1 00:00:00,090 --> 00:00:02,490 The following content is provided under a Creative 2 00:00:02,490 --> 00:00:04,030 Commons license. 3 00:00:04,030 --> 00:00:06,330 Your support will help MIT OpenCourseWare 4 00:00:06,330 --> 00:00:10,720 continue to offer high-quality educational resources for free. 5 00:00:10,720 --> 00:00:13,320 To make a donation or view additional materials 6 00:00:13,320 --> 00:00:17,280 from hundreds of MIT courses, visit MIT OpenCourseWare 7 00:00:17,280 --> 00:00:18,450 at ocw.mit.edu. 8 00:00:21,450 --> 00:00:27,960 ABBY NOYCE: So going over quickly what we did on Tuesday. 9 00:00:27,960 --> 00:00:31,230 So we talked about the fact that this long-term-- 10 00:00:31,230 --> 00:00:33,540 hanging onto long-term information, long-term memory. 11 00:00:33,540 --> 00:00:36,780 Learning of new facts seems to depend on changes 12 00:00:36,780 --> 00:00:39,120 in how strong a synapse is. 13 00:00:39,120 --> 00:00:41,250 So we know that when you have two cells that 14 00:00:41,250 --> 00:00:46,210 are joined into the synapse, then 15 00:00:46,210 --> 00:00:49,054 if the first cell fires, then it releases 16 00:00:49,054 --> 00:00:50,970 a neurotransmitter that changes the likelihood 17 00:00:50,970 --> 00:00:52,624 of the second cell firing. 18 00:00:52,624 --> 00:00:54,165 So you can think of a strong synapse, 19 00:00:54,165 --> 00:00:56,250 a strong excitatory synapse is one 20 00:00:56,250 --> 00:00:59,730 where if the first cell fires, the second cell is really 21 00:00:59,730 --> 00:01:03,330 likely to fire, whereas a weaker excitatory synapse would be one 22 00:01:03,330 --> 00:01:05,400 where if the first cell fires, the second cell is 23 00:01:05,400 --> 00:01:09,200 only a little bit more likely to fire than it was before. 24 00:01:09,200 --> 00:01:11,484 And we talked in some detail about one mechanism 25 00:01:11,484 --> 00:01:12,900 that seems to underlie this, which 26 00:01:12,900 --> 00:01:15,540 is long-term potentiation, which happens 27 00:01:15,540 --> 00:01:18,900 at these excitatory glutamatergic synapses. 28 00:01:18,900 --> 00:01:21,930 Depends on two types of receptors, 29 00:01:21,930 --> 00:01:25,020 one which is the AMPA receptors, which 30 00:01:25,020 --> 00:01:29,070 are a sodium channel, an ionotropic sodium 31 00:01:29,070 --> 00:01:32,220 channel, and which will open and allow sodium 32 00:01:32,220 --> 00:01:36,150 in any time that glutamate is released into a synapse. 33 00:01:36,150 --> 00:01:40,652 And then these pickier NMDA receptors. 34 00:01:40,652 --> 00:01:41,360 That's backwards. 35 00:01:41,360 --> 00:01:42,480 I'm sorry. 36 00:01:42,480 --> 00:01:46,170 And NMDA receptors only open when 37 00:01:46,170 --> 00:01:50,790 the potential across the membrane at the synapse 38 00:01:50,790 --> 00:01:53,310 depolarizes to about minus 35 millivolts. 39 00:01:53,310 --> 00:01:55,620 So when there's a lot of sodium already coming in, 40 00:01:55,620 --> 00:01:59,620 then the NMDA receptors open up. 41 00:01:59,620 --> 00:02:01,620 And they also allow in as well as sodium, 42 00:02:01,620 --> 00:02:07,533 what other ion that causes long-term changes in the cell? 43 00:02:07,533 --> 00:02:08,380 AUDIENCE: Calcium. 44 00:02:08,380 --> 00:02:09,764 ABBY NOYCE: Calcium. 45 00:02:09,764 --> 00:02:10,970 AUDIENCE: Calcium. 46 00:02:10,970 --> 00:02:12,400 ABBY NOYCE: Calcium. 47 00:02:12,400 --> 00:02:12,900 Right. 48 00:02:12,900 --> 00:02:16,225 Remember, calcium acts as a second messenger, 49 00:02:16,225 --> 00:02:18,600 and it triggers the activity of all these other proteins. 50 00:02:21,840 --> 00:02:25,960 So summing up what we talked about with memory. 51 00:02:25,960 --> 00:02:29,940 So we know that when you experience something, then 52 00:02:29,940 --> 00:02:32,730 you get these patterns of cortical activity 53 00:02:32,730 --> 00:02:36,420 starting in your sensory areas, and moving outwards from there 54 00:02:36,420 --> 00:02:40,890 from primary sensory cortex to these association areas that 55 00:02:40,890 --> 00:02:43,140 do things like object recognition, 56 00:02:43,140 --> 00:02:44,940 or parse the language that you're hearing. 57 00:02:49,110 --> 00:02:52,320 And so you can think of any given experience 58 00:02:52,320 --> 00:02:56,940 as being a particular pattern of neuron firing. 59 00:02:56,940 --> 00:02:59,610 And we know from last week that this area 60 00:02:59,610 --> 00:03:02,820 in the prefrontal cortex is key for attending to experiences, 61 00:03:02,820 --> 00:03:06,420 to allowing one or another aspect of what we're 62 00:03:06,420 --> 00:03:08,190 experiencing to come to the surface, 63 00:03:08,190 --> 00:03:12,210 to be the thing that gets processed. 64 00:03:12,210 --> 00:03:15,300 And so we can model what's happening in memory as allowing 65 00:03:15,300 --> 00:03:20,610 the prefrontal cortex, and to take all of these experience 66 00:03:20,610 --> 00:03:23,940 representations, and funnel them into this medial temporal area 67 00:03:23,940 --> 00:03:25,770 where the hippocampus is. 68 00:03:25,770 --> 00:03:30,270 And the hippocampus takes all of this input, 69 00:03:30,270 --> 00:03:32,430 and it strengthens the connections between them 70 00:03:32,430 --> 00:03:35,100 so that you can form a memory of an experience. 71 00:03:39,620 --> 00:03:42,970 So the way we think this happens is 72 00:03:42,970 --> 00:03:45,430 that when you first form like an episodic memory, 73 00:03:45,430 --> 00:03:49,540 a memory of something that happens to you, then 74 00:03:49,540 --> 00:03:52,450 what happens is that synapse strings between the hippocampus 75 00:03:52,450 --> 00:03:54,880 and other parts of cortex get changed so 76 00:03:54,880 --> 00:03:57,190 that the hippocampus can bring up 77 00:03:57,190 --> 00:04:01,470 all the pieces of that memory at once 78 00:04:01,470 --> 00:04:04,780 so that you bring up that pattern of activity 79 00:04:04,780 --> 00:04:07,300 that you had when you actually had the experience. 80 00:04:07,300 --> 00:04:11,470 That episodic memory involves recreating, recapitulating 81 00:04:11,470 --> 00:04:13,420 this pattern of activity. 82 00:04:16,120 --> 00:04:20,019 And so when you have one aspect of a memory cue other parts 83 00:04:20,019 --> 00:04:22,000 of it, the idea is that what's happening 84 00:04:22,000 --> 00:04:24,640 is that by activating the representation for this one 85 00:04:24,640 --> 00:04:29,500 part, that then brings out that hippocampal 86 00:04:29,500 --> 00:04:31,210 through the hippocampus then activates 87 00:04:31,210 --> 00:04:34,510 all of these other parts, and let's you get the whole memory. 88 00:04:34,510 --> 00:04:38,890 And over time, as you remember things either over time 89 00:04:38,890 --> 00:04:42,370 or through repeated retrieval of these memories, 90 00:04:42,370 --> 00:04:44,980 then the links between the cortical representations 91 00:04:44,980 --> 00:04:46,335 themselves become stronger. 92 00:04:46,335 --> 00:04:48,460 And you get this idea where memory is consolidated, 93 00:04:48,460 --> 00:04:51,850 where it moves away from depending on the hippocampus, 94 00:04:51,850 --> 00:04:53,100 and into just cortex. 95 00:04:53,100 --> 00:04:58,000 If we look at our retrograde amnesic patients like HM, 96 00:04:58,000 --> 00:05:01,090 remember that his memory for distant past events, childhood 97 00:05:01,090 --> 00:05:02,200 events was really good. 98 00:05:02,200 --> 00:05:05,110 His memory for events a few years before his surgery 99 00:05:05,110 --> 00:05:05,965 was really bad. 100 00:05:05,965 --> 00:05:10,080 So there's a lot of evidence from other patients 101 00:05:10,080 --> 00:05:13,690 in similar situations that you need the hippocampus 102 00:05:13,690 --> 00:05:15,886 for recent episodic memories. 103 00:05:15,886 --> 00:05:17,260 You don't need it for older ones. 104 00:05:21,340 --> 00:05:22,630 Whoa, memory. 105 00:05:27,810 --> 00:05:28,840 Does that make sense? 