 --> 01:04:14,550 Little kids will say stuff, single word utterances, 1433 01:04:14,550 --> 01:04:16,200 to get something they want. 1434 01:04:16,200 --> 01:04:18,630 And kids usually, around six months 1435 01:04:18,630 --> 01:04:20,850 after they start doing that, will kind of 1436 01:04:20,850 --> 01:04:25,070 tend to graduate to two or maybe three words utterances. 1437 01:04:25,070 --> 01:04:28,740 So they'll say things like, juice good, 1438 01:04:28,740 --> 01:04:34,350 or want ball, or mama mine, or no. 1439 01:04:34,350 --> 01:04:37,740 They get no real early on a lot of the time. 1440 01:04:37,740 --> 01:04:41,190 And somewhere right after that, like between two and three, 1441 01:04:41,190 --> 01:04:44,550 they go from like, these two words structures to sentences. 1442 01:04:44,550 --> 01:04:47,640 Like, yesterday we went to the park. 1443 01:04:47,640 --> 01:04:49,620 Or I want the thing. 1444 01:04:49,620 --> 01:04:51,420 Or-- and they'll-- 1445 01:04:51,420 --> 01:04:53,130 there's this shift. 1446 01:04:53,130 --> 01:04:55,620 And once they produce sentences, within the next year, 1447 01:04:55,620 --> 01:04:58,680 they're producing sentences with subordinate clauses. 1448 01:04:58,680 --> 01:05:01,380 And so Chomsky said that the way the kids can learn this 1449 01:05:01,380 --> 01:05:03,520 is just too good. 1450 01:05:03,520 --> 01:05:08,310 That it's not possible or reasonable that kids who like, 1451 01:05:08,310 --> 01:05:11,370 still can't manage to operate a spoon without dumping things 1452 01:05:11,370 --> 01:05:15,110 all over themselves, can master language. 1453 01:05:15,110 --> 01:05:17,340 And so Chomsky said that in some way, 1454 01:05:17,340 --> 01:05:21,930 humans have to have a universal grammar, some kind 1455 01:05:21,930 --> 01:05:25,140 of genetically determined module in your brain 1456 01:05:25,140 --> 01:05:29,400 that is wired to look for language. 1457 01:05:29,400 --> 01:05:32,160 To look through the different kinds of input coming in 1458 01:05:32,160 --> 01:05:34,110 and pick up the linguistic bits and slot 1459 01:05:34,110 --> 01:05:37,200 them into kind of certain preexisting patterns 1460 01:05:37,200 --> 01:05:38,760 and learn it. 1461 01:05:42,149 --> 01:05:43,690 This is one of the things that's been 1462 01:05:43,690 --> 01:05:46,870 kind of an ongoing kerfuffle in evolutionary psych 1463 01:05:46,870 --> 01:05:49,810 for the last 40, 50 years. 1464 01:05:49,810 --> 01:05:52,510 Kerfuffle, like a squabble. 1465 01:05:52,510 --> 01:05:55,630 Like lots of people making lots of noise. 1466 01:05:55,630 --> 01:05:57,850 Sometimes by writing nasty papers at each other. 1467 01:06:01,161 --> 01:06:01,910 This is academics. 1468 01:06:01,910 --> 01:06:04,080 It doesn't usually get into actual yelling matches. 1469 01:06:04,080 --> 01:06:06,330 I've heard some very pointed questions at conferences, 1470 01:06:06,330 --> 01:06:08,162 though. 1471 01:06:08,162 --> 01:06:10,000 AUDIENCE: [INAUDIBLE] 1472 01:06:10,000 --> 01:06:10,990 ABBY NOYCE: Yeah. 1473 01:06:10,990 --> 01:06:13,480 He's also like, a liberal political guy, like, 1474 01:06:13,480 --> 01:06:16,510 who does a lot of stuff on that world. 1475 01:06:16,510 --> 01:06:20,369 So people-- it took me a while to figure out 1476 01:06:20,369 --> 01:06:21,910 that that guy and the linguistics guy 1477 01:06:21,910 --> 01:06:22,600 were the same guy. 1478 01:06:22,600 --> 01:06:23,099 They are. 1479 01:06:23,099 --> 01:06:25,646 He's actually here at MIT, Chomsky is. 1480 01:06:25,646 --> 01:06:26,540 AUDIENCE: For real? 1481 01:06:26,540 --> 01:06:27,248 ABBY NOYCE: Yeah. 1482 01:06:27,248 --> 01:06:28,570 I mean probably not now. 1483 01:06:28,570 --> 01:06:29,710 It's 7:00 in the evening. 1484 01:06:29,710 --> 01:06:31,966 But yeah, he works here. 1485 01:06:31,966 --> 01:06:33,340 There's a lot of cool linguistics 1486 01:06:33,340 --> 01:06:35,440 folks here and at Harvard. 1487 01:06:35,440 --> 01:06:38,632 AUDIENCE: [INAUDIBLE] Can we question you? 1488 01:06:38,632 --> 01:06:40,000 Please. 1489 01:06:40,000 --> 01:06:42,292 ABBY NOYCE: So Chomsky said the speakers of a language, 1490 01:06:42,292 --> 01:06:44,249 OK, so that when you speak a language, he said, 1491 01:06:44,249 --> 01:06:45,880 you have to know its vocabulary. 1492 01:06:45,880 --> 01:06:47,352 And learning vocabulary is probably 1493 01:06:47,352 --> 01:06:48,310 pretty straightforward. 1494 01:06:48,310 --> 01:06:49,660 You learn a word, you learn its meaning. 1495 01:06:49,660 --> 01:06:51,493 This is just a matter of sticking this stuff 1496 01:06:51,493 --> 01:06:52,750 into your brain. 1497 01:06:52,750 --> 01:07:00,520 And it's not hugely improbable to think that the sort of-- 1498 01:07:00,520 --> 01:07:03,790 humans are good at all sorts of ways of finding patterns, 1499 01:07:03,790 --> 01:07:07,060 finding meaning, building kind of representations 1500 01:07:07,060 --> 01:07:09,520 of category knowledge and all of this other stuff. 1501 01:07:09,520 --> 01:07:11,860 And it's not to far off to say that we're probably wired 1502 01:07:11,860 --> 01:07:13,270 to be good at finding word-- 1503 01:07:13,270 --> 01:07:17,112 matching word phonetics to meaning information, 1504 01:07:17,112 --> 01:07:19,570 that there might be a critical development period where you 1505 01:07:19,570 --> 01:07:21,520 can just soak this stuff up. 1506 01:07:21,520 --> 01:07:23,740 Kids learn like, 10 words a day between the ages 1507 01:07:23,740 --> 01:07:24,820 of three and six. 1508 01:07:24,820 --> 01:07:28,287 Their vocabulary just skyrockets. 1509 01:07:28,287 --> 01:07:30,370 More interestingly, if you're speaking a language, 1510 01:07:30,370 --> 01:07:33,400 you also have what Chomsky calls its generative grammar. 1511 01:07:33,400 --> 01:07:36,580 A generative grammar is all of the rules 1512 01:07:36,580 --> 01:07:38,700 for how you can put things together 1513 01:07:38,700 --> 01:07:41,680 and how you can put kinds of words together in English. 1514 01:07:41,680 --> 01:07:45,910 So English standards of grammar says that nouns-- 1515 01:07:45,910 --> 01:07:47,740 that subjects should go before verbs, 1516 01:07:47,740 --> 01:07:49,940 the verb should go before objects. 1517 01:07:49,940 --> 01:07:52,910 Propositions should go after the verb and before the thing that 1518 01:07:52,910 --> 01:07:53,450 they're-- 1519 01:07:53,450 --> 01:07:57,899 the noun that they're-- that is the object of the proposition. 1520 01:07:57,899 --> 01:08:00,190 All of these are part of the generative grammar for how 1521 01:08:00,190 --> 01:08:01,504 you build English sentences. 1522 01:08:01,504 --> 01:08:03,670 And of course, nobody thinks about this when they're 1523 01:08:03,670 --> 01:08:04,878 building an English sentence. 1524 01:08:04,878 --> 01:08:06,760 Because if you are a native English speaker, 1525 01:08:06,760 --> 01:08:09,370 you can do this by, this sounds right. 1526 01:08:09,370 --> 01:08:10,925 This does not sound right. 1527 01:08:10,925 --> 01:08:13,300 If I give you a sentence that says something like, Johnny 1528 01:08:13,300 --> 01:08:16,794 cat sat mat, you go, what? 1529 01:08:16,794 --> 01:08:17,710 It doesn't make sense. 1530 01:08:17,710 --> 01:08:18,668 It doesn't sound right. 1531 01:08:18,668 --> 01:08:20,920 You can tell. 1532 01:08:20,920 --> 01:08:22,750 So kids learn both the vocabulary 1533 01:08:22,750 --> 01:08:24,100 and the generative grammar. 1534 01:08:24,100 --> 01:08:26,183 And they do it-- they're clearly not just doing it 1535 01:08:26,183 --> 01:08:27,640 by things that they have heard. 1536 01:08:27,640 --> 01:08:28,569 They're learning the rules. 1537 01:08:28,569 --> 01:08:30,902 So you'll hear little kids saying, they say things like, 1538 01:08:30,902 --> 01:08:32,770 we go to the park. 1539 01:08:32,770 --> 01:08:34,779 They probably did not hear that as an exemplar 1540 01:08:34,779 --> 01:08:36,580 from one of the adults in their life. 1541 01:08:36,580 --> 01:08:38,640 They're probably-- they've learned the add 1542 01:08:38,640 --> 01:08:41,100 an -ed to things to mark the past tense, 1543 01:08:41,100 --> 01:08:42,990 and they're over applying it. 1544 01:08:42,990 --> 01:08:47,920 They'll say-- they'll say things like, we went on an airplane 1545 01:08:47,920 --> 01:08:50,830 and it flied. 1546 01:08:50,830 --> 01:08:52,779 So that they over regularize. 1547 01:08:52,779 --> 01:08:55,200 They take-- or they'll talk about-- 1548 01:08:55,200 --> 01:08:58,479 if I have one mouse, I have one mouse. 1549 01:08:58,479 --> 01:09:01,060 I have two mouses. 1550 01:09:01,060 --> 01:09:02,310 They know how to make plurals. 1551 01:09:02,310 --> 01:09:04,226 They haven't always learned the special cases. 1552 01:09:04,226 --> 01:09:06,438 So English is full of things that aren't-- 1553 01:09:06,438 --> 01:09:08,580 AUDIENCE: Yeah, what do people say besides mouses? 1554 01:09:08,580 --> 01:09:09,901 ABBY NOYCE: Mice. 1555 01:09:09,901 --> 01:09:10,442 AUDIENCE: Oh. 1556 01:09:14,939 --> 01:09:16,439 ABBY NOYCE: So what's going on here? 1557 01:09:16,439 --> 01:09:21,160 So Chomsky says that humans have a universal grammar. 1558 01:09:21,160 --> 01:09:27,410 That we have a language ability in our brain that limits, 1559 01:09:27,410 --> 01:09:29,840 because the possible of like, all imaginable grammars 1560 01:09:29,840 --> 01:09:31,735 is just too big. 1561 01:09:31,735 --> 01:09:33,109 So there are limits to what sorts 1562 01:09:33,109 --> 01:09:36,500 of languages-- what sorts of grammatical structure 1563 01:09:36,500 --> 01:09:38,233 a language can have. 1564 01:09:38,233 --> 01:09:40,399 And so kids who are trying to learn a language, kids 1565 01:09:40,399 --> 01:09:46,010 who are trying to take the input of their environment and slot 1566 01:09:46,010 --> 01:09:51,590 it and figure out how the structure that's underlying it, 1567 01:09:51,590 --> 01:09:54,510 have a head start by knowing what some of the limits are. 1568 01:09:54,510 --> 01:09:55,910 And so they know things. 1569 01:09:55,910 --> 01:10:00,890 And so Chomsky proposes that things like, so-- 1570 01:10:00,890 --> 01:10:02,050 English has a lot of rules. 1571 01:10:02,050 --> 01:10:05,630 You say that subjects go before verbs objects go 1572 01:10:05,630 --> 01:10:08,870 after verbs you have propositions 1573 01:10:08,870 --> 01:10:13,474 that go before the noun that they're describing. 1574 01:10:13,474 --> 01:10:15,890 In Japanese, all of those things are the other way around, 1575 01:10:15,890 --> 01:10:16,880 right? 1576 01:10:16,880 --> 01:10:22,550 Verbs go before subjects, verbs go before objects. 1577 01:10:22,550 --> 01:10:26,150 Japanese has post positions. 1578 01:10:26,150 --> 01:10:27,496 And Chomsky proposes-- 1579 01:10:27,496 --> 01:10:29,120 AUDIENCE: Actually verbs generally go-- 1580 01:10:29,120 --> 01:10:31,102 subjects then object then verb. 1581 01:10:31,102 --> 01:10:32,060 ABBY NOYCE: Then verbs. 1582 01:10:32,060 --> 01:10:33,747 It's subject, object, verb, Yeah. 1583 01:10:33,747 --> 01:10:36,080 And it has like, post positions instead of prepositions, 1584 01:10:36,080 --> 01:10:37,490 where it's the noun and then the thing that's 1585 01:10:37,490 --> 01:10:39,919 describing its relationship to the rest of the sentence. 1586 01:10:39,919 --> 01:10:45,188 AUDIENCE: [INAUDIBLE] 1587 01:10:45,188 --> 01:10:47,104 AUDIENCE: A bit. 1588 01:10:47,104 --> 01:10:49,950 I know a bit of a lot of languages. 1589 01:10:49,950 --> 01:10:52,450 ABBY NOYCE: Yeah, when I was being a bit more of linguistics 1590 01:10:52,450 --> 01:10:54,283 geek than I have been lately, I knew a lot-- 1591 01:10:54,283 --> 01:10:57,190 I had a much better sense of what tended to be 1592 01:10:57,190 --> 01:10:58,150 what word order. 1593 01:10:58,150 --> 01:11:00,070 But anyway, one of the things that Chomsky points out is you 1594 01:11:00,070 --> 01:11:01,220 tend to see patterns. 1595 01:11:01,220 --> 01:11:02,590 So words that have-- 1596 01:11:02,590 --> 01:11:05,290 English is-- languages that are like English the subject, 1597 01:11:05,290 --> 01:11:06,100 object, verb-- 1598 01:11:06,100 --> 01:11:09,430 subject, verb, object, tend to also have prepositions, not 1599 01:11:09,430 --> 01:11:10,570 post positions. 1600 01:11:10,570 --> 01:11:13,034 Languages that are a subject, object, verb, like Japanese, 1601 01:11:13,034 --> 01:11:14,200 tend to have post positions. 1602 01:11:14,200 --> 01:11:16,870 So you can kind of get one big master 1603 01:11:16,870 --> 01:11:20,140 switch that can go to one or the other of a few basic word 1604 01:11:20,140 --> 01:11:21,700 orders, that then tends to direct 1605 01:11:21,700 --> 01:11:24,764 all of these other things. 1606 01:11:24,764 --> 01:11:26,180 How can this be genetically coded? 1607 01:11:26,180 --> 01:11:29,120 Well we're not 100% sure of that. 1608 01:11:29,120 --> 01:11:32,000 There aren't any really solid answers to this. 1609 01:11:32,000 --> 01:11:33,650 But the best one is that this is one 1610 01:11:33,650 --> 01:11:36,200 of those critical development things where you've 1611 01:11:36,200 --> 01:11:38,990 got to have input of a certain kind at a certain stage 1612 01:11:38,990 --> 01:11:42,399 in order to allow the brain to wire itself up properly. 1613 01:11:42,399 --> 01:11:43,940 Remember we talked about what happens 1614 01:11:43,940 --> 01:11:45,731 with vision if you don't get input from one 1615 01:11:45,731 --> 01:11:47,416 eye for a period of time? 1616 01:11:47,416 --> 01:11:49,040 All of the cortex that might originally 1617 01:11:49,040 --> 01:11:51,320 have been going to get input from that eye 1618 01:11:51,320 --> 01:11:54,580 switches allegiance and delivers its input to the other eye. 1619 01:11:54,580 --> 01:11:58,100 Or we talked about if you only see edges in one orientation, 1620 01:11:58,100 --> 01:11:59,827 it doesn't develop. 1621 01:11:59,827 --> 01:12:01,160 So there's some kind of innate-- 1622 01:12:03,890 --> 01:12:08,660 and this is not in any means a 100% agreed upon idea. 1623 01:12:08,660 --> 01:12:10,790 This is a point of controversy, a point 1624 01:12:10,790 --> 01:12:14,240 of ongoing dispute in the linguistics 1625 01:12:14,240 --> 01:12:15,870 and like, language acquisition world. 1626 01:12:20,937 --> 01:12:22,270 Oh I know what else [INAUDIBLE]. 1627 01:12:22,270 --> 01:12:24,800 OK so-- what do we have for time? 1628 01:12:24,800 --> 01:12:25,940 Christ. 1629 01:12:25,940 --> 01:12:28,050 OK, we're good. 1630 01:12:28,050 --> 01:12:30,250 So Hawaii. 1631 01:12:30,250 --> 01:12:32,900 So there's this idea that universal grammar gives you 1632 01:12:32,900 --> 01:12:34,650 some kind of structure in which to slot 1633 01:12:34,650 --> 01:12:35,691 the input you're getting. 1634 01:12:35,691 --> 01:12:39,620 And this makes up for the idea that the input is not 1635 01:12:39,620 --> 01:12:42,530 good enough to give kids enough-- 1636 01:12:42,530 --> 01:12:44,630 to alone give kids enough information 1637 01:12:44,630 --> 01:12:47,700 to figure out the grammar of their language. 1638 01:12:47,700 --> 01:12:49,774 That there's got to be some other information-- 1639 01:12:52,717 --> 01:12:55,050 the input coming in doesn't give kids enough information 1640 01:12:55,050 --> 01:12:57,020 to understand their [INAUDIBLE],, to figure out 1641 01:12:57,020 --> 01:12:58,186 the grammar of the language. 1642 01:12:58,186 --> 01:13:02,330 There's got to be some extra help coming just 1643 01:13:02,330 --> 01:13:04,550 from people's innate abilities. 1644 01:13:04,550 --> 01:13:06,860 So this should be even more extreme 1645 01:13:06,860 --> 01:13:10,500 when the linguistic input has no consistent grammar. 1646 01:13:10,500 --> 01:13:12,840 When the linguistic input is very poor. 1647 01:13:12,840 --> 01:13:16,190 It's not just like the level of kind of poor 1648 01:13:16,190 --> 01:13:18,950 that most people have. 1649 01:13:18,950 --> 01:13:21,560 So Hawaii was-- let's see. 1650 01:13:21,560 --> 01:13:26,000 Hawaii was annexed to the United States in 1898. 1651 01:13:26,000 --> 01:13:29,420 And slightly, about 20 years earlier than that, 1652 01:13:29,420 --> 01:13:31,820 Hawaii's economy had really taken off. 1653 01:13:31,820 --> 01:13:34,640 There'd started being free trade between the US and Hawaii. 1654 01:13:34,640 --> 01:13:37,400 And so you got, as you often do when you see an economy take 1655 01:13:37,400 --> 01:13:38,774 off, you got a pool of immigrants 1656 01:13:38,774 --> 01:13:41,810 from all over the Pacific coming in. 1657 01:13:41,810 --> 01:13:44,060 So you had Americans, you had English speaking people. 1658 01:13:44,060 --> 01:13:46,052 You had folks from China. 1659 01:13:46,052 --> 01:13:47,510 You had folks from the Philippines. 1660 01:13:47,510 --> 01:13:50,210 You had folks from Japan. 1661 01:13:50,210 --> 01:13:53,114 You had folks from, native Hawaiian islanders. 1662 01:13:53,114 --> 01:13:54,530 And so you got all of these people 1663 01:13:54,530 --> 01:13:56,870 with different linguistic backgrounds. 1664 01:13:56,870 --> 01:13:58,750 And what you tend to see happening 1665 01:13:58,750 --> 01:14:03,410 is that they develop a, not quite a real language. 1666 01:14:03,410 --> 01:14:05,420 But what's called a pigeon for communicating in. 1667 01:14:05,420 --> 01:14:09,440 So pigeons tend to have really restricted vocabularies, very 1668 01:14:09,440 --> 01:14:10,940 few words. 1669 01:14:10,940 --> 01:14:15,320 They tend to have very simple syntactic rules. 1670 01:14:15,320 --> 01:14:17,360 They often don't have, for example, a fixed word 1671 01:14:17,360 --> 01:14:21,410 order the way a lot of, not all, but a lot of languages do. 1672 01:14:21,410 --> 01:14:23,720 They don't have any structures for like, recursion, 1673 01:14:23,720 --> 01:14:24,785 for subordinate clauses. 1674 01:14:28,600 --> 01:14:33,280 And speakers of a pigeon often, there aren't like, 1675 01:14:33,280 --> 01:14:36,130 rules about what words are obligatory or are not 1676 01:14:36,130 --> 01:14:37,242 obligatory. 1677 01:14:37,242 --> 01:14:39,200 So you can like, leave out parts of a sentence. 1678 01:14:39,200 --> 01:14:42,334 It's not like English. 1679 01:14:42,334 --> 01:14:43,750 So for example, in English, if you 1680 01:14:43,750 --> 01:14:45,833 want to talk about the weather, you've got to say, 1681 01:14:45,833 --> 01:14:46,952 it's raining. 