106 00:05:28,840 --> 00:05:30,630 This idea that at first, the hippocampus 107 00:05:30,630 --> 00:05:32,379 can bring up all of these different pieces 108 00:05:32,379 --> 00:05:34,380 of a representation, than over time, one of them 109 00:05:34,380 --> 00:05:36,255 can trigger all the others as the connections 110 00:05:36,255 --> 00:05:39,440 between the cortical representations get stronger? 111 00:05:45,344 --> 00:05:46,328 All right. 112 00:05:59,620 --> 00:06:01,720 Moving along. 113 00:06:01,720 --> 00:06:04,220 Numbers. 114 00:06:04,220 --> 00:06:08,080 So we know that people can in general do math. 115 00:06:08,080 --> 00:06:09,260 Anyone here like math? 116 00:06:09,260 --> 00:06:11,337 Anyone think they're kind of good at it? 117 00:06:11,337 --> 00:06:12,740 Whoa. 118 00:06:12,740 --> 00:06:14,670 All right. 119 00:06:14,670 --> 00:06:18,610 Got probably a non-typical sample of the population here. 120 00:06:18,610 --> 00:06:22,154 So doing basic arithmetic types of math, 121 00:06:22,154 --> 00:06:24,070 which is, of course, not the only kind of math 122 00:06:24,070 --> 00:06:26,440 or even the most interesting kind of math, 123 00:06:26,440 --> 00:06:30,400 but it is, one, the kind of math that most people know about, 124 00:06:30,400 --> 00:06:33,107 and that most people encounter in their daily lives, 125 00:06:33,107 --> 00:06:34,690 and that most people are taught to do. 126 00:06:34,690 --> 00:06:37,900 So it's the one that's been most studied and the most tested. 127 00:06:37,900 --> 00:06:42,440 Seems to require three things that most human beings can do. 128 00:06:42,440 --> 00:06:45,550 There's this innate sense of number. 129 00:06:45,550 --> 00:06:50,170 This sense of what numbers are little, what numbers are big. 130 00:06:50,170 --> 00:06:52,390 This ability for small collections of items 131 00:06:52,390 --> 00:06:54,520 to look at them and know, without counting, 132 00:06:54,520 --> 00:06:55,810 how many items are there. 133 00:07:00,790 --> 00:07:03,210 And then there's this second step. 134 00:07:03,210 --> 00:07:08,080 It's this ability to, for example, count, 135 00:07:08,080 --> 00:07:14,350 to connect number names with numbers, with items. 136 00:07:14,350 --> 00:07:16,930 And thirdly, you need to have an ability 137 00:07:16,930 --> 00:07:19,480 to follow algorithms, to follow the steps in a process. 138 00:07:19,480 --> 00:07:22,784 Think about doing addition, and carrying numbers. 139 00:07:22,784 --> 00:07:23,950 Something we know how to do. 140 00:07:23,950 --> 00:07:26,210 Or even the basic if you're multiplying it by 10, 141 00:07:26,210 --> 00:07:28,300 you stick a 0 on the end. 142 00:07:28,300 --> 00:07:32,200 All of us probably learned these little tricks really early, 143 00:07:32,200 --> 00:07:34,780 and we know how to do them. 144 00:07:34,780 --> 00:07:38,077 And this fact that the ability to follow algorithms 145 00:07:38,077 --> 00:07:39,910 is so important is you see a lot of students 146 00:07:39,910 --> 00:07:43,270 who can follow all the steps in the math they're doing, 147 00:07:43,270 --> 00:07:45,310 and can get the right answer, and don't really 148 00:07:45,310 --> 00:07:47,434 understand why they're doing this, or how it works. 149 00:07:47,434 --> 00:07:49,330 And you often see people who feel 150 00:07:49,330 --> 00:07:51,880 like it doesn't make any sense, because they 151 00:07:51,880 --> 00:07:56,560 don't have a good feel for what the underlying numbers they're 152 00:07:56,560 --> 00:07:58,670 working with are. 153 00:07:58,670 --> 00:08:06,040 So let's talk first about this basic sense of number idea. 154 00:08:06,040 --> 00:08:10,870 I'm going to briefly show you about half a 155 00:08:10,870 --> 00:08:15,790 second, a little less, show you an image with some dots in it. 156 00:08:15,790 --> 00:08:17,100 I want you to tell me how many. 157 00:08:17,100 --> 00:08:18,940 Ready? 158 00:08:18,940 --> 00:08:19,580 How many? 159 00:08:19,580 --> 00:08:20,260 AUDIENCE: One. 160 00:08:20,260 --> 00:08:20,926 ABBY NOYCE: One. 161 00:08:20,926 --> 00:08:21,780 Was that easy? 162 00:08:21,780 --> 00:08:23,266 AUDIENCE: Yes. 163 00:08:23,266 --> 00:08:25,330 ABBY NOYCE: Want to do another one? 164 00:08:25,330 --> 00:08:25,830 How many? 165 00:08:25,830 --> 00:08:26,734 AUDIENCE: [INAUDIBLE] 166 00:08:26,734 --> 00:08:27,650 ABBY NOYCE: All right. 167 00:08:27,650 --> 00:08:29,420 Good. 168 00:08:29,420 --> 00:08:29,920 How many? 169 00:08:29,920 --> 00:08:31,990 [INTERPOSING VOICES] 170 00:08:31,990 --> 00:08:33,559 ABBY NOYCE: How many? 171 00:08:33,559 --> 00:08:35,376 [INTERPOSING VOICES] 172 00:08:35,376 --> 00:08:36,640 ABBY NOYCE: I hear fives. 173 00:08:36,640 --> 00:08:37,150 I hear six. 174 00:08:37,150 --> 00:08:38,650 How many people think it was five? 175 00:08:38,650 --> 00:08:39,150 Go ahead. 176 00:08:39,150 --> 00:08:39,919 Don't worry about being wrong. 177 00:08:39,919 --> 00:08:41,690 Raise your hand with what you actually thought it was. 178 00:08:41,690 --> 00:08:43,390 How many people think it was six? 179 00:08:43,390 --> 00:08:45,620 Sevens? 180 00:08:45,620 --> 00:08:46,412 What did you think? 181 00:08:46,412 --> 00:08:47,536 AUDIENCE: I wasn't looking. 182 00:08:47,536 --> 00:08:48,380 ABBY NOYCE: [LAUGHS] 183 00:08:48,380 --> 00:08:49,379 AUDIENCE: I looked away. 184 00:08:49,379 --> 00:08:52,010 ABBY NOYCE: So I noticed that our consensus went down. 185 00:08:52,010 --> 00:08:54,532 Was that one harder? 186 00:08:54,532 --> 00:08:55,880 AUDIENCE: How many was it? 187 00:08:55,880 --> 00:08:56,240 ABBY NOYCE: Six. 188 00:08:56,240 --> 00:08:57,130 AUDIENCE: Yes! 189 00:08:57,130 --> 00:08:58,910 [LAUGHS] 190 00:08:58,910 --> 00:09:00,830 ABBY NOYCE: Want another one? 191 00:09:00,830 --> 00:09:01,330 How many? 192 00:09:01,330 --> 00:09:02,593 AUDIENCE: 11. 193 00:09:02,593 --> 00:09:04,610 ABBY NOYCE: [LAUGHS] I don't think it was. 194 00:09:04,610 --> 00:09:05,860 I think it was more than that. 195 00:09:05,860 --> 00:09:08,254 1, 2, 3, 4, 5, 6, 7-- 196 00:09:08,254 --> 00:09:08,920 I think it's 12. 197 00:09:08,920 --> 00:09:09,600 AUDIENCE: Yes! 198 00:09:09,600 --> 00:09:10,670 ABBY NOYCE: But notice that at this point, 199 00:09:10,670 --> 00:09:11,734 we have to count them. 200 00:09:11,734 --> 00:09:12,900 AUDIENCE: Well, I was close. 201 00:09:12,900 --> 00:09:14,024 ABBY NOYCE: You were close. 202 00:09:14,024 --> 00:09:15,590 You're estimating is good. 203 00:09:15,590 --> 00:09:18,420 So for numbers up to about three or four, 204 00:09:18,420 --> 00:09:20,150 depending on whose papers you're reading, 205 00:09:20,150 --> 00:09:21,950 people are fast and accurate. 206 00:09:21,950 --> 00:09:23,579 And they're equally fast. 207 00:09:23,579 --> 00:09:26,120 The difference it takes you to say, that was one, versus that 208 00:09:26,120 --> 00:09:29,240 was two, versus that was three, versus that was four, 209 00:09:29,240 --> 00:09:29,880 it's flat. 210 00:09:29,880 --> 00:09:32,780 It's the same amount of time for any of these responses. 211 00:09:32,780 --> 00:09:35,660 For larger numbers, people's estimate 212 00:09:35,660 --> 00:09:39,320 for doing it fast like this drops off. 213 00:09:39,320 --> 00:09:44,714 People can get pretty close, as you guys all demonstrated when 214 00:09:44,714 --> 00:09:47,130 we had six, and we had a couple of numbers on either side, 215 00:09:47,130 --> 00:09:51,105 or when we got that bigger number and I heard 11, and 13, 216 00:09:51,105 --> 00:09:53,580 and all right in there, around 12. 217 00:09:53,580 --> 00:09:57,320 But for larger numbers, to get an accurate answer were slower. 218 00:09:57,320 --> 00:09:59,840 And we're not as accurate. 219 00:09:59,840 --> 00:10:01,850 And one thing that happens with larger numbers 220 00:10:01,850 --> 00:10:04,860 is that as a collection increases in size, 221 00:10:04,860 --> 00:10:08,480 the time it takes people to answer how many things are you 222 00:10:08,480 --> 00:10:11,090 showing me goes up linearly. 