1682 01:14:46,952 --> 01:14:49,160 It in that sentence doesn't really refer to anything. 1683 01:14:49,160 --> 01:14:51,070 It's just that you can't have raining without the subject 1684 01:14:51,070 --> 01:14:51,820 there before it. 1685 01:14:51,820 --> 01:14:53,740 English requires this. 1686 01:14:53,740 --> 01:14:55,950 Anyone here take Spanish in school? 1687 01:14:55,950 --> 01:15:00,080 How do you say it's raining in Spanish? 1688 01:15:00,080 --> 01:15:00,580 Huh? 1689 01:15:00,580 --> 01:15:02,012 AUDIENCE: Esta lloviendo. 1690 01:15:02,012 --> 01:15:03,160 AUDIENCE: Lloviendo. 1691 01:15:03,160 --> 01:15:04,150 ABBY NOYCE: Yeah but you don't need the-- you 1692 01:15:04,150 --> 01:15:05,274 don't need the esta, right? 1693 01:15:05,274 --> 01:15:07,039 You can just say the verb part. 1694 01:15:07,039 --> 01:15:08,580 Isn't Spanish one of those languages? 1695 01:15:08,580 --> 01:15:10,370 AUDIENCE: Yeah, you don't need a subject. 1696 01:15:10,370 --> 01:15:11,350 ABBY NOYCE: Yeah you don't need the subject part. 1697 01:15:11,350 --> 01:15:13,300 You can just say whatever that verb is. 1698 01:15:13,300 --> 01:15:18,142 AUDIENCE: [INAUDIBLE] like, [INAUDIBLE] 1699 01:15:18,142 --> 01:15:18,850 ABBY NOYCE: Yeah. 1700 01:15:18,850 --> 01:15:21,580 AUDIENCE: [INAUDIBLE] 1701 01:15:21,580 --> 01:15:22,960 ABBY NOYCE: Yeah, OK. 1702 01:15:22,960 --> 01:15:24,640 So Spanish doesn't require subjects 1703 01:15:24,640 --> 01:15:25,641 in the way English does. 1704 01:15:25,641 --> 01:15:27,681 So the words-- the rules for these are different. 1705 01:15:27,681 --> 01:15:29,740 English requires more pieces of the sentence. 1706 01:15:29,740 --> 01:15:32,510 So pigeons often don't have like, this sort of, 1707 01:15:32,510 --> 01:15:33,760 this part of it is obligatory. 1708 01:15:33,760 --> 01:15:36,350 You can leave out whatever piece-- 1709 01:15:36,350 --> 01:15:39,790 whatever pieces you think might make it clear-- 1710 01:15:39,790 --> 01:15:42,590 might already be clear to the person you're talking to. 1711 01:15:42,590 --> 01:15:44,220 So if I was coming into the store, 1712 01:15:44,220 --> 01:15:45,970 you know, and I was trying to buy bananas, 1713 01:15:45,970 --> 01:15:47,553 you might, in English, you might go up 1714 01:15:47,553 --> 01:15:50,350 and say, I want bananas, right? 1715 01:15:50,350 --> 01:15:55,350 If you spoke a pigeon, you could see people just saying bananas. 1716 01:15:55,350 --> 01:15:55,990 You know? 1717 01:15:55,990 --> 01:15:57,610 If you walked into a store and said bananas, 1718 01:15:57,610 --> 01:15:59,260 like, if you walked into Laverde's at the student center 1719 01:15:59,260 --> 01:16:00,879 and said, bananas, they'd probably 1720 01:16:00,879 --> 01:16:02,170 look at you a little bit funny. 1721 01:16:02,170 --> 01:16:03,461 They might get you your banana. 1722 01:16:03,461 --> 01:16:06,952 But this would not be like, a normal conversation 1723 01:16:06,952 --> 01:16:07,660 in this language. 1724 01:16:10,522 --> 01:16:13,460 AUDIENCE: [INAUDIBLE] 1725 01:16:13,460 --> 01:16:14,198 ABBY NOYCE: Yeah? 1726 01:16:14,198 --> 01:16:17,070 AUDIENCE: People tend to leave out as much as possible. 1727 01:16:17,070 --> 01:16:18,820 ABBY NOYCE: Yeah. 1728 01:16:18,820 --> 01:16:21,570 I don't know what the obligatory word rules are in Japanese. 1729 01:16:21,570 --> 01:16:22,080 At all. 1730 01:16:22,080 --> 01:16:24,056 You probably have better sense of this than I. 1731 01:16:24,056 --> 01:16:25,770 AUDIENCE: They can leave out as much as possible. 1732 01:16:25,770 --> 01:16:27,380 They leave out the subjects, objects. 1733 01:16:27,380 --> 01:16:30,870 And sometimes they don't have a verb if the verb is obvious. 1734 01:16:30,870 --> 01:16:31,420 And-- 1735 01:16:31,420 --> 01:16:32,128 ABBY NOYCE: Cool. 1736 01:16:32,128 --> 01:16:36,040 AUDIENCE: [INAUDIBLE] 1737 01:16:36,040 --> 01:16:37,120 ABBY NOYCE: All right. 1738 01:16:37,120 --> 01:16:39,328 So when you get a mixed pool of immigrants like this, 1739 01:16:39,328 --> 01:16:42,209 they tend to develop a kind of a common pigeon, sort 1740 01:16:42,209 --> 01:16:43,000 of a mini language. 1741 01:16:43,000 --> 01:16:44,041 It's not a full language. 1742 01:16:44,041 --> 01:16:46,720 It doesn't have the capabilities that a full language does. 1743 01:16:46,720 --> 01:16:48,130 But you can get by in it. 1744 01:16:48,130 --> 01:16:51,490 So what happens then is if you see 1745 01:16:51,490 --> 01:16:55,140 kids who are brought up in this polyglot community, where 1746 01:16:55,140 --> 01:16:56,882 there's lots of languages being spoken. 1747 01:16:56,882 --> 01:16:58,840 But the only common language, the only language 1748 01:16:58,840 --> 01:17:01,173 they might share with their neighbors or with their kids 1749 01:17:01,173 --> 01:17:02,860 at school, is this pidgin. 1750 01:17:02,860 --> 01:17:05,080 Probably not at school if it's Hawaii in 1898, 1751 01:17:05,080 --> 01:17:07,426 let's be honest. 1752 01:17:07,426 --> 01:17:08,800 So what happens when kids develop 1753 01:17:08,800 --> 01:17:11,410 in this is that they take this pidgin language 1754 01:17:11,410 --> 01:17:13,500 and they expand the vocabulary. 1755 01:17:13,500 --> 01:17:15,250 So you get something up to what, you know, 1756 01:17:15,250 --> 01:17:17,070 a real language, a language that's 1757 01:17:17,070 --> 01:17:20,090 been around for a few hundred years even looks like. 1758 01:17:20,090 --> 01:17:22,570 And they'll-- it'll start having this kind of syntax. 1759 01:17:22,570 --> 01:17:24,430 And so this is called a Creole. 1760 01:17:24,430 --> 01:17:27,950 A Creole is a language that develops from a pidgin, 1761 01:17:27,950 --> 01:17:36,946 but has all of these markers of real languages and real syntax. 1762 01:17:36,946 --> 01:17:40,404 Where are my examples? 1763 01:17:40,404 --> 01:17:43,368 So, where is it? 1764 01:17:52,260 --> 01:17:53,002 So yeah. 1765 01:17:53,002 --> 01:17:54,960 So Creole gets used in some other contexts too. 1766 01:17:54,960 --> 01:17:57,001 But in this case, this is for kinds of languages. 1767 01:17:57,001 --> 01:17:59,525 And you'll see Creoles in a lot of places. 1768 01:17:59,525 --> 01:18:00,900 You'll see Creoles-- and that are 1769 01:18:00,900 --> 01:18:02,108 based on different languages. 1770 01:18:02,108 --> 01:18:05,210 So Hawaiian-- the Hawaiian creole is based on English. 1771 01:18:05,210 --> 01:18:07,710 There are Creoles that are based on French in a lot of like, 1772 01:18:07,710 --> 01:18:10,386 parts of the world the French once colonized. 1773 01:18:10,386 --> 01:18:11,760 There's a lot of creole languages 1774 01:18:11,760 --> 01:18:13,134 that are spoken in the Caribbean. 1775 01:18:13,134 --> 01:18:15,450 If you look at like, the history of these former sugar 1776 01:18:15,450 --> 01:18:16,990 plantation islands. 1777 01:18:16,990 --> 01:18:19,560 Look at like, the slave trade and a bunch of people taken 1778 01:18:19,560 --> 01:18:21,360 who didn't have a common language, 1779 01:18:21,360 --> 01:18:23,190 all dumped together in one place and forced 1780 01:18:23,190 --> 01:18:25,680 to work together and communicate with other people. 1781 01:18:25,680 --> 01:18:27,150 You'll see these pidgins develop. 1782 01:18:27,150 --> 01:18:29,040 Kids grow up in this community and then 1783 01:18:29,040 --> 01:18:32,280 they'll take that pidgin and expand it. 1784 01:18:37,236 --> 01:18:37,736 Go. 1785 01:18:37,736 --> 01:18:38,236 Go back. 1786 01:18:38,236 --> 01:18:39,224 All right. 1787 01:18:47,160 --> 01:18:48,896 AUDIENCE: So if you could take French, be 1788 01:18:48,896 --> 01:18:51,624 able to speak French, bonjour, would you 1789 01:18:51,624 --> 01:18:53,519 be able to understand French creole? 1790 01:18:53,519 --> 01:18:54,560 ABBY NOYCE: Probably not. 1791 01:18:54,560 --> 01:18:55,976 Or you probably-- like what I find 1792 01:18:55,976 --> 01:18:59,470 is that with English based creoles, if I'm reading them, 1793 01:18:59,470 --> 01:19:01,220 like, I'm reading somebody's gloss of one, 1794 01:19:01,220 --> 01:19:02,270 I can usually figure it out. 1795 01:19:02,270 --> 01:19:04,460 If I'm trying to follow somebody who's speaking it, 1796 01:19:04,460 --> 01:19:05,310 it's really hard. 1797 01:19:05,310 --> 01:19:07,640 So it was somewhere between when I was in high school 1798 01:19:07,640 --> 01:19:10,640 and in college, I worked on a farm stand. 1799 01:19:10,640 --> 01:19:13,430 And the guys who, they're like seasonal labor for like, 1800 01:19:13,430 --> 01:19:15,890 all of the farms in my parts of the Southern New Hampshire, 1801 01:19:15,890 --> 01:19:16,939 are Jamaican. 1802 01:19:16,939 --> 01:19:17,480 I don't know. 1803 01:19:17,480 --> 01:19:18,980 There's this whole pool of Jamaican workers 1804 01:19:18,980 --> 01:19:20,720 who come up to-- who come here for the summer, 1805 01:19:20,720 --> 01:19:21,469 work in the farms. 1806 01:19:21,469 --> 01:19:23,649 And they go home in like, November. 