223 00:10:11,090 --> 00:10:13,730 For each extra item, there's a certain amount of processing 224 00:10:13,730 --> 00:10:15,350 time that's added. 225 00:10:15,350 --> 00:10:19,010 So what seems to be happening is that we're counting, 226 00:10:19,010 --> 00:10:22,670 is that there's some good behavioral evidence 227 00:10:22,670 --> 00:10:27,620 that for collections of items above about 4, 228 00:10:27,620 --> 00:10:30,080 that people have to do this extra bit of processing 229 00:10:30,080 --> 00:10:31,400 per each item that's shown. 230 00:10:34,100 --> 00:10:38,560 So this ability to look at a screen very quickly 231 00:10:38,560 --> 00:10:41,410 and be like, there's three, there's two, there's four, 232 00:10:41,410 --> 00:10:44,830 is called subitizing. 233 00:10:44,830 --> 00:10:45,770 Cool word for you. 234 00:10:45,770 --> 00:10:50,710 S-U-B-I-T-I-Z-I-N-G. It's spelled like it's British, 235 00:10:50,710 --> 00:10:52,880 mostly because I think the people who do real math 236 00:10:52,880 --> 00:10:55,470 psychology work are mostly Brits, 237 00:10:55,470 --> 00:10:58,300 or they kind of started it. 238 00:10:58,300 --> 00:11:02,380 And there's one theory that says what's 239 00:11:02,380 --> 00:11:07,660 really happening in subitizing is that we're looking at shape. 240 00:11:07,660 --> 00:11:11,142 So one dot, pretty easy to spot, right? 241 00:11:11,142 --> 00:11:12,100 Don't need to count it. 242 00:11:12,100 --> 00:11:13,690 It's just one dot. 243 00:11:13,690 --> 00:11:17,470 Two dots, no matter how you arrange them, 244 00:11:17,470 --> 00:11:21,520 are always going to form a straight line. 245 00:11:21,520 --> 00:11:27,255 Three dots, unless you have the totally unlikely thing 246 00:11:27,255 --> 00:11:29,410 of lining them all up in a straight line 247 00:11:29,410 --> 00:11:33,890 are going to form triangles. 248 00:11:33,890 --> 00:11:35,270 No matter what. 249 00:11:35,270 --> 00:11:39,220 Four dots randomly arranged are going to form 250 00:11:39,220 --> 00:11:41,710 some kind of quadrilateral. 251 00:11:41,710 --> 00:11:44,650 Now, notice that as we go up, there are other possibilities. 252 00:11:44,650 --> 00:11:47,740 You could do three dots in a straight line. 253 00:11:47,740 --> 00:11:51,790 You could do four dots as a triangle 254 00:11:51,790 --> 00:11:55,702 with a dot in the middle, or as a straight line. 255 00:11:55,702 --> 00:11:57,160 And four, again, is where you start 256 00:11:57,160 --> 00:11:59,620 seeing people's performance get a little bit sloppier, 257 00:11:59,620 --> 00:12:02,830 but still in the subitizing quick range, and not 258 00:12:02,830 --> 00:12:06,979 so much in the linear addition of time counting range. 259 00:12:06,979 --> 00:12:08,770 So one thing that says what's going on here 260 00:12:08,770 --> 00:12:12,850 is really shape analysis, and other folks 261 00:12:12,850 --> 00:12:15,149 who say, no, no, no, it's an intrinsic number sense. 262 00:12:15,149 --> 00:12:16,690 But either way, this is a distinction 263 00:12:16,690 --> 00:12:18,480 that we are really good at. 264 00:12:21,490 --> 00:12:23,590 What about other kinds of critters? 265 00:12:23,590 --> 00:12:25,750 So we know that humans are good with numbers. 266 00:12:25,750 --> 00:12:27,460 We like to think that humans are special. 267 00:12:29,854 --> 00:12:32,020 Humans are not always as special as we think we are. 268 00:12:32,020 --> 00:12:34,700 Numbers is one of the place where you can't do it. 269 00:12:34,700 --> 00:12:38,520 Otto Kohler was working in the '40s and '50s. 270 00:12:38,520 --> 00:12:41,050 German guy, as you might have guessed. 271 00:12:41,050 --> 00:12:43,300 And he trained birds to distinguish 272 00:12:43,300 --> 00:12:46,150 between numbers in the two to six 273 00:12:46,150 --> 00:12:50,300 range using dots, collections of dots. 274 00:12:50,300 --> 00:12:53,800 He was working with ravens. 275 00:12:53,800 --> 00:12:56,710 And he would show them a card with between two and six dots 276 00:12:56,710 --> 00:12:58,000 arranged on it. 277 00:12:58,000 --> 00:13:01,120 And then there were two food receptacles 278 00:13:01,120 --> 00:13:04,150 with different numbers of dots covering the food receptacles. 279 00:13:04,150 --> 00:13:06,900 And the one that matched the number of dots 280 00:13:06,900 --> 00:13:09,650 that the bird was originally shown, not in the same pattern, 281 00:13:09,650 --> 00:13:12,370 just the same number of dots, was 282 00:13:12,370 --> 00:13:14,440 the one that had food in it. 283 00:13:14,440 --> 00:13:15,940 And these birds could learn to do 284 00:13:15,940 --> 00:13:19,900 that kind of comparison between this group of dots 285 00:13:19,900 --> 00:13:21,010 and that group of dots. 286 00:13:24,130 --> 00:13:30,550 Alex, the noted gray parrot who died last fall, maybe? 287 00:13:30,550 --> 00:13:31,640 Recently, anyway. 288 00:13:31,640 --> 00:13:35,200 It was all over the news. 289 00:13:35,200 --> 00:13:37,990 So a lot of people have always thought that parrots are just 290 00:13:37,990 --> 00:13:38,920 mimicking. 291 00:13:38,920 --> 00:13:41,170 Parrots who say words are just mimicking. 292 00:13:41,170 --> 00:13:43,960 Irene Pepperberg believed strongly 293 00:13:43,960 --> 00:13:47,350 that this is not true, that parrots can at least learn 294 00:13:47,350 --> 00:13:49,900 some basic concepts. 295 00:13:49,900 --> 00:13:54,230 And she taught Alex to distinguish some colors, 296 00:13:54,230 --> 00:13:58,870 shapes, and numbers up to five. 297 00:13:58,870 --> 00:14:00,940 So Alex, probably for this tray, could not 298 00:14:00,940 --> 00:14:03,670 say how many things there were. 299 00:14:03,670 --> 00:14:06,411 But for a small group of items, you 300 00:14:06,411 --> 00:14:07,660 could say, how many are there? 301 00:14:07,660 --> 00:14:12,820 And he would say, two, or four, or three. 302 00:14:12,820 --> 00:14:15,940 Definitely, again, with reasonable accuracy, 303 00:14:15,940 --> 00:14:19,215 could distinguish between these different groups of numbers. 304 00:14:19,215 --> 00:14:20,590 This probably doesn't necessarily 305 00:14:20,590 --> 00:14:23,714 mean that either of these birds are counting, per se, 306 00:14:23,714 --> 00:14:25,630 in the sense where we think of it where you're 307 00:14:25,630 --> 00:14:29,110 going one, two, three, four. 308 00:14:29,110 --> 00:14:30,790 But just like we have an intuitive 309 00:14:30,790 --> 00:14:34,750 grasp of the sizes of different amounts of things, 310 00:14:34,750 --> 00:14:36,664 different numbers of things, these critters 311 00:14:36,664 --> 00:14:37,580 can do the same thing. 312 00:14:40,200 --> 00:14:43,770 One course that did not do arithmetic, 313 00:14:43,770 --> 00:14:46,380 you guys may have heard of Clever Hans. 314 00:14:46,380 --> 00:14:50,550 So this is way back right around the turn of the 20th century. 315 00:14:50,550 --> 00:14:54,210 Wilhelm von Osten claimed that he had taught arithmetic 316 00:14:54,210 --> 00:14:56,550 to his horse. 317 00:14:56,550 --> 00:14:57,720 And they toured the country. 318 00:14:57,720 --> 00:14:59,430 They toured the continent. 319 00:14:59,430 --> 00:15:00,930 They drew admiring crowds. 320 00:15:03,930 --> 00:15:06,930 He would say, Hans, what is 2 plus 2? 321 00:15:06,930 --> 00:15:09,836 And Hans would sit there and tap his front foot. 322 00:15:09,836 --> 00:15:13,410 [TAPPING FOOT] 323 00:15:13,410 --> 00:15:13,935 Four times. 324 00:15:16,609 --> 00:15:17,650 It was pretty convincing. 325 00:15:17,650 --> 00:15:18,930 He could handle fractions. 326 00:15:18,930 --> 00:15:23,192 He could handle addition, subtraction, multiplication. 327 00:15:23,192 --> 00:15:26,920 AUDIENCE: How would you demonstrate a fraction? 328 00:15:26,920 --> 00:15:29,041 What's 1/3 plus 1/3? 