1807 01:19:23,649 --> 01:19:25,190 Talking to the Jamaican guys, they're 1808 01:19:25,190 --> 01:19:28,602 seeking something that is based on the same English language 1809 01:19:28,602 --> 01:19:29,810 that my language is based on. 1810 01:19:29,810 --> 01:19:32,630 But it's not the same. 1811 01:19:32,630 --> 01:19:35,416 And it's hard to follow if you're not paying attention. 1812 01:19:35,416 --> 01:19:37,040 So we all slow way down for each other. 1813 01:19:37,040 --> 01:19:38,498 AUDIENCE: I think their creole is-- 1814 01:19:38,498 --> 01:19:40,709 cause it's more [INAUDIBLE] on structural, 1815 01:19:40,709 --> 01:19:42,375 like the way they structure their words, 1816 01:19:42,375 --> 01:19:44,530 like, they'll kind of stop halfway. 1817 01:19:44,530 --> 01:19:45,280 ABBY NOYCE: Right. 1818 01:19:45,280 --> 01:19:47,030 And also like, the phonemes are different. 1819 01:19:47,030 --> 01:19:48,841 Like their phonemes have shifted somewhat. 1820 01:19:48,841 --> 01:19:49,340 Anyway. 1821 01:19:49,340 --> 01:19:51,129 AUDIENCE: [INAUDIBLE] 1822 01:19:51,129 --> 01:19:52,420 ABBY NOYCE: Yeah, I don't know. 1823 01:19:52,420 --> 01:19:53,060 It's been a while. 1824 01:19:53,060 --> 01:19:54,580 I don't know if I could pull out what I think 1825 01:19:54,580 --> 01:19:55,663 are the defining features. 1826 01:19:55,663 --> 01:19:57,550 But I'd recognize it if I heard it. 1827 01:19:57,550 --> 01:19:58,942 But it was definitely like-- 1828 01:19:58,942 --> 01:20:00,650 we just kind of would be like, wait stop. 1829 01:20:00,650 --> 01:20:02,180 Can you say that again? 1830 01:20:02,180 --> 01:20:07,680 AUDIENCE: [INAUDIBLE] but like, after you hear [INAUDIBLE] 1831 01:20:07,680 --> 01:20:08,975 ABBY NOYCE: To some extent. 1832 01:20:08,975 --> 01:20:11,177 AUDIENCE: That's how [INAUDIBLE] 1833 01:20:11,177 --> 01:20:11,885 ABBY NOYCE: Yeah. 1834 01:20:11,885 --> 01:20:12,531 AUDIENCE: I have a lot of Jamaican friends, 1835 01:20:12,531 --> 01:20:13,825 and I speak like that. 1836 01:20:13,825 --> 01:20:16,622 And I do kind of understand them most of the time. 1837 01:20:16,622 --> 01:20:18,580 ABBY NOYCE: You might do more of it than I did. 1838 01:20:18,580 --> 01:20:19,810 I know that, because this was like, 1839 01:20:19,810 --> 01:20:22,119 you'd have just kind of these quick couple sentence 1840 01:20:22,119 --> 01:20:22,660 interchanges. 1841 01:20:22,660 --> 01:20:23,950 Like, because there was the store crew, 1842 01:20:23,950 --> 01:20:25,930 which was mostly high school kids from town. 1843 01:20:25,930 --> 01:20:27,070 And then there was the farm crew, which 1844 01:20:27,070 --> 01:20:28,030 was all these Jamaican guys. 1845 01:20:28,030 --> 01:20:29,696 And these were like, separate castes who 1846 01:20:29,696 --> 01:20:31,340 didn't interact a whole lot. 1847 01:20:31,340 --> 01:20:32,990 It was a little weird. 1848 01:20:32,990 --> 01:20:35,650 But that's kind of my immediate example of interacting closely 1849 01:20:35,650 --> 01:20:38,530 with somebody who's speaking something that is 1850 01:20:38,530 --> 01:20:42,266 based on English, but is not. 1851 01:20:42,266 --> 01:20:43,390 It's a functional language. 1852 01:20:43,390 --> 01:20:44,200 It's got its own rules. 1853 01:20:44,200 --> 01:20:45,580 It's not like it's English that's wrong. 1854 01:20:45,580 --> 01:20:46,955 It's a different language, and it 1855 01:20:46,955 --> 01:20:49,950 sounds just enough like English to kind of trip you up. 1856 01:20:49,950 --> 01:20:50,500 All right. 1857 01:20:50,500 --> 01:20:51,890 So where does this syntax come from? 1858 01:20:51,890 --> 01:20:53,230 So we know that these kids start out 1859 01:20:53,230 --> 01:20:54,896 speaking this pidgin, this language that 1860 01:20:54,896 --> 01:20:59,259 doesn't have syntax, that doesn't have a good vocabulary. 1861 01:20:59,259 --> 01:21:01,300 And if you get kids growing up in this community, 1862 01:21:01,300 --> 01:21:03,220 they have a language that's regularized, 1863 01:21:03,220 --> 01:21:07,200 that has a fixed word order, that has syntax. 1864 01:21:07,200 --> 01:21:08,170 What happened? 1865 01:21:08,170 --> 01:21:12,520 So one possibility is that the various donor languages, like, 1866 01:21:12,520 --> 01:21:14,842 these kids are probably interacting with their parents 1867 01:21:14,842 --> 01:21:16,300 enough to grow up speaking whatever 1868 01:21:16,300 --> 01:21:18,770 language their parents speak. 1869 01:21:18,770 --> 01:21:22,030 And are they just bringing these rules over to the creole? 1870 01:21:22,030 --> 01:21:24,730 But a lot of people have done a lot of analysis of this, 1871 01:21:24,730 --> 01:21:27,400 looking at the creole syntax versus the donor 1872 01:21:27,400 --> 01:21:29,410 languages, the potential languages that 1873 01:21:29,410 --> 01:21:31,120 make up the community. 1874 01:21:31,120 --> 01:21:35,191 And what they found is that they're really not the same. 1875 01:21:35,191 --> 01:21:37,690 The rules that govern the syntax in the creole and the rules 1876 01:21:37,690 --> 01:21:40,840 that govern the syntax in the other languages are different. 1877 01:21:40,840 --> 01:21:44,890 So Derek Bickerton, who's a linguistics guy who 1878 01:21:44,890 --> 01:21:46,750 did a lot of this work on Hawaii, 1879 01:21:46,750 --> 01:21:48,820 in particular because the creole in Hawaii 1880 01:21:48,820 --> 01:21:50,830 developed recently enough that you 1881 01:21:50,830 --> 01:21:52,360 could compare people who are kind 1882 01:21:52,360 --> 01:21:55,810 of in that first generation of people 1883 01:21:55,810 --> 01:21:58,470 who grew up speaking the creole versus people who 1884 01:21:58,470 --> 01:22:00,220 are immigrants at about the same time, 1885 01:22:00,220 --> 01:22:03,670 and can actually compare their language use. 1886 01:22:03,670 --> 01:22:05,260 He did this in the '70s. 1887 01:22:05,260 --> 01:22:07,210 So looking at people from the early-- 1888 01:22:07,210 --> 01:22:09,940 people who were probably in their 70s then. 1889 01:22:09,940 --> 01:22:12,190 And so what Bickerton claims is going 1890 01:22:12,190 --> 01:22:13,930 on here is he says that Creoles-- 1891 01:22:13,930 --> 01:22:16,120 and if you compare Creoles around the world, 1892 01:22:16,120 --> 01:22:18,730 in terms of their syntax, they're more like each other 1893 01:22:18,730 --> 01:22:21,070 than they are like any other group of languages. 1894 01:22:21,070 --> 01:22:22,630 And Bickerton claims that what you're 1895 01:22:22,630 --> 01:22:24,370 seeing when a Creole develops is you're 1896 01:22:24,370 --> 01:22:27,160 seeing kind of the default switches 1897 01:22:27,160 --> 01:22:29,830 for a grammatical structure. 1898 01:22:29,830 --> 01:22:31,490 The default settings. 1899 01:22:31,490 --> 01:22:34,720 So with the input-- without any kind of grammatical input 1900 01:22:34,720 --> 01:22:38,320 coming in for this language, these kids 1901 01:22:38,320 --> 01:22:43,700 default to this particular set of rules and go by that. 1902 01:22:43,700 --> 01:22:46,690 So if you think of the grammar switches like, 1903 01:22:46,690 --> 01:22:48,400 universal grammar is taking the input 1904 01:22:48,400 --> 01:22:49,840 and trying to figure out which-- 1905 01:22:49,840 --> 01:22:53,470 more or less which option it maps onto. 1906 01:22:53,470 --> 01:22:55,637 There's got to be a default setting for that switch. 1907 01:22:55,637 --> 01:22:57,469 And that this is what it is, it's what you-- 1908 01:22:57,469 --> 01:22:59,625 the structure you get when you speak a Creole, when 1909 01:22:59,625 --> 01:23:02,860 you have a Creole. 1910 01:23:02,860 --> 01:23:05,250 So how the brain separates sounds. 1911 01:23:05,250 --> 01:23:06,900 These guys are interested in what's 1912 01:23:06,900 --> 01:23:11,190 called both auditory streaming processes 1913 01:23:11,190 --> 01:23:14,770 and also in concurrent sound segregation. 1914 01:23:14,770 --> 01:23:18,332 So when, in the world, there are always lots of noises going-- 1915 01:23:18,332 --> 01:23:20,790 there are usually lots of noises going on at the same time. 1916 01:23:20,790 --> 01:23:22,710 We know this in class because we usually have the window open 1917 01:23:22,710 --> 01:23:24,900 and traffic on Mass Ave. is noisy, right? 1918 01:23:24,900 --> 01:23:27,780 So you guys have a real time trying to segregate these two-- 1919 01:23:27,780 --> 01:23:29,530 you've got two streams of sound coming in. 1920 01:23:29,530 --> 01:23:31,740 You've got me babbling, which hopefully you 1921 01:23:31,740 --> 01:23:33,570 want to listen to, and you've also got all of the noise 1922 01:23:33,570 --> 01:23:33,930 out there. 1923 01:23:33,930 --> 01:23:35,930 So you're trying to pick out one of these things 1924 01:23:35,930 --> 01:23:37,080 from the other one. 1925 01:23:37,080 --> 01:23:39,384 And these guys are interested, in particular, in the, 1926 01:23:39,384 --> 01:23:40,800 not just what kinds of cues people 1927 01:23:40,800 --> 01:23:44,700 use, but really in what kind of the neuroscience 1928 01:23:44,700 --> 01:23:45,930 underlying this ability is. 1929 01:23:45,930 --> 01:23:47,763 What parts of the brain seem to be involved, 1930 01:23:47,763 --> 01:23:50,170 what kinds of things they do. 1931 01:23:50,170 --> 01:23:55,650 So we're doing the usual thing where some folks presented. 1932 01:23:55,650 --> 01:23:57,930 I also asked everybody to read the whole thing 1933 01:23:57,930 --> 01:24:00,750 and be prepared to ask questions of your classmates. 1934 01:24:03,300 --> 01:24:06,149 Sarah and Wayne, I guess you guys get to follow along. 1935 01:24:06,149 --> 01:24:08,190 And if you have questions, feel free to sing out. 1936 01:24:14,092 --> 01:24:16,050 I'm not quite evil enough to give you a section 1937 01:24:16,050 --> 01:24:17,460 and put you on the spot with it. 1938 01:24:17,460 --> 01:24:19,830 But you should ask-- try and ask questions. 1939 01:24:19,830 --> 01:24:23,100 So first up, what cues do listeners use to segregate 1940 01:24:23,100 --> 01:24:25,420 sounds that Zachariah and Jess. 1941 01:24:25,420 --> 01:24:30,500 And you guys also get that big box on 467. 1942 01:24:30,500 --> 01:24:31,645 AUDIENCE: I'll start. 1943 01:24:31,645 --> 01:24:32,270 ABBY NOYCE: OK. 1944 01:24:32,270 --> 01:24:36,174 AUDIENCE: So auditory streaming is the ability 1945 01:24:36,174 --> 01:24:40,078 to follow one speaker over a period of time [INAUDIBLE] 1946 01:24:40,078 --> 01:24:41,054 other sounds. 1947 01:24:41,054 --> 01:24:43,982 And to study it, they play two tones 1948 01:24:43,982 --> 01:24:45,450 with different frequencies. 1949 01:24:45,450 --> 01:24:46,533 ABBY NOYCE: Ooh, ooh, ooh. 1950 01:24:46,533 --> 01:24:48,540 Want to hear one of these? 1951 01:24:48,540 --> 01:24:50,890 I found the website that they use for this. 1952 01:24:50,890 --> 01:24:53,580 So, bonk, come on. 1953 01:24:53,580 --> 01:24:54,670 I know you switched. 1954 01:24:54,670 --> 01:24:56,590 We're very proud of you. 1955 01:24:56,590 --> 01:24:57,949 So here's a galloping one. 1956 01:24:57,949 --> 01:24:59,740 So these are models of the two frequencies. 1957 01:24:59,740 --> 01:25:02,600 So when the frequencies are close together, 1958 01:25:02,600 --> 01:25:04,810 you get a rhythm that sounds like this. 1959 01:25:07,870 --> 01:25:10,811 Whoa, what is it doing to my like, video app? 1960 01:25:10,811 --> 01:25:11,310 Anyway. 1961 01:25:11,310 --> 01:25:15,320 But you hear how it's like, da-da-dot, da-da-dot. 1962 01:25:15,320 --> 01:25:16,570 OK. 1963 01:25:16,570 --> 01:25:21,930 And if you compare that to-- 1964 01:25:21,930 --> 01:25:23,650 it's the same timing. 1965 01:25:23,650 --> 01:25:25,480 But the frequencies are further apart. 1966 01:25:28,794 --> 01:25:29,960 All right, don't look at it. 1967 01:25:37,990 --> 01:25:38,970 So the [INAUDIBLE]. 1968 01:25:38,970 --> 01:25:41,450 AUDIENCE: [INAUDIBLE] they don't mean the rate at which 1969 01:25:41,450 --> 01:25:41,950 [INAUDIBLE]? 1970 01:25:41,950 --> 01:25:42,574 ABBY NOYCE: No. 1971 01:25:42,574 --> 01:25:46,630 They mean the frequencies of the two independent tones. 1972 01:25:46,630 --> 01:25:49,290 So it's not nearly-- yeah, the pitch of the two tones. 1973 01:25:51,970 --> 01:25:55,900 So the further apart you split them, the less strong 1974 01:25:55,900 --> 01:25:57,550 that galloping effect is. 1975 01:25:57,550 --> 01:25:59,195 If you put them even further apart, 1976 01:25:59,195 --> 01:26:04,040 and that one isn't terribly far apart, you'll even hear-- 1977 01:26:04,040 --> 01:26:05,898 I want to see what these guys have. 1978 01:26:05,898 --> 01:26:07,273 You might-- you'll almost hear it 1979 01:26:07,273 --> 01:26:08,700 as just two concurrent streams. 1980 01:26:08,700 --> 01:26:10,210 And I want to see if this one is further 1981 01:26:10,210 --> 01:26:11,376 apart than the previous one. 1982 01:26:11,376 --> 01:26:14,350 So that was an integrated one versus the same. 1983 01:26:18,800 --> 01:26:21,910 So here the-- so that one, you can parse either way, 1984 01:26:21,910 --> 01:26:24,610 as either be-de-deep, be-de-deep, or one tone 1985 01:26:24,610 --> 01:26:28,420 going beep, beep, beep, and one going bip, bip, bip. 1986 01:26:28,420 --> 01:26:30,870 And you can pull it out into two separate streams. 1987 01:26:30,870 --> 01:26:31,510 All right. 1988 01:26:31,510 --> 01:26:32,040 Sorry Jess. 1989 01:26:32,040 --> 01:26:32,540 Go ahead. 1990 01:26:43,735 --> 01:26:45,715 AUDIENCE: In another study, they found 1991 01:26:45,715 --> 01:26:49,180 that two sounds with the same frequency 1992 01:26:49,180 --> 01:26:52,680 can be listened to separately. 1993 01:26:52,680 --> 01:26:56,592 But when they start at the same time and stop at the same time, 1994 01:26:56,592 --> 01:26:58,080 then you can't tell. 1995 01:27:02,048 --> 01:27:08,000 Um, also, the pitch also helps people to differentiate 1996 01:27:08,000 --> 01:27:10,976 between the two sounds. 1997 01:27:10,976 --> 01:27:14,449 [INAUDIBLE] 1998 01:27:14,449 --> 01:27:16,240 ABBY NOYCE: So pulling out some of the cues 1999 01:27:16,240 --> 01:27:17,656 that people use to split them out. 2000 01:27:17,656 --> 01:27:19,330 So possibly what kind of cues-- 2001 01:27:19,330 --> 01:27:22,330 the neural stuff we'll talk about later in the paper 2002 01:27:22,330 --> 01:27:23,380 should be looking for. 2003 01:27:23,380 --> 01:27:25,231 Cool. 2004 01:27:25,231 --> 01:27:25,730 Zachariah. 2005 01:27:25,730 --> 01:27:26,750 AUDIENCE: [INAUDIBLE] summarize the box. 2006 01:27:26,750 --> 01:27:27,455 ABBY NOYCE: You're on the box. 2007 01:27:27,455 --> 01:27:28,000 All right. 2008 01:27:28,000 --> 01:27:30,770 Tell me about the box. 2009 01:27:30,770 --> 01:27:32,720 AUDIENCE: It mentioned that one way 2010 01:27:32,720 --> 01:27:36,562 that a listener can kind of determine 2011 01:27:36,562 --> 01:27:40,990 if a person's right in front of you versus like, [INAUDIBLE] 2012 01:27:40,990 --> 01:27:47,632 is by the [INAUDIBLE] whether the energy of how 2013 01:27:47,632 --> 01:27:50,830 they get open ears are equal or if they're different. 2014 01:27:50,830 --> 01:27:54,766 So, like if the person's right in front of you, then 2015 01:27:54,766 --> 01:27:59,194 both ears-- well they should get the sound 2016 01:27:59,194 --> 01:28:03,130 with the same amount of volume. 2017 01:28:03,130 --> 01:28:05,884 If the person's over to, say, the left then 2018 01:28:05,884 --> 01:28:08,750 your left ear gets it [INAUDIBLE].. 2019 01:28:08,750 --> 01:28:13,126 So they did an experiment where they had a listener come 2020 01:28:13,126 --> 01:28:14,460 with seven headphones. 2021 01:28:14,460 --> 01:28:18,350 And they played the exact same-- 2022 01:28:18,350 --> 01:28:22,560 I think the exact same sort of pattern into each ear. 2023 01:28:22,560 --> 01:28:27,420 And the person couldn't really-- 2024 01:28:27,420 --> 01:28:30,790 I guess couldn't detect if, [INAUDIBLE] like the sound 2025 01:28:30,790 --> 01:28:32,905 was coming from any particular direction. 2026 01:28:32,905 --> 01:28:33,530 ABBY NOYCE: OK. 2027 01:28:33,530 --> 01:28:36,700 AUDIENCE: And then, what-- and then, in part C, 2028 01:28:36,700 --> 01:28:41,460 it shows that one of the waves is inverted so that where 2029 01:28:41,460 --> 01:28:44,980 the other one is having its peak be-- 2030 01:28:44,980 --> 01:28:48,760 we'll say where the left ear perceives a peak in energy, 2031 01:28:48,760 --> 01:28:53,648 the right ear perceives a bottom value. 2032 01:28:53,648 --> 01:28:55,572 And then, the person's able to tell, 2033 01:28:55,572 --> 01:28:58,565 like, it's coming from a certain direction. 2034 01:28:58,565 --> 01:28:59,190 ABBY NOYCE: OK. 2035 01:29:02,210 --> 01:29:04,532 Tell me about these two graphs. 2036 01:29:04,532 --> 01:29:10,980 AUDIENCE: Um, so, the first graph, 2037 01:29:10,980 --> 01:29:14,452 [INAUDIBLE] I believe the blue represents-- 2038 01:29:28,340 --> 01:29:32,804 sorry, I [INAUDIBLE]. 2039 01:29:32,804 --> 01:29:33,796 ABBY NOYCE: It's OK. 2040 01:29:33,796 --> 01:29:34,788 Take a sec. 2041 01:29:51,156 --> 01:29:55,006 AUDIENCE: OK, so, for the [INAUDIBLE],, 2042 01:29:55,006 --> 01:29:59,810 the horizontal dashed lines represent the point at which 2043 01:29:59,810 --> 01:30:01,020 they-- 2044 01:30:01,020 --> 01:30:06,120 the range away from the mean [INAUDIBLE] outside 2045 01:30:06,120 --> 01:30:08,570 of those two lines, they'll say, OK, 2046 01:30:08,570 --> 01:30:12,110 the subject is now responding to the signal. 2047 01:30:12,110 --> 01:30:17,130 And it shows that when the sounds are inverted, 2048 01:30:17,130 --> 01:30:19,250 so one's playing one way, and the other's playing 2049 01:30:19,250 --> 01:30:21,520 a different way, the subject will 2050 01:30:21,520 --> 01:30:32,472 tend to respond to the signal earlier at a less frequency 2051 01:30:32,472 --> 01:30:34,635 than in the other case when-- 2052 01:30:34,635 --> 01:30:38,070 ABBY NOYCE: At less volume, right? 2053 01:30:38,070 --> 01:30:41,750 Signal level and decibel somethings. 2054 01:30:41,750 --> 01:30:42,250 Yeah. 2055 01:30:42,250 --> 01:30:43,200 AUDIENCE: Volume. 2056 01:30:43,200 --> 01:30:48,845 And then, whereas in the curves, the subject 2057 01:30:48,845 --> 01:30:53,346 has to hear it at a higher volume for when the two waves 2058 01:30:53,346 --> 01:30:57,202 are [INAUDIBLE] energy levels. 2059 01:30:59,830 --> 01:31:00,580 ABBY NOYCE: Right. 