329 00:15:29,041 --> 00:15:31,987 [TAPPING FOOT] 330 00:15:32,970 --> 00:15:36,820 As two pieces with a pause in the middle. 331 00:15:36,820 --> 00:15:39,120 AUDIENCE: How would they know [INAUDIBLE] 332 00:15:39,120 --> 00:15:40,410 ABBY NOYCE: How would they know the horse was doing it? 333 00:15:40,410 --> 00:15:42,034 The horse is the one doing the tapping. 334 00:15:42,034 --> 00:15:43,246 The horse would tap his hoof. 335 00:15:43,246 --> 00:15:45,070 AUDIENCE: I thought there was another guy sitting there. 336 00:15:45,070 --> 00:15:45,570 ABBY NOYCE: No. 337 00:15:45,570 --> 00:15:46,990 AUDIENCE: I saw a cartoon with that once. 338 00:15:46,990 --> 00:15:48,765 I think that's The Wild Thornberrys. 339 00:15:48,765 --> 00:15:51,056 They had the horse, except there was somebody backstage 340 00:15:51,056 --> 00:15:52,100 kicking the horse. 341 00:15:52,100 --> 00:15:53,620 ABBY NOYCE: I saw that episode. 342 00:15:53,620 --> 00:15:54,120 Yes. 343 00:15:54,120 --> 00:15:54,960 AUDIENCE: I saw that. 344 00:15:54,960 --> 00:15:56,960 ABBY NOYCE: This was a little subtler than that. 345 00:15:56,960 --> 00:16:01,020 So in 1904, a committee of psychological experts 346 00:16:01,020 --> 00:16:03,120 investigated this, observed, decided that there 347 00:16:03,120 --> 00:16:05,340 was no trickery involved. 348 00:16:05,340 --> 00:16:07,590 Nobody was poking the horse from backstage or anything 349 00:16:07,590 --> 00:16:08,880 like that. 350 00:16:08,880 --> 00:16:11,910 And concluded that it was for real. 351 00:16:11,910 --> 00:16:14,800 And a guy named Oskar Pfungst. 352 00:16:14,800 --> 00:16:16,890 And I don't think I can pronounce that right, 353 00:16:16,890 --> 00:16:17,765 but I'm going to try. 354 00:16:17,765 --> 00:16:18,850 Pfungst. 355 00:16:18,850 --> 00:16:20,337 In 1907. 356 00:16:20,337 --> 00:16:21,420 He didn't really buy this. 357 00:16:21,420 --> 00:16:23,640 He'd been part of that original study committee, 358 00:16:23,640 --> 00:16:27,450 and the committee had come to the consensus that it was real. 359 00:16:27,450 --> 00:16:31,410 And he was just like, yeah, I can still see some issues here. 360 00:16:31,410 --> 00:16:35,520 And he tested Hans in a variety of situations. 361 00:16:35,520 --> 00:16:40,740 He had situations where somebody other than von Osten 362 00:16:40,740 --> 00:16:42,870 was asking the questions. 363 00:16:42,870 --> 00:16:47,040 He had somebody other than von Osten ask the questions when 364 00:16:47,040 --> 00:16:49,170 von Osten was not in the room. 365 00:16:49,170 --> 00:16:54,060 He had somebody state a question in von Osten's hearing, 366 00:16:54,060 --> 00:16:57,990 and then write it on a board which was oriented 367 00:16:57,990 --> 00:17:00,420 so von Osten couldn't see it, but Hans could, 368 00:17:00,420 --> 00:17:01,920 so that Hans was seeing the question 369 00:17:01,920 --> 00:17:05,099 other than the one that had originally been proposed. 370 00:17:05,099 --> 00:17:08,880 And the conclusion that he came to 371 00:17:08,880 --> 00:17:12,210 was that whenever the trainer did not 372 00:17:12,210 --> 00:17:15,240 know what the question was, Hans was wrong. 373 00:17:15,240 --> 00:17:19,530 When the trainer knew what the question was, Hans was right. 374 00:17:19,530 --> 00:17:21,720 So the conclusion that these guys came to 375 00:17:21,720 --> 00:17:25,690 is that consciously or unconsciously, 376 00:17:25,690 --> 00:17:29,100 von Osten was cueing Hans in some way, 377 00:17:29,100 --> 00:17:31,920 cueing him when to stop tapping, more 378 00:17:31,920 --> 00:17:35,820 or less, to tap until he relaxed and it was the right answer. 379 00:17:35,820 --> 00:17:39,000 The general consensus seems to be that the trainer would 380 00:17:39,000 --> 00:17:42,120 get tensor, and tensor, and tensor as the horse approached 381 00:17:42,120 --> 00:17:44,490 the right number of taps, and when he got to the right 382 00:17:44,490 --> 00:17:48,420 one would relax, and the horse could pick up 383 00:17:48,420 --> 00:17:53,457 on this cue, which was probably a very subtle cue, and figure 384 00:17:53,457 --> 00:17:55,290 out that the right thing to do at that point 385 00:17:55,290 --> 00:17:58,650 was to stop tapping, which is very clever, 386 00:17:58,650 --> 00:18:00,112 but it is not arithmetic. 387 00:18:02,960 --> 00:18:05,310 This kind of cueing is one of the reasons for things 388 00:18:05,310 --> 00:18:07,650 like very careful double blind clinical studies, 389 00:18:07,650 --> 00:18:09,810 the fact that when one of the people in the room 390 00:18:09,810 --> 00:18:12,434 knows what's going on, no matter how careful they think they're 391 00:18:12,434 --> 00:18:15,480 being, other people can often pick up 392 00:18:15,480 --> 00:18:19,020 on what that person knows. 393 00:18:19,020 --> 00:18:21,450 And it's one reason. 394 00:18:21,450 --> 00:18:24,630 So Clever Hans, one of the things that this incident did 395 00:18:24,630 --> 00:18:27,690 is it made a lot of psychologists very dubious 396 00:18:27,690 --> 00:18:30,680 about any and all claims of animal arithmetic abilities. 397 00:18:30,680 --> 00:18:33,570 So those number abilities tests that we started 398 00:18:33,570 --> 00:18:36,360 seeing in the '40s and '50s, everyone 399 00:18:36,360 --> 00:18:39,120 was really skeptical for a while until the results 400 00:18:39,120 --> 00:18:40,560 started being really solid. 401 00:18:40,560 --> 00:18:42,810 AUDIENCE: Was this actually proved? 402 00:18:42,810 --> 00:18:44,740 ABBY NOYCE: Was what actually proved? 403 00:18:44,740 --> 00:18:47,650 AUDIENCE: That we was cueing by getting tensor. 404 00:18:50,560 --> 00:18:53,074 Because he could've been blinking 405 00:18:53,074 --> 00:18:54,482 like one, two, [INAUDIBLE] 406 00:18:54,482 --> 00:18:55,190 ABBY NOYCE: Yeah. 407 00:18:59,090 --> 00:19:02,360 So from what I've read about this, 408 00:19:02,360 --> 00:19:03,920 the general consensus seems to have 409 00:19:03,920 --> 00:19:06,010 been that von Osten thought it was for real. 410 00:19:06,010 --> 00:19:09,470 von Osten seems to have at least convinced people 411 00:19:09,470 --> 00:19:12,230 that he was not deliberately cueing the horse. 412 00:19:12,230 --> 00:19:19,160 He really thought Hans could do this, 413 00:19:19,160 --> 00:19:20,750 that it was not deliberate. 414 00:19:20,750 --> 00:19:23,480 So there could have been other cueing things 415 00:19:23,480 --> 00:19:26,140 that Hans was picking up on. 416 00:19:26,140 --> 00:19:28,700 Pfungst, the guy who finally disproved 417 00:19:28,700 --> 00:19:30,980 it did something similar with human subjects, 418 00:19:30,980 --> 00:19:36,950 where he'd have somebody show them a arithmetic problem, 419 00:19:36,950 --> 00:19:39,080 and he would answer by tapping. 420 00:19:39,080 --> 00:19:41,870 And he said, he could pull enough cues off 421 00:19:41,870 --> 00:19:47,810 of random people to know when to stop tapping about 80% 422 00:19:47,810 --> 00:19:49,700 of the time-ish. 423 00:19:49,700 --> 00:19:52,280 And he said he described it as this growing tension 424 00:19:52,280 --> 00:19:55,850 in the person who is listening. 425 00:19:55,850 --> 00:19:57,170 It's 1907. 426 00:19:57,170 --> 00:20:00,980 It's not quite up to modern research standards. 427 00:20:00,980 --> 00:20:03,920 Horses don't go into laboratories very well. 428 00:20:03,920 --> 00:20:06,120 But nonetheless, it's an interesting story. 429 00:20:06,120 --> 00:20:07,550 It's an interesting case of where 430 00:20:07,550 --> 00:20:09,050 there was a confound that slipped 431 00:20:09,050 --> 00:20:11,450 right past people the first time they looked at it. 432 00:20:11,450 --> 00:20:12,740 Remember, a confound is something else 433 00:20:12,740 --> 00:20:14,948 that's varying along with what you're trying to study 434 00:20:14,948 --> 00:20:16,415 that can mess up your results. 435 00:20:20,010 --> 00:20:20,940 So Hans is clever. 