2060 01:31:00,580 --> 01:31:05,600 So they're playing it a tone like a single frequency 2061 01:31:05,600 --> 01:31:07,580 tone, right, like a beep. 2062 01:31:07,580 --> 01:31:09,950 And they're also playing like a white noise over it. 2063 01:31:09,950 --> 01:31:13,490 And so what they're showing is that when these two things are 2064 01:31:13,490 --> 01:31:16,230 in sync, when they are in phase at both ears, 2065 01:31:16,230 --> 01:31:19,000 so that the beep is on the upside of a sine wave 2066 01:31:19,000 --> 01:31:26,680 at the same time in both ears, then they're 2067 01:31:26,680 --> 01:31:30,430 graphing the response of neurons in the inferior colliculus. 2068 01:31:30,430 --> 01:31:31,990 And the blue line is the in phase, 2069 01:31:31,990 --> 01:31:33,531 and the red line is the out-of-phase. 2070 01:31:33,531 --> 01:31:36,100 And so if you look at graph D, right, you 2071 01:31:36,100 --> 01:31:40,660 can see that in response to the out-of-phase condition, 2072 01:31:40,660 --> 01:31:46,720 the red crosses that line sooner at a lower volume 2073 01:31:46,720 --> 01:31:47,470 than for the blue. 2074 01:31:47,470 --> 01:31:50,530 You've got to make the tone much louder relative to the noise 2075 01:31:50,530 --> 01:31:53,680 in order to distinguish it when they're in phase. 2076 01:31:53,680 --> 01:31:57,250 And then E is the same thing for a neuron that 2077 01:31:57,250 --> 01:31:59,770 responds the other way, the neuron 2078 01:31:59,770 --> 01:32:02,980 increases its firing rate for both out-of-phase and in phase 2079 01:32:02,980 --> 01:32:03,994 signals. 2080 01:32:03,994 --> 01:32:06,160 Yeah, that's done in honest to goodness Guinea pigs. 2081 01:32:08,750 --> 01:32:11,080 You can't stick electrodes into people. 2082 01:32:11,080 --> 01:32:12,770 But there's a-- the evidence seems 2083 01:32:12,770 --> 01:32:14,270 to be that this is pretty much what 2084 01:32:14,270 --> 01:32:15,730 happens with human neurons. 2085 01:32:15,730 --> 01:32:17,600 Good, cool. 2086 01:32:17,600 --> 01:32:20,690 So there are neurons that respond to pulling a signal out 2087 01:32:20,690 --> 01:32:23,930 of a noise situation. 2088 01:32:23,930 --> 01:32:24,980 Who's up? 2089 01:32:24,980 --> 01:32:30,610 Neural and cognitive bases of auditory streaming. 2090 01:32:30,610 --> 01:32:34,320 And that's Naaman and Natasha, and Vladamir if he was here, 2091 01:32:34,320 --> 01:32:35,000 but he's not. 2092 01:32:35,000 --> 01:32:38,869 So you guys are on it. 2093 01:32:38,869 --> 01:32:39,410 AUDIENCE: OK. 2094 01:32:42,875 --> 01:32:48,815 So basically, most neurons are respond 2095 01:32:48,815 --> 01:32:52,775 to different frequencies of sound. 2096 01:32:52,775 --> 01:32:57,291 And so it basically refers to the [INAUDIBLE] 2097 01:32:57,291 --> 01:33:03,266 on the first page where, 1A, where there's 2098 01:33:03,266 --> 01:33:05,190 two tones that are played. 2099 01:33:05,190 --> 01:33:07,390 It's like the same type of [INAUDIBLE].. 2100 01:33:07,390 --> 01:33:10,286 So when they're very-- 2101 01:33:10,286 --> 01:33:17,400 when the difference between the tones are [INAUDIBLE] together, 2102 01:33:17,400 --> 01:33:21,040 then the neurons that respond to one tone 2103 01:33:21,040 --> 01:33:25,390 will respond quite a lot of to the other tone as well. 2104 01:33:25,390 --> 01:33:29,750 When they're really far apart, then it will respond weakly. 2105 01:33:29,750 --> 01:33:33,040 And if they're in the middle, it will respond a bit. 2106 01:33:33,040 --> 01:33:38,560 And it said, if you speed the sequence up 2107 01:33:38,560 --> 01:33:40,546 for an intermediate-- 2108 01:33:40,546 --> 01:33:46,370 for an intermediate difference, then your neurons 2109 01:33:46,370 --> 01:33:48,260 will tune to one of-- 2110 01:33:51,900 --> 01:33:53,450 your neurons will tune to the tone. 2111 01:33:53,450 --> 01:33:57,400 So it would basically kind of like, 2112 01:33:57,400 --> 01:34:00,440 it would reduce, like, response to one of the tones. 2113 01:34:00,440 --> 01:34:03,523 And all the tones would be more-- 2114 01:34:03,523 --> 01:34:07,190 it would respond more to one of-- like A 2115 01:34:07,190 --> 01:34:12,100 and respond less to B. In my example. 2116 01:34:12,100 --> 01:34:27,812 And, OK. 2117 01:34:27,812 --> 01:34:32,231 [INAUDIBLE] explain the [INAUDIBLE].. 2118 01:34:32,231 --> 01:34:34,740 ABBY NOYCE: No, no, no, no, no, no. 2119 01:34:34,740 --> 01:34:36,260 Yeah. 2120 01:34:36,260 --> 01:34:40,665 AUDIENCE: So that's [INAUDIBLE] continuity illusion, which is, 2121 01:34:40,665 --> 01:34:47,720 basically, if a tone is played in-- 2122 01:34:47,720 --> 01:34:50,088 or say a slide is being played. 2123 01:34:50,088 --> 01:34:56,014 And then slide stops and then replaced by some other noise, 2124 01:34:56,014 --> 01:35:01,270 but there's no gap between the original sound 2125 01:35:01,270 --> 01:35:06,580 and the interrupting one, then even if the original sound 2126 01:35:06,580 --> 01:35:09,020 isn't being [? played ?] behind the interrupting one, 2127 01:35:09,020 --> 01:35:11,960 your brain will kind of just fill it in for you. 2128 01:35:17,948 --> 01:36:09,395 And-- [INAUDIBLE] understand like the other part. 2129 01:36:09,395 --> 01:36:11,770 ABBY NOYCE: All right, so you've got a tone glide, right? 2130 01:36:11,770 --> 01:36:14,560 And a noise that is masking a gap. 2131 01:36:14,560 --> 01:36:18,610 So a tone glide is like, a sound that goes whoop! 2132 01:36:18,610 --> 01:36:20,092 Or woo. 2133 01:36:20,092 --> 01:36:21,050 Or something like that. 2134 01:36:21,050 --> 01:36:22,191 It just slides right down. 2135 01:36:22,191 --> 01:36:23,690 And if you take out the middle of it 2136 01:36:23,690 --> 01:36:25,398 and put just like a burst of white noise, 2137 01:36:25,398 --> 01:36:29,842 like [STATIC SOUND] over it, then what you'll hear usually, 2138 01:36:29,842 --> 01:36:31,300 depending on a few things about it, 2139 01:36:31,300 --> 01:36:33,424 is that you'll hear that the tone glide keeps going 2140 01:36:33,424 --> 01:36:35,000 kind of behind the white noise. 2141 01:36:35,000 --> 01:36:37,460 You'll hear that you're hearing both. 2142 01:36:37,460 --> 01:36:39,970 And so they were looking for neural correlates 2143 01:36:39,970 --> 01:36:43,950 of the continuity illusion which is neurons that would keep-- 2144 01:36:43,950 --> 01:36:45,450 that would respond to the tone glide 2145 01:36:45,450 --> 01:36:49,040 and keep responding behind the noise masker. 2146 01:36:49,040 --> 01:36:54,096 And they say there isn't a lot of evidence for this. 2147 01:36:54,096 --> 01:37:03,450 And the thing with the behavioral response 2148 01:37:03,450 --> 01:37:08,060 is when you hear that tone glide, that tone continuity, 2149 01:37:08,060 --> 01:37:13,022 it depends on the noise that is masking it having 2150 01:37:13,022 --> 01:37:15,230 part of that noise being in the frequency region that 2151 01:37:15,230 --> 01:37:18,290 would have been in the glide that you 2152 01:37:18,290 --> 01:37:19,800 think you keep hearing. 2153 01:37:19,800 --> 01:37:21,120 If that made sense. 2154 01:37:21,120 --> 01:37:24,600 So that the white noise mask has got to have frequencies, 2155 01:37:24,600 --> 01:37:27,870 it's got to include energy in those frequencies. 2156 01:37:27,870 --> 01:37:30,130 And if it doesn't, if it's missing those frequencies 2157 01:37:30,130 --> 01:37:32,570 in the noise, then you won't hear the continuous tone 2158 01:37:32,570 --> 01:37:33,150 behind it. 2159 01:37:33,150 --> 01:37:36,860 So it depends on properties of the masking sound. 2160 01:37:39,440 --> 01:37:44,440 And so they talk about one study, Sugita, Sugeeta, 2161 01:37:44,440 --> 01:37:45,620 I don't know. 2162 01:37:45,620 --> 01:37:49,160 Who studies-- who is looking for neurons that respond to this. 2163 01:37:49,160 --> 01:37:55,090 And theirs does not depend on the frequency-- 2164 01:37:55,090 --> 01:37:58,340 the noise having the right frequency region. 2165 01:37:58,340 --> 01:38:01,364 So it may or may not be a neural correlate of this region. 2166 01:38:01,364 --> 01:38:02,780 And there's another one who thinks 2167 01:38:02,780 --> 01:38:08,284 they have found it because they do have that frequency content. 2168 01:38:08,284 --> 01:38:10,188 AUDIENCE: What does it mean by [INAUDIBLE]?? 2169 01:38:14,010 --> 01:38:15,630 ABBY NOYCE: All set? 2170 01:38:15,630 --> 01:38:16,380 All right. 2171 01:38:16,380 --> 01:38:17,380 Is it your turn to talk? 2172 01:38:17,380 --> 01:38:18,225 AUDIENCE: Yes. 2173 01:38:18,225 --> 01:38:18,850 ABBY NOYCE: OK. 2174 01:38:18,850 --> 01:38:21,295 AUDIENCE: So they were trying to see 2175 01:38:21,295 --> 01:38:23,251 if the streaming was occurring. 2176 01:38:23,251 --> 01:38:26,680 So they decided to use an [INAUDIBLE] 2177 01:38:26,680 --> 01:38:33,044 So that measures neural responses to sound sequences. 2178 01:38:33,044 --> 01:38:40,448 And they're looking for the mismatched [? activity ?] 2179 01:38:40,448 --> 01:38:40,948 which-- 2180 01:38:44,406 --> 01:38:47,864 so if you hear [INAUDIBLE] sounds, which 2181 01:38:47,864 --> 01:38:56,096 is a short sequence of sounds, then [INAUDIBLE] oh yeah, 2182 01:38:56,096 --> 01:39:00,340 they presented it in the sequence of other [INAUDIBLE].. 