436 00:20:25,900 --> 00:20:27,970 So we know this idea that humans started 437 00:20:27,970 --> 00:20:29,290 developing number sense. 438 00:20:29,290 --> 00:20:31,300 When did they do it? 439 00:20:31,300 --> 00:20:35,800 Well, so Jean Piaget is the classic developmental guy. 440 00:20:39,050 --> 00:20:42,010 In the '50s, he set up this set of stages, 441 00:20:42,010 --> 00:20:43,667 developmental stages. 442 00:20:43,667 --> 00:20:45,250 So at this age, children will do this. 443 00:20:45,250 --> 00:20:46,610 And at that age, children will do that. 444 00:20:46,610 --> 00:20:48,190 And one of the things he pointed out, 445 00:20:48,190 --> 00:20:53,350 he claimed was that four and five-year-olds, if you take 446 00:20:53,350 --> 00:20:57,580 a bunch of items, say, six items twice, 447 00:20:57,580 --> 00:21:07,000 and we line them up in rows, and say, which row has more items? 448 00:21:07,000 --> 00:21:10,680 The kids will say, they're the same. 449 00:21:10,680 --> 00:21:13,320 And if the experimenter then, with a kid watching them, 450 00:21:13,320 --> 00:21:16,196 takes one of the rows and spreads them out-- 451 00:21:16,196 --> 00:21:18,570 and I'm going to make my things wider, which isn't really 452 00:21:18,570 --> 00:21:19,020 what I want to do. 453 00:21:19,020 --> 00:21:19,520 Spread out. 454 00:21:23,040 --> 00:21:25,440 And says, now which one has more rows? 455 00:21:25,440 --> 00:21:27,810 The kid will say, the longer one. 456 00:21:27,810 --> 00:21:29,552 So Piaget pointed this out. 457 00:21:29,552 --> 00:21:31,260 And Piaget says that this means that kids 458 00:21:31,260 --> 00:21:34,350 who are under about five don't have 459 00:21:34,350 --> 00:21:38,190 this core idea of numerosity where unless you add or remove 460 00:21:38,190 --> 00:21:40,650 something to the set, the number stays the same. 461 00:21:43,780 --> 00:21:46,170 Some people have been kind of skeptical of this. 462 00:21:46,170 --> 00:21:48,030 One reason for being skeptical of it 463 00:21:48,030 --> 00:21:52,710 is if you do this with two-year-olds, 464 00:21:52,710 --> 00:21:54,360 they get it right. 465 00:21:54,360 --> 00:21:56,220 Three-year-Olds get it right. 466 00:21:56,220 --> 00:21:59,950 Four and five-year-olds get it wrong. 467 00:21:59,950 --> 00:22:02,550 So one hypothesis is that human children 468 00:22:02,550 --> 00:22:06,200 lose number ability between the ages of four and six. 469 00:22:06,200 --> 00:22:08,910 That seems unlikely. 470 00:22:08,910 --> 00:22:12,840 Other people have pointed out that around age five 471 00:22:12,840 --> 00:22:16,740 is when kids start really thinking about this idea 472 00:22:16,740 --> 00:22:19,920 that other people have motivations and beliefs that 473 00:22:19,920 --> 00:22:21,614 underlie what they're doing. 474 00:22:21,614 --> 00:22:24,030 They start being able to reason about why other people are 475 00:22:24,030 --> 00:22:26,850 doing and saying things in a way that's 476 00:22:26,850 --> 00:22:29,580 more sophisticated than younger kids can. 477 00:22:29,580 --> 00:22:32,220 And that logic says that-- 478 00:22:36,330 --> 00:22:38,550 so I just told this grown up, who 479 00:22:38,550 --> 00:22:40,830 is clearly smarter and more clueful than me, 480 00:22:40,830 --> 00:22:41,820 that they're the same. 481 00:22:41,820 --> 00:22:44,280 And she moved them, and they're clearly still the same. 482 00:22:44,280 --> 00:22:46,410 Why is she asking you this question again? 483 00:22:46,410 --> 00:22:49,782 She must mean something other than what she just asked. 484 00:22:49,782 --> 00:22:51,990 I must have misunderstood the question she is asking. 485 00:22:51,990 --> 00:22:56,014 She must really be asking about the size of the row. 486 00:22:56,014 --> 00:22:57,930 And will then trying to give the right answer, 487 00:22:57,930 --> 00:23:00,120 trying to make the adult happy will 488 00:23:00,120 --> 00:23:02,463 pointing to the longer row. 489 00:23:02,463 --> 00:23:04,920 AUDIENCE: [INAUDIBLE] 490 00:23:04,920 --> 00:23:06,670 ABBY NOYCE: Somebody. 491 00:23:06,670 --> 00:23:08,285 So we said the two and three-year-olds 492 00:23:08,285 --> 00:23:09,660 who are too young to do that kind 493 00:23:09,660 --> 00:23:11,610 of complicated reasoning about the motivations 494 00:23:11,610 --> 00:23:18,810 behind other people's behavior will get this right. 495 00:23:18,810 --> 00:23:21,720 Four and five-year-olds, that age where if you are doing this 496 00:23:21,720 --> 00:23:23,559 by asking them questions-- 497 00:23:23,559 --> 00:23:26,100 has anyone here ever tried to have a conversation with a four 498 00:23:26,100 --> 00:23:27,990 or five-year-old? 499 00:23:27,990 --> 00:23:31,120 It's a little weird, isn't it? 500 00:23:31,120 --> 00:23:33,600 So one of the problems with Piaget's original experiment 501 00:23:33,600 --> 00:23:38,340 is that it depends on having a verbal conversation with a four 502 00:23:38,340 --> 00:23:41,310 or five-year-old, which we know that little kids, 503 00:23:41,310 --> 00:23:44,580 and little kids and language can be a little wonky at times. 504 00:23:44,580 --> 00:23:48,840 So somebody else said, what happens if we give them M&Ms, 505 00:23:48,840 --> 00:23:51,090 and we show them two rows of M&Ms, 506 00:23:51,090 --> 00:23:59,520 and one of them has four M&Ms, and one of them has six M&Ms? 507 00:24:02,420 --> 00:24:05,820 And instead of saying, which one has more, we say, 508 00:24:05,820 --> 00:24:09,600 you can pick one row of M&Ms and eat them. 509 00:24:09,600 --> 00:24:12,030 This gets around the language problem. 510 00:24:12,030 --> 00:24:14,580 This gives the kids a nice, strong incentive 511 00:24:14,580 --> 00:24:16,440 to pick the one that has more. 512 00:24:16,440 --> 00:24:18,990 And when you frame this problem in this way, four 513 00:24:18,990 --> 00:24:21,150 and five-year-olds, the kids who Piaget says 514 00:24:21,150 --> 00:24:23,070 don't understand this core concept of number, 515 00:24:23,070 --> 00:24:25,320 that number is unrelated to the amount of space things 516 00:24:25,320 --> 00:24:29,040 take up, get it right every time. 517 00:24:29,040 --> 00:24:31,260 So this is another piece of evidence 518 00:24:31,260 --> 00:24:33,240 for that other hypothesis that these kids 519 00:24:33,240 --> 00:24:36,056 are figuring that they misunderstood the question, 520 00:24:36,056 --> 00:24:38,430 and trying to parse it in some way that makes more sense. 521 00:24:41,440 --> 00:24:43,080 So we know that toddlers have this kind 522 00:24:43,080 --> 00:24:48,120 of core sense of number of what's more, what's last. 523 00:24:48,120 --> 00:24:50,790 Two-year-olds can't count, by the way, 524 00:24:50,790 --> 00:24:54,810 but they can do this intuitive judgment. 525 00:24:54,810 --> 00:24:57,420 What about littler kids, once we start getting really 526 00:24:57,420 --> 00:24:59,550 down into baby size? 527 00:24:59,550 --> 00:25:02,880 So this guy named Starkey did a bunch of experiments 528 00:25:02,880 --> 00:25:06,480 with babies between 16 and 30 weeks of age. 529 00:25:06,480 --> 00:25:09,821 So that's between four and eight months. 530 00:25:09,821 --> 00:25:11,820 And when you want to work with babies, with kids 531 00:25:11,820 --> 00:25:14,100 who are too young to talk, too young to have 532 00:25:14,100 --> 00:25:16,500 a whole lot of hand-eye coordination, 533 00:25:16,500 --> 00:25:19,410 you start having to get clever experiment designs in order 534 00:25:19,410 --> 00:25:21,510 to figure out what they're doing. 535 00:25:21,510 --> 00:25:28,467 So Starkey said, aha. 536 00:25:28,467 --> 00:25:30,300 We know that babies are interested in things 537 00:25:30,300 --> 00:25:33,190 that are new and different. 538 00:25:33,190 --> 00:25:35,400 And that if I give them they've seen 539 00:25:35,400 --> 00:25:38,285 a lot of times and something that is new, 540 00:25:38,285 --> 00:25:39,660 they will spend more time looking 541 00:25:39,660 --> 00:25:44,130 at the thing that is new than the thing they've seen before. 