2183 01:39:00,340 --> 01:39:01,090 ABBY NOYCE: Right. 2184 01:39:01,090 --> 01:39:03,464 So what would an example of like, a sequence of standards 2185 01:39:03,464 --> 01:39:07,155 with one deviant sound be then? 2186 01:39:07,155 --> 01:39:08,580 AUDIENCE: They're [INAUDIBLE]. 2187 01:39:11,430 --> 01:39:12,855 So is that like-- 2188 01:39:12,855 --> 01:39:14,330 ABBY NOYCE: That's the-- so they're 2189 01:39:14,330 --> 01:39:17,410 using for like a stimulus here is like beep beep beep boop 2190 01:39:17,410 --> 01:39:18,321 beep beep beep beep. 2191 01:39:18,321 --> 01:39:20,570 Or you've got a string of sounds that are all the same 2192 01:39:20,570 --> 01:39:22,250 and then one is different. 2193 01:39:22,250 --> 01:39:24,350 And then what they're looking at, 2194 01:39:24,350 --> 01:39:26,330 remember EG is measuring electrical activity 2195 01:39:26,330 --> 01:39:27,590 in the brain, right. 2196 01:39:27,590 --> 01:39:32,060 So this mismatched negativity is a negative wave. 2197 01:39:32,060 --> 01:39:35,060 So it's the electrical activity in the brain goes-- 2198 01:39:35,060 --> 01:39:39,590 has a distinct negative pattern at a certain point in time 2199 01:39:39,590 --> 01:39:45,403 after that deviant tone is presented. 2200 01:39:45,403 --> 01:39:46,882 AUDIENCE: All right. 2201 01:39:46,882 --> 01:39:54,277 And then, they asked the patient, or, yeah, [INAUDIBLE] 2202 01:39:54,277 --> 01:39:57,048 to identify a deviant. 2203 01:39:57,048 --> 01:40:03,470 And they thought that if they measured the MMN to a deviance, 2204 01:40:03,470 --> 01:40:08,430 they could see if there's a string. 2205 01:40:08,430 --> 01:40:22,842 So they found that the MMN one [INAUDIBLE] successive tone, 2206 01:40:22,842 --> 01:40:25,350 so the [INAUDIBLE] like, continues. 2207 01:40:25,350 --> 01:40:26,100 ABBY NOYCE: Right. 2208 01:40:29,140 --> 01:40:32,230 AUDIENCE: Suggested that there was streaming when-- 2209 01:40:32,230 --> 01:40:34,619 when was that 100 MS? 2210 01:40:34,619 --> 01:40:35,660 ABBY NOYCE: Milliseconds. 2211 01:40:35,660 --> 01:40:36,618 AUDIENCE: Milliseconds. 2212 01:40:36,618 --> 01:40:41,240 So but when the SRA was 750 milliseconds, 2213 01:40:41,240 --> 01:40:43,882 they thought that the [INAUDIBLE] should be weaker 2214 01:40:43,882 --> 01:40:47,368 and they found there's no MMN. 2215 01:40:47,368 --> 01:40:54,340 So thought that these differences in the MMN was more 2216 01:40:54,340 --> 01:40:56,830 differences in streaming. 2217 01:40:56,830 --> 01:41:01,312 So they thought that this was occurring outside their focus 2218 01:41:01,312 --> 01:41:04,671 of attention. 2219 01:41:04,671 --> 01:41:05,296 ABBY NOYCE: OK. 2220 01:41:08,782 --> 01:41:14,350 AUDIENCE: And then more recently, they 2221 01:41:14,350 --> 01:41:20,068 thought that the MMN could be affecting eye attention. 2222 01:41:20,068 --> 01:41:28,534 So they're not sure that people can be like, [INAUDIBLE] 2223 01:41:28,534 --> 01:41:31,024 actually ignoring their instructions 2224 01:41:31,024 --> 01:41:34,012 and they're actually listening to sounds. 2225 01:41:34,012 --> 01:41:40,774 But [INAUDIBLE]. 2226 01:41:40,774 --> 01:41:41,482 ABBY NOYCE: Yeah. 2227 01:41:41,482 --> 01:41:43,474 Fair enough. 2228 01:41:43,474 --> 01:41:44,470 Questions? 2229 01:41:56,440 --> 01:41:58,620 OK. 2230 01:41:58,620 --> 01:42:00,130 Ladies? 2231 01:42:00,130 --> 01:42:04,780 AUDIENCE: [INAUDIBLE] OK. 2232 01:42:04,780 --> 01:42:08,440 So they talked about like, affects of streaming 2233 01:42:08,440 --> 01:42:09,880 on [INAUDIBLE]. 2234 01:42:09,880 --> 01:42:11,940 So this guy Jones and his colleagues 2235 01:42:11,940 --> 01:42:15,248 had like, a bunch of people try to recall like, visually 2236 01:42:15,248 --> 01:42:19,012 presented [INAUDIBLE] while listening to a stream of tones. 2237 01:42:19,012 --> 01:42:21,520 And they had like-- 2238 01:42:21,520 --> 01:42:23,805 OK, so the [INAUDIBLE] was disrupted only 2239 01:42:23,805 --> 01:42:25,380 by tones that changed over time. 2240 01:42:25,380 --> 01:42:27,503 So like, if they listened to this one stream that 2241 01:42:27,503 --> 01:42:29,836 would have like, two different beeps, like, alternating. 2242 01:42:29,836 --> 01:42:31,129 Or not alternating. 2243 01:42:31,129 --> 01:42:31,670 I don't know. 2244 01:42:31,670 --> 01:42:33,830 So if they, like, if it changed, they would be like, 2245 01:42:33,830 --> 01:42:34,996 oh well, that's distracting. 2246 01:42:34,996 --> 01:42:36,889 But like if they're just like one tone that 2247 01:42:36,889 --> 01:42:38,472 was like repeated over and over again, 2248 01:42:38,472 --> 01:42:40,180 they could just kind of like tune it out. 2249 01:42:40,180 --> 01:42:41,860 And it wouldn't really both them. 2250 01:42:41,860 --> 01:42:43,892 But like, it also like, well like, 2251 01:42:43,892 --> 01:42:45,600 if it listed two different streams, like, 2252 01:42:45,600 --> 01:42:47,618 each with their own repeated tone, 2253 01:42:47,618 --> 01:42:51,090 it still doesn't bother them, even though [INAUDIBLE].. 2254 01:42:51,090 --> 01:42:52,082 ABBY NOYCE: OK. 2255 01:42:52,082 --> 01:42:59,026 AUDIENCE: And OK. 2256 01:42:59,026 --> 01:43:02,498 And [INAUDIBLE]. 2257 01:43:02,498 --> 01:43:03,490 OK. 2258 01:43:03,490 --> 01:43:08,450 And then [INAUDIBLE] 2259 01:43:08,450 --> 01:43:10,930 AUDIENCE: Then the, for manipulating attention 2260 01:43:10,930 --> 01:43:12,770 during the [INAUDIBLE] of streaming, 2261 01:43:12,770 --> 01:43:17,237 the main focus was that a tone sequence has 2262 01:43:17,237 --> 01:43:21,536 a similar stream, like, beep, boop, beep, like what we said. 2263 01:43:21,536 --> 01:43:23,800 After several seconds, it could eventually 2264 01:43:23,800 --> 01:43:26,050 turn into two streams. 2265 01:43:26,050 --> 01:43:28,970 So you could hear like, it separating, and it 2266 01:43:28,970 --> 01:43:30,220 would be like, beep beep beep. 2267 01:43:30,220 --> 01:43:34,380 And then boop, boop, boop-- like two frequencies. 2268 01:43:34,380 --> 01:43:38,760 And in one condition, or test, there 2269 01:43:38,760 --> 01:43:43,770 was a 20 second sequence of repeating ABA triplets. 2270 01:43:43,770 --> 01:43:46,380 So two different frequencies were heard. 2271 01:43:46,380 --> 01:43:50,040 And this was presented to subjects' left ears. 2272 01:43:50,040 --> 01:43:52,080 And they had to tell the researchers 2273 01:43:52,080 --> 01:43:54,560 how many streams they heard. 2274 01:43:54,560 --> 01:43:57,806 And then in the second test, the subjects 2275 01:43:57,806 --> 01:44:02,110 performed a series of tasks on a series of noise bursts 2276 01:44:02,110 --> 01:44:04,510 in their right ear during the first 10 seconds-- 2277 01:44:04,510 --> 01:44:06,010 ABBY NOYCE: So what kind of stimulus 2278 01:44:06,010 --> 01:44:07,455 do they mean by a noise burst? 2279 01:44:07,455 --> 01:44:10,766 AUDIENCE: Like a sudden? 2280 01:44:10,766 --> 01:44:12,890 ABBY NOYCE: Yeah like, you know, like white noises. 2281 01:44:12,890 --> 01:44:15,210 So it's like a random assortment of-- 2282 01:44:15,210 --> 01:44:16,590 it's like [STATIC SOUND]. 2283 01:44:16,590 --> 01:44:18,840 It's just a static noise is usually 2284 01:44:18,840 --> 01:44:20,430 what they mean for that. 2285 01:44:20,430 --> 01:44:24,730 So this isn't just like noise in the casual use of like a sound. 2286 01:44:24,730 --> 01:44:28,220 It's like a particular kind of thing. 2287 01:44:28,220 --> 01:44:32,350 AUDIENCE: So this was presented in their right ear 2288 01:44:32,350 --> 01:44:34,195 during the first 10 seconds of the sequence. 2289 01:44:34,195 --> 01:44:36,810 So then they switched the tones to the left ear. 2290 01:44:36,810 --> 01:44:40,650 And they had to make their own opinion on what they heard. 2291 01:44:40,650 --> 01:44:49,545 And the researchers found that the streaming was as if the-- 2292 01:44:49,545 --> 01:44:53,220 it was the start of the attended sequence, 2293 01:44:53,220 --> 01:44:56,615 like, I don't know what that really meant. 2294 01:44:59,120 --> 01:45:00,620 ABBY NOYCE: Did you guys notice when 2295 01:45:00,620 --> 01:45:02,910 we listened to the galloping streams, 2296 01:45:02,910 --> 01:45:04,980 if you listen to the A and B one. 2297 01:45:04,980 --> 01:45:05,880 I'll play it again. 2298 01:45:05,880 --> 01:45:07,350 And I'm going to apologize for the way the screen's going 2299 01:45:07,350 --> 01:45:08,435 to get all wonky. 2300 01:45:08,435 --> 01:45:10,470 It starts out, and you can almost hear-- 2301 01:45:20,029 --> 01:45:21,530 So sometimes with stuff like this, 2302 01:45:21,530 --> 01:45:24,430 it will start out sounding like a gallop, like beep beep beep. 2303 01:45:24,430 --> 01:45:26,060 And it will eventually, as you keep listening to it, 2304 01:45:26,060 --> 01:45:28,101 start to separate so that you eventually-- later, 2305 01:45:28,101 --> 01:45:30,510 you hear it as two separate streams, a low stream 2306 01:45:30,510 --> 01:45:32,061 and a high stream. 2307 01:45:32,061 --> 01:45:33,810 And I don't know if anyone perceives that. 2308 01:45:33,810 --> 01:45:35,435 But it's one pattern that you tend to-- 2309 01:45:35,435 --> 01:45:38,040 people tend to see with these kinds of tones. 2310 01:45:38,040 --> 01:45:40,960 And so they're saying-- so that's what they're looking at. 2311 01:45:40,960 --> 01:45:45,012 So having subjects and they're comparing this kind of. 2312 01:45:45,012 --> 01:45:47,220 So when you first hear it, you hear it as one stream. 2313 01:45:47,220 --> 01:45:49,800 And you later split it into two streams. 2314 01:45:49,800 --> 01:45:52,550 And they wanted to see if they-- 2315 01:45:52,550 --> 01:45:54,419 so if they play just the tones, then they 2316 01:45:54,419 --> 01:45:55,710 ask people what they perceived. 2317 01:45:55,710 --> 01:45:57,720 And they said, OK, we hear this one stream at first and later 2318 01:45:57,720 --> 01:45:58,362 as two. 2319 01:45:58,362 --> 01:45:59,820 And then they played them the tones 2320 01:45:59,820 --> 01:46:02,195 at the same time, asked them to spend 10 seconds thinking 2321 01:46:02,195 --> 01:46:04,350 about noise bursts in the other ear 2322 01:46:04,350 --> 01:46:06,240 and then shift their attention to the tones. 2323 01:46:06,240 --> 01:46:08,130 And the tones had been there all along. 2324 01:46:08,130 --> 01:46:09,120 But they weren't attending to them 2325 01:46:09,120 --> 01:46:11,370 because they had to do this other noise discrimination 2326 01:46:11,370 --> 01:46:13,560 task. 2327 01:46:13,560 --> 01:46:16,134 and then they wanted to see if streaming-- 2328 01:46:16,134 --> 01:46:17,550 and their theory was, if streaming 2329 01:46:17,550 --> 01:46:18,840 happens without attention, than it 2330 01:46:18,840 --> 01:46:21,300 should sound like it would've 10 seconds into the control 2331 01:46:21,300 --> 01:46:22,997 condition. 2332 01:46:22,997 --> 01:46:23,580 But it didn't. 2333 01:46:23,580 --> 01:46:26,820 It sounded like when they first started hearing the tones. 2334 01:46:26,820 --> 01:46:29,100 So they're saying that this means attention is not 2335 01:46:29,100 --> 01:46:29,814 happening. 2336 01:46:29,814 --> 01:46:31,230 When you're doing-- that streaming 2337 01:46:31,230 --> 01:46:34,140 is not happening when you're not paying attention to it. 2338 01:46:34,140 --> 01:46:36,000 Does that make sense? 2339 01:46:36,000 --> 01:46:36,739 OK. 2340 01:46:36,739 --> 01:46:40,232 AUDIENCE: And then in a later study, 2341 01:46:40,232 --> 01:46:43,725 researchers found that streaming can be disrupted even 2342 01:46:43,725 --> 01:46:47,218 by non-auditory competing tasks. 2343 01:46:47,218 --> 01:46:50,320 And the results show that attention 2344 01:46:50,320 --> 01:46:53,664 can have a strong influence on auditory streaming. 2345 01:46:53,664 --> 01:46:56,144 And some streaming can occur without full attention. 2346 01:46:56,144 --> 01:46:59,864 But it can be strongly affected when subjects have 2347 01:46:59,864 --> 01:47:02,959 to perform a demanding task. 2348 01:47:02,959 --> 01:47:03,584 ABBY NOYCE: OK. 2349 01:47:07,056 --> 01:47:08,070 AUDIENCE: Yup. 2350 01:47:08,070 --> 01:47:11,360 ABBY NOYCE: Can you paraphrase that in your own words Jen? 2351 01:47:11,360 --> 01:47:16,400 AUDIENCE: So even though auditory streaming 2352 01:47:16,400 --> 01:47:22,110 can be affected by attention, some streaming 2353 01:47:22,110 --> 01:47:24,618 can occur without your full attention. 2354 01:47:27,606 --> 01:47:32,760 But the streaming can be affected 2355 01:47:32,760 --> 01:47:36,460 while you're doing like, a really-- a demanding task 2356 01:47:36,460 --> 01:47:39,384 that's competing with your senses, I guess. 2357 01:47:39,384 --> 01:47:40,300 ABBY NOYCE: Good yeah. 2358 01:47:40,300 --> 01:47:42,190 Cool. 2359 01:47:42,190 --> 01:47:44,570 Non auditory areas and auditory streaming. 2360 01:47:44,570 --> 01:47:47,354 AUDIENCE: Well this book basically said that like, 2361 01:47:47,354 --> 01:47:50,140 [INAUDIBLE] is not [INAUDIBLE] like auditory processing 2362 01:47:50,140 --> 01:47:52,368 could still play a role in streaming. 2363 01:47:52,368 --> 01:47:55,770 Some people who had [INAUDIBLE] can't see things 2364 01:47:55,770 --> 01:47:58,200 on your left side so well. 2365 01:47:58,200 --> 01:47:59,820 Like, they also had reduced streaming 2366 01:47:59,820 --> 01:48:02,400 for like, the sounds in their left ear. 2367 01:48:02,400 --> 01:48:04,400 And, like, the same thing happened 2368 01:48:04,400 --> 01:48:08,592 when like, different [INAUDIBLE] one specific area. 2369 01:48:08,592 --> 01:48:15,578 And it also said like, the intraparietal sulcus, the IPS, 2370 01:48:15,578 --> 01:48:20,568 [INAUDIBLE] with perception of two streams. 2371 01:48:20,568 --> 01:48:23,562 So yeah. 2372 01:48:23,562 --> 01:48:28,552 So it's [INAUDIBLE] IPS like changed depending on-- 2373 01:48:28,552 --> 01:48:31,228 OK, so like, they would play, like, they would have beeps. 2374 01:48:31,228 --> 01:48:32,978 And then sometimes, they would play, like, 2375 01:48:32,978 --> 01:48:35,908 I guess one stream of the same beat or like, 2376 01:48:35,908 --> 01:48:37,320 two streams of the same beep. 2377 01:48:37,320 --> 01:48:38,153 ABBY NOYCE: It was-- 2378 01:48:38,153 --> 01:48:39,829 I think it was all the same sequence. 2379 01:48:39,829 --> 01:48:42,120 But your perception of whether it was one stream or two 2380 01:48:42,120 --> 01:48:43,770 separate streams would change. 2381 01:48:43,770 --> 01:48:45,390 So you'd either hear it as a gallop 2382 01:48:45,390 --> 01:48:48,010 or as a high stream and a slow stream. 2383 01:48:48,010 --> 01:48:49,664 And this sometimes can be, you know, 2384 01:48:49,664 --> 01:48:51,330 like looking at a Necker Cube, and it'll 2385 01:48:51,330 --> 01:48:53,379 pop back and forth between-- 2386 01:48:53,379 --> 01:48:55,170 Necker Cube is that wire frame cube, right, 2387 01:48:55,170 --> 01:48:56,850 that can look either way, depending 2388 01:48:56,850 --> 01:48:58,764 on how you're looking at it. 2389 01:48:58,764 --> 01:49:00,097 You know what I'm talking about? 2390 01:49:00,097 --> 01:49:01,380 AUDIENCE: Oh like the boxing-- 2391 01:49:01,380 --> 01:49:03,505 ABBY NOYCE: The boxing cube that's just the outline 2392 01:49:03,505 --> 01:49:05,400 and it can pop either this way or that way. 2393 01:49:05,400 --> 01:49:05,900 Right. 2394 01:49:05,900 --> 01:49:06,750 So this can be like that. 2395 01:49:06,750 --> 01:49:08,550 Where you can either hear it just one stream or as two 2396 01:49:08,550 --> 01:49:08,850 streams. 2397 01:49:08,850 --> 01:49:10,266 And sometimes, it'll spontaneously 2398 01:49:10,266 --> 01:49:12,510 shift back and forth. 2399 01:49:12,510 --> 01:49:14,244 So and they're looking at whether-- 2400 01:49:14,244 --> 01:49:15,660 and they're saying that activation 2401 01:49:15,660 --> 01:49:19,020 in the inner parietal sulcus, the IPS, 2402 01:49:19,020 --> 01:49:21,390 depends on what perception people are having. 2403 01:49:21,390 --> 01:49:23,940 So they found a piece of the brain who's activation actually 2404 01:49:23,940 --> 01:49:26,430 matches with like, self-reported perception of one or two 2405 01:49:26,430 --> 01:49:27,990 streams. 2406 01:49:27,990 --> 01:49:30,320 Which is cool. 2407 01:49:30,320 --> 01:49:32,180 Does that make sense? 2408 01:49:32,180 --> 01:49:34,450 Sort of. 2409 01:49:34,450 --> 01:49:37,070 So they're just playing beep boop beep boop, or probably as 2410 01:49:37,070 --> 01:49:37,570 a gallop. 2411 01:49:37,570 --> 01:49:40,790 So like beep beep beep, beep beep beep, beep beep beep. 2412 01:49:40,790 --> 01:49:43,280 And people can hear that either as one 2413 01:49:43,280 --> 01:49:45,080 galloping stream-- da-duh-da. 2414 01:49:45,080 --> 01:49:47,160 Or as two separate streams, one going beep, beep, 2415 01:49:47,160 --> 01:49:49,490 and one going beep-beep-beep-beep-beep. 2416 01:49:49,490 --> 01:49:52,220 Even though it's the same input, right? 2417 01:49:52,220 --> 01:49:53,855 And so what they found is that-- 2418 01:49:53,855 --> 01:49:55,521 so what you might expect to see is that, 2419 01:49:55,521 --> 01:49:57,530 so there's some neurons that would respond 2420 01:49:57,530 --> 01:49:59,300 only to single but not to two streams, 2421 01:49:59,300 --> 01:50:00,830 or vise versa, something in your brain that 2422 01:50:00,830 --> 01:50:03,246 actually changes depending on which way you are perceiving 2423 01:50:03,246 --> 01:50:05,170 it. 2424 01:50:05,170 --> 01:50:06,640 And they found one. 2425 01:50:06,640 --> 01:50:08,960 So that the pattern of firing in this particular part 2426 01:50:08,960 --> 01:50:12,075 of your brain changes, not because the input changed, 2427 01:50:12,075 --> 01:50:13,950 but because how you're perceiving it changed. 2428 01:50:13,950 --> 01:50:16,220 Or probably causes how you're perceiving it to change, 2429 01:50:16,220 --> 01:50:18,127 might be a better way of thinking about that. 2430 01:50:18,127 --> 01:50:20,210 That it's picking out a different kind of pattern. 2431 01:50:23,160 --> 01:50:23,680 Yeah. 2432 01:50:23,680 --> 01:50:25,555 Brains are like pattern recognition machines. 2433 01:50:25,555 --> 01:50:26,530 It's what we do. 2434 01:50:26,530 --> 01:50:28,080 OK.