542 00:25:44,130 --> 00:25:47,850 So they had kids sit on their parents' lap, 543 00:25:47,850 --> 00:25:49,890 and they videotaped them so they could confirm 544 00:25:49,890 --> 00:25:52,544 where their eyes were looking. 545 00:25:52,544 --> 00:25:54,460 You probably can't use eye trackers on babies. 546 00:25:54,460 --> 00:25:56,070 I think they move around too much. 547 00:25:56,070 --> 00:25:57,570 But this was in the '80s before they 548 00:25:57,570 --> 00:25:59,310 had good eye-tracking software. 549 00:25:59,310 --> 00:26:03,480 And they had the babies look at a screen. 550 00:26:03,480 --> 00:26:06,360 And they'd show a picture of two dots, 551 00:26:06,360 --> 00:26:08,490 and the babies would look at it for a little while, 552 00:26:08,490 --> 00:26:10,320 and get bored, and start looking around the room. 553 00:26:10,320 --> 00:26:11,370 And they'd show a different picture 554 00:26:11,370 --> 00:26:13,770 of two dots, two dots in a different arrangement. 555 00:26:13,770 --> 00:26:15,311 And the kids would look at it a while 556 00:26:15,311 --> 00:26:16,590 and they'd get bored faster. 557 00:26:16,590 --> 00:26:18,600 And you show them two dots a few more times, 558 00:26:18,600 --> 00:26:20,640 and they get bored faster and faster each time. 559 00:26:20,640 --> 00:26:23,190 Their attention starts wandering. 560 00:26:23,190 --> 00:26:26,790 If you then change it up to a picture of three dots, 561 00:26:26,790 --> 00:26:28,792 the kids look at it again for a long time, 562 00:26:28,792 --> 00:26:30,750 like they did the first time they saw two dots. 563 00:26:34,230 --> 00:26:35,170 Once, it's random. 564 00:26:35,170 --> 00:26:36,975 But if you do it across, say, 20, 565 00:26:36,975 --> 00:26:40,170 30 babies across several trials, it starts really seeming 566 00:26:40,170 --> 00:26:42,120 like the kids are understanding that there 567 00:26:42,120 --> 00:26:45,030 is a difference between two of something 568 00:26:45,030 --> 00:26:46,726 and three of something. 569 00:26:51,132 --> 00:26:52,840 AUDIENCE: Is there a way to differentiate 570 00:26:52,840 --> 00:26:55,350 between the kids just sensing that there's 571 00:26:55,350 --> 00:27:00,430 a different picture on there, but not really understanding 572 00:27:00,430 --> 00:27:00,972 the quantity? 573 00:27:00,972 --> 00:27:02,596 ABBY NOYCE: Well, remember that they're 574 00:27:02,596 --> 00:27:04,540 showing them several different two dot slides. 575 00:27:04,540 --> 00:27:06,165 So two dots arranged in different ways. 576 00:27:06,165 --> 00:27:08,212 So they're not responding to just the change. 577 00:27:08,212 --> 00:27:09,670 They seem to be responding actually 578 00:27:09,670 --> 00:27:11,140 to the change in quantity. 579 00:27:11,140 --> 00:27:13,090 They're probably not counting, but they 580 00:27:13,090 --> 00:27:16,870 seem to have this recognition that what's being shown 581 00:27:16,870 --> 00:27:18,670 is different from what was shown before. 582 00:27:18,670 --> 00:27:21,460 It captures their attention again. 583 00:27:21,460 --> 00:27:24,340 Somebody did something similar with younger infants, 584 00:27:24,340 --> 00:27:27,340 like under a week-old babies. 585 00:27:27,340 --> 00:27:29,021 And with kids that age, you can't even 586 00:27:29,021 --> 00:27:30,770 sit them up and watch them look at things. 587 00:27:30,770 --> 00:27:32,414 They just kind of flop. 588 00:27:32,414 --> 00:27:35,140 [LAUGHTER] 589 00:27:35,140 --> 00:27:36,115 You know what I mean? 590 00:27:36,115 --> 00:27:40,750 Like really, really, newborns are like, "deh." 591 00:27:40,750 --> 00:27:43,220 They just don't have any motor control at all. 592 00:27:43,220 --> 00:27:47,830 But what they can do is suck on stuff. 593 00:27:47,830 --> 00:27:48,810 Little kids, newborns. 594 00:27:48,810 --> 00:27:51,190 It's one of the things that babies can do right away. 595 00:27:51,190 --> 00:27:52,780 And what researchers have done when 596 00:27:52,780 --> 00:27:55,890 they want to work with really young babies, 597 00:27:55,890 --> 00:27:58,960 these newborn babies, is set something up 598 00:27:58,960 --> 00:28:02,980 where the kid has a little bottle nipple to suck on, 599 00:28:02,980 --> 00:28:04,870 and it's wired up to something that 600 00:28:04,870 --> 00:28:10,590 controls what's being displayed to the baby in some way. 601 00:28:10,590 --> 00:28:14,590 And newborns don't have good visual perception yet. 602 00:28:14,590 --> 00:28:16,840 They can't focus their eyes on anything very far away. 603 00:28:16,840 --> 00:28:18,541 You guys probably know this. 604 00:28:18,541 --> 00:28:20,290 That's why everyone always has the mobiles 605 00:28:20,290 --> 00:28:22,000 in the crib for the baby to practice 606 00:28:22,000 --> 00:28:24,350 focusing on and looking at. 607 00:28:24,350 --> 00:28:26,989 AUDIENCE: [INAUDIBLE] 608 00:28:26,989 --> 00:28:29,030 ABBY NOYCE: Or with the black and white patterns, 609 00:28:29,030 --> 00:28:30,410 the high-contrast patterns. 610 00:28:30,410 --> 00:28:33,092 AUDIENCE: I thought it was to make parents feel better. 611 00:28:33,092 --> 00:28:34,700 ABBY NOYCE: [LAUGHS] Many child-raising things are 612 00:28:34,700 --> 00:28:35,760 to make parents feel better. 613 00:28:35,760 --> 00:28:36,843 I will totally grant that. 614 00:28:36,843 --> 00:28:40,490 Anyway, so they had these little kids, these newborns. 615 00:28:40,490 --> 00:28:42,884 And they had one of these nipple setups 616 00:28:42,884 --> 00:28:44,300 so that when they sucked on it, it 617 00:28:44,300 --> 00:28:46,707 would play a nonsense syllable. 618 00:28:46,707 --> 00:28:49,040 And so they would do, again, the two versus three thing. 619 00:28:49,040 --> 00:28:51,650 They'd have it play a two-syllable nonsense 620 00:28:51,650 --> 00:28:54,620 word every time the babies sucked on it. 621 00:28:54,620 --> 00:28:56,510 And the babies would eventually be like, hey, 622 00:28:56,510 --> 00:28:58,718 look, I can make noises happen, and suck on it a lot. 623 00:28:58,718 --> 00:29:01,970 And eventually, the rate would slow down. 624 00:29:01,970 --> 00:29:06,350 And if they changed it to three-syllable nonsense words, 625 00:29:06,350 --> 00:29:08,450 the babies would, again, respond to the change 626 00:29:08,450 --> 00:29:10,670 by acting more interested. 627 00:29:10,670 --> 00:29:12,170 In this case, sucking on the nipple 628 00:29:12,170 --> 00:29:13,910 again to get these new words, these new 629 00:29:13,910 --> 00:29:16,880 sounds that were different from what was just being played 630 00:29:16,880 --> 00:29:21,440 to be played at a higher rate. 631 00:29:21,440 --> 00:29:23,030 So that's some evidence, at least, 632 00:29:23,030 --> 00:29:25,400 that very, very young children can 633 00:29:25,400 --> 00:29:30,230 make this distinction between two different small numbers. 634 00:29:30,230 --> 00:29:33,740 Numbers seems to be one of these things that's intuitive. 635 00:29:33,740 --> 00:29:36,230 Not counting numbers, not arithmetic numbers, per se, 636 00:29:36,230 --> 00:29:38,930 but this intuitive grasp of small numbers 637 00:29:38,930 --> 00:29:40,730 and the differences between them seems 638 00:29:40,730 --> 00:29:44,210 to be one of these really deeply-ingrained 639 00:29:44,210 --> 00:29:48,530 human capabilities, and not just humans, for other animals, too. 640 00:29:48,530 --> 00:29:51,870 But it's cooler in humans, or something. 641 00:29:51,870 --> 00:29:53,570 Anyway, the other thing that Starkey 642 00:29:53,570 --> 00:29:57,140 show that was really cool is that not only could babies 643 00:29:57,140 --> 00:30:00,320 make this distinction between two dots and three dots. 644 00:30:03,352 --> 00:30:05,060 There's some evidence showing that babies 645 00:30:05,060 --> 00:30:10,100 seem to make the connection between a display of two 646 00:30:10,100 --> 00:30:13,790 items or three items, and the soundtrack of two 647 00:30:13,790 --> 00:30:16,910 beats or three beats. 648 00:30:16,910 --> 00:30:21,530 And this was, again, with older infants, 48-month-old kids. 649 00:30:21,530 --> 00:30:23,810 And again, the sitting on mom's lap. 650 00:30:23,810 --> 00:30:26,830 And they had two screens for them to look at, 651 00:30:26,830 --> 00:30:29,150 one which showed two items, I believe, actually, 652 00:30:29,150 --> 00:30:31,580 like two apples, or two balls, or something, 653 00:30:31,580 --> 00:30:33,470 and one showed three. 654 00:30:33,470 --> 00:30:37,649 And they would play a soundtrack of two or three drumbeats. 655 00:30:37,649 --> 00:30:39,440 And what they discovered is that the babies 656 00:30:39,440 --> 00:30:42,980 would spend more time looking at the screen that 657 00:30:42,980 --> 00:30:46,940 had the same number of items as the drumbeats, 658 00:30:46,940 --> 00:30:49,910 which Starkey takes as evidence showing that the kids are 659 00:30:49,910 --> 00:30:52,160 able to make this connection that both of these things 660 00:30:52,160 --> 00:30:56,470 are two in some abstract way, even though they're coming in 661 00:30:56,470 --> 00:30:57,470 in different modalities. 662 00:30:57,470 --> 00:30:59,630 One of these is visual and one of them is auditory. 663 00:31:02,390 --> 00:31:04,160 Babies, they're smarter than you think. 664 00:31:07,058 --> 00:31:09,956 AUDIENCE: My parents say that on the day that my brother 665 00:31:09,956 --> 00:31:17,511 switched from mother to the bottle, 666 00:31:17,511 --> 00:31:22,890 he was listening to the Handel's Messiah. 667 00:31:22,890 --> 00:31:24,357 AUDIENCE: The what? 668 00:31:24,357 --> 00:31:25,736 AUDIENCE: Handel's Messiah. 669 00:31:25,736 --> 00:31:26,777 You've never heard of it? 670 00:31:26,777 --> 00:31:28,470 It's the [INAUDIBLE] 671 00:31:28,470 --> 00:31:30,660 ABBY NOYCE: Not everybody is classical music people. 672 00:31:30,660 --> 00:31:32,126 That's OK. 673 00:31:32,126 --> 00:31:34,030 AUDIENCE: They like to tell a funny story. 674 00:31:34,030 --> 00:31:38,230 They say that my mom had to go out, 675 00:31:38,230 --> 00:31:42,790 so dad was stuck with the job of feeding my older brother. 676 00:31:42,790 --> 00:31:47,860 And then for the first part of the night, 677 00:31:47,860 --> 00:31:49,245 he'd refuse the milk bottle. 678 00:31:49,245 --> 00:31:51,974 And he actually used his tongue to push it out of his mouth. 679 00:31:51,974 --> 00:31:54,703 But then when they got to the part of the "Hallelujah 680 00:31:54,703 --> 00:31:58,675 Chorus," he'd calm down. 681 00:31:58,675 --> 00:31:59,625 ABBY NOYCE: Cute. 682 00:32:02,960 --> 00:32:04,598 All right. 683 00:32:04,598 --> 00:32:08,619 AUDIENCE: [INAUDIBLE] 684 00:32:08,619 --> 00:32:09,910 ABBY NOYCE: Moving right along. 685 00:32:09,910 --> 00:32:13,945 So we have this innate number sense of smallish numbers. 686 00:32:16,450 --> 00:32:18,010 The lack of having this is called 687 00:32:18,010 --> 00:32:22,469 dyscalculia, kind of equivalent to dyslexia, 688 00:32:22,469 --> 00:32:23,260 difficulty reading. 689 00:32:23,260 --> 00:32:26,980 Dyscalculia, inability to work with numbers. 690 00:32:26,980 --> 00:32:31,150 These are some cases by, oh, shoot, Brian-- 691 00:32:31,150 --> 00:32:33,070 something that starts with B-- 692 00:32:33,070 --> 00:32:36,370 who is at Cambridge in England, and has been working 693 00:32:36,370 --> 00:32:37,630 on this for a long time. 694 00:32:37,630 --> 00:32:40,450 Like I said, the Brits did a lot of this work 695 00:32:40,450 --> 00:32:42,940 in the last couple of decades really putting it together. 696 00:32:42,940 --> 00:32:46,740 And he started collecting interesting patients. 697 00:32:46,740 --> 00:32:49,990 So [INAUDIBLE] had an operation to remove a tumor 698 00:32:49,990 --> 00:32:51,190 in her left parietal lobe. 699 00:32:51,190 --> 00:32:53,852 And after that, her general intelligence 700 00:32:53,852 --> 00:32:55,060 remained more or less intact. 701 00:32:55,060 --> 00:32:58,810 She scored well within the normal range on most IQ tests. 702 00:32:58,810 --> 00:33:01,510 Her linguistic ability remained intact. 703 00:33:01,510 --> 00:33:02,950 She could talk to the examiners. 704 00:33:02,950 --> 00:33:06,490 She could answer all sorts of logical questions 705 00:33:06,490 --> 00:33:07,280 about her world. 706 00:33:10,360 --> 00:33:13,720 But she didn't have this sense of numbers anymore. 707 00:33:13,720 --> 00:33:16,180 She couldn't do that that's one thing, that's 708 00:33:16,180 --> 00:33:18,280 two things, that's three things, that subitizing, 709 00:33:18,280 --> 00:33:22,420 that instant recognition task that we can do. 710 00:33:22,420 --> 00:33:26,980 She couldn't do arithmetic, other than the fast facts 711 00:33:26,980 --> 00:33:31,090 that she'd managed to retain. 712 00:33:31,090 --> 00:33:34,670 So she could still recite off her multiplication tables, 713 00:33:34,670 --> 00:33:37,810 but it didn't make mean anything to her. 714 00:33:37,810 --> 00:33:42,160 And they tried to make the connection between the written 715 00:33:42,160 --> 00:33:46,750 numerals and this string of words that she knew. 716 00:33:46,750 --> 00:33:48,030 And it just didn't click. 717 00:33:48,030 --> 00:33:51,370 She couldn't make this comparison between 1 times 1 718 00:33:51,370 --> 00:33:54,900 is 1, and 1 times 2 is 2, with the written-out version of it, 719 00:33:54,900 --> 00:33:56,740 or the fact that it meant anything. 720 00:33:56,740 --> 00:33:59,152 It was just a string of words that she remembered. 721 00:34:03,370 --> 00:34:07,000 So some of the cases that you'll see with this 722 00:34:07,000 --> 00:34:10,480 are stroke victims, Alzheimer's victims, 723 00:34:10,480 --> 00:34:12,580 people who've had a tumor removed, 724 00:34:12,580 --> 00:34:14,590 people who've had some kind of damage 725 00:34:14,590 --> 00:34:17,050 to the left parietal lobe. 726 00:34:17,050 --> 00:34:19,179 There's also a few cases, they're much rarer, 727 00:34:19,179 --> 00:34:22,900 of people who are actually born without this sense of number. 728 00:34:22,900 --> 00:34:24,790 Again, this guy talks about a guy 729 00:34:24,790 --> 00:34:32,170 named Charles who can count, do simple arithmetic slowly 730 00:34:32,170 --> 00:34:35,790 on his fingers, like single digit addition and subtraction, 731 00:34:35,790 --> 00:34:38,510 uses a calculator for everything. 732 00:34:38,510 --> 00:34:43,600 If you show him, for example, 7 and 4, and say, 733 00:34:43,600 --> 00:34:44,980 which one is bigger? 734 00:34:44,980 --> 00:34:47,920 He can say, one, two, three, four. 735 00:34:47,920 --> 00:34:48,909 I get to four first. 736 00:34:48,909 --> 00:34:52,250 Seven must be bigger. 737 00:34:52,250 --> 00:34:54,736 For numbers that are outside of finger-counting range, 738 00:34:54,736 --> 00:34:55,360 he can't do it. 739 00:34:58,520 --> 00:35:02,770 So these are rare inabilities, but this 740 00:35:02,770 --> 00:35:05,200 does seem to happen that some people just-- 741 00:35:05,200 --> 00:35:07,570 this innate sense that most of us have, 742 00:35:07,570 --> 00:35:09,780 it's just not there for these people. 743 00:35:09,780 --> 00:35:11,980 And the extent to which people can work around this 744 00:35:11,980 --> 00:35:14,480 is pretty impressive to me. 745 00:35:14,480 --> 00:35:16,490 I'm one of those innately fast at math people, 746 00:35:16,490 --> 00:35:20,230 and I can't imagine not being able to do this. 747 00:35:20,230 --> 00:35:23,950 But definitely impressive. 748 00:35:27,330 --> 00:35:27,830 Counting. 749 00:35:27,830 --> 00:35:31,220 So this is moving from this innate, unnamed groups 750 00:35:31,220 --> 00:35:34,800 of number sense to this more precise sense of math, 751 00:35:34,800 --> 00:35:37,030 in which we're used to thinking about numbers. 752 00:35:37,030 --> 00:35:38,750 So counting let's you determine how many 753 00:35:38,750 --> 00:35:39,650 things are in a collection. 754 00:35:39,650 --> 00:35:41,191 How many chairs are in the first row? 755 00:35:41,191 --> 00:35:46,700 1, 2, 3, 4, 5, 6, and then one up front here. 756 00:35:46,700 --> 00:35:49,940 Really little kids have a hard time with this idea. 757 00:35:49,940 --> 00:35:52,070 You say, how many blocks are on the table? 758 00:35:52,070 --> 00:35:54,404 They'll say, five. 759 00:35:54,404 --> 00:35:55,820 You say, can you count the blocks? 760 00:35:55,820 --> 00:36:00,027 And they'll say, 1, 2, 3, 4, 5, 6, 7, 8, 9. 761 00:36:00,027 --> 00:36:02,360 And you'll say, OK, so how many blocks are on the table? 762 00:36:02,360 --> 00:36:04,145 They'll say, five. 763 00:36:07,364 --> 00:36:08,780 The number that you get up to when 764 00:36:08,780 --> 00:36:12,590 you're counting the collection and the number of items that 765 00:36:12,590 --> 00:36:19,450 are in the set are equivalent doesn't click for, again, 766 00:36:19,450 --> 00:36:20,840 two and three-year-old kids. 767 00:36:23,276 --> 00:36:24,650 Or they'll count something twice. 768 00:36:24,650 --> 00:36:29,350 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15. 769 00:36:33,920 --> 00:36:37,202 So counting is a learned skill. 770 00:36:37,202 --> 00:36:38,410 Little kids don't do it well. 771 00:36:38,410 --> 00:36:40,790 Bigger kids do do it well. 772 00:36:40,790 --> 00:36:44,120 Some cultures, even nowadays, have number words 773 00:36:44,120 --> 00:36:51,290 that go something like 1, 2, a lot, or 1, 2, 3, a lot, 774 00:36:51,290 --> 00:36:54,310 and really don't have counting in any kind of useful way 775 00:36:54,310 --> 00:36:57,140 any way that we would think of as counting. 776 00:36:57,140 --> 00:36:59,570 Presumably, a couple thousand years ago, 777 00:36:59,570 --> 00:37:01,670 there were a lot more cultures without these kind 778 00:37:01,670 --> 00:37:03,860 of intrinsic counting words. 779 00:37:03,860 --> 00:37:04,940 You can still hear this. 780 00:37:04,940 --> 00:37:06,648 If you look at languages, you'll probably 781 00:37:06,648 --> 00:37:10,130 notice that the numbers for 1, and 2, and 3 782 00:37:10,130 --> 00:37:12,920 behave differently than the numbers for later words. 783 00:37:12,920 --> 00:37:17,840 For example, in English, for numbers above four, 784 00:37:17,840 --> 00:37:19,940 how do you get the ordinal for that? 785 00:37:19,940 --> 00:37:22,160 Fourth, fifth, sixth. 786 00:37:22,160 --> 00:37:24,080 You just add that "th" right on to the end, 787 00:37:24,080 --> 00:37:25,560 and it goes right on up. 788 00:37:25,560 --> 00:37:27,560 But for the first three numbers, it's different. 789 00:37:27,560 --> 00:37:30,217 First, second, third. 790 00:37:30,217 --> 00:37:32,300 And you'll see this pattern in a lot of languages. 791 00:37:36,050 --> 00:37:37,840 You'll see it in French. 792 00:37:37,840 --> 00:37:40,130 Primer, seconder. 793 00:37:40,130 --> 00:37:44,214 And then it goes up with the "em" endings. 794 00:37:44,214 --> 00:37:46,130 I don't know any other number words that well. 795 00:37:46,130 --> 00:37:47,396 I used to. 796 00:37:47,396 --> 00:37:48,770 But anyway, you'll see this thing 797 00:37:48,770 --> 00:37:51,050 where the first three numbers are treated differently 798 00:37:51,050 --> 00:37:51,716 in the language. 799 00:37:54,470 --> 00:37:59,330 You'll see this pattern is taken as reasoning that these numbers 800 00:37:59,330 --> 00:38:01,626 are treated differently in the brain, 801 00:38:01,626 --> 00:38:03,500 or that they're older than the other numbers. 802 00:38:06,560 --> 00:38:09,380 So counting doesn't seem to be an innate human ability the way 803 00:38:09,380 --> 00:38:11,864 that that low-level number recognition is. 804 00:38:11,864 --> 00:38:13,530 Working with these higher level numbers, 805 00:38:13,530 --> 00:38:17,210 these more precise ways of thinking about numbers 806 00:38:17,210 --> 00:38:18,235 is different. 807 00:38:21,114 --> 00:38:23,030 Thinking about more precise levels of numbers. 808 00:38:23,030 --> 00:38:24,700 We're going to do the dots again, 809 00:38:24,700 --> 00:38:27,364 but I'm going to show you two numbers. 810 00:38:27,364 --> 00:38:28,780 One on the left, one on the right. 811 00:38:28,780 --> 00:38:31,090 AUDIENCE: What do you mean by bigger? 812 00:38:31,090 --> 00:38:33,490 ABBY NOYCE: Larger in quantity. 813 00:38:33,490 --> 00:38:36,970 If the numbers are 11 and four, 11 is bigger. 814 00:38:36,970 --> 00:38:40,120 They're printed in the same size font. 815 00:38:40,120 --> 00:38:42,010 Which reminds me of an interesting experiment 816 00:38:42,010 --> 00:38:43,090 that I'll tell you about after we do this. 817 00:38:43,090 --> 00:38:44,830 So this is going to be like the dots. 818 00:38:44,830 --> 00:38:45,860 One on the left, one on the right. 819 00:38:45,860 --> 00:38:48,160 I want you to raise your left hand or your right hand 820 00:38:48,160 --> 00:38:49,150 depending on which one is bigger. 821 00:38:49,150 --> 00:38:49,760 AUDIENCE: [INAUDIBLE] 822 00:38:49,760 --> 00:38:51,470 ABBY NOYCE: And I you to do it as fast as you can. 823 00:38:51,470 --> 00:38:53,170 It's the one that's on the same side as the number that's 824 00:38:53,170 --> 00:38:54,257 bigger. 825 00:38:54,257 --> 00:38:55,090 You can handle that. 826 00:38:55,090 --> 00:38:56,870 AUDIENCE: [INAUDIBLE] 827 00:38:56,870 --> 00:38:57,760 ABBY NOYCE: So if the one on the left is bigger, 828 00:38:57,760 --> 00:38:58,510 you raise your left hand. 829 00:38:58,510 --> 00:38:59,926 If the one on the right is bigger, 830 00:38:59,926 --> 00:39:01,410 you raise your right hand. 831 00:39:01,410 --> 00:39:02,330 And try and be fast. 832 00:39:02,330 --> 00:39:02,830 Ready? 833 00:39:02,830 --> 00:39:05,200 And I'm going to, again, show them pretty quickly. 834 00:39:05,200 --> 00:39:07,584 AUDIENCE: [LAUGHTER] 835 00:39:07,584 --> 00:39:08,500 ABBY NOYCE: All right. 836 00:39:08,500 --> 00:39:09,706 Want to do another one? 837 00:39:13,610 --> 00:39:14,586 All right. 838 00:39:18,990 --> 00:39:21,252 All right. 839 00:39:21,252 --> 00:39:22,228 [LAUGHS] 840 00:39:22,228 --> 00:39:23,269 AUDIENCE: Ow. 841 00:39:23,269 --> 00:39:24,060 ABBY NOYCE: You OK? 842 00:39:24,060 --> 00:39:25,226 AUDIENCE: That was my elbow. 843 00:39:25,226 --> 00:39:26,170 ABBY NOYCE: All right. 844 00:39:26,170 --> 00:39:28,179 Was the last one harder than the other ones? 845 00:39:28,179 --> 00:39:28,720 AUDIENCE: No. 846 00:39:28,720 --> 00:39:29,880 ABBY NOYCE: No? 847 00:39:29,880 --> 00:39:31,630 So I think this is one of the things where 848 00:39:31,630 --> 00:39:32,730 if you do a bunch of them, they get harder. 849 00:39:32,730 --> 00:39:34,840 One thing that you'll see in general 850 00:39:34,840 --> 00:39:36,550 is that as people do this, this seems 851 00:39:36,550 --> 00:39:39,742 to be harder than you might think, that often, 852 00:39:39,742 --> 00:39:41,200 once you get into the rhythm of it, 853 00:39:41,200 --> 00:39:42,910 you'll see response times of up to half 854 00:39:42,910 --> 00:39:45,800 a second to decide which one is bigger, which one is not. 855 00:39:45,800 --> 00:39:49,240 It also tends to be that for two numbers that are close 856 00:39:49,240 --> 00:39:52,780 together, people are slower and less accurate 857 00:39:52,780 --> 00:39:54,580 than for two numbers that are far apart. 858 00:39:54,580 --> 00:39:56,770 If the numbers are 1 and 25, people 859 00:39:56,770 --> 00:39:59,130 are very fast and very accurate. 860 00:39:59,130 --> 00:40:04,450 If the numbers are 24 and 25, people are slower and they mess 861 00:40:04,450 --> 00:40:06,250 up more. 862 00:40:06,250 --> 00:40:10,510 So one of the ideas underlying this 863 00:40:10,510 --> 00:40:14,420 is that we have this number line representation of numbers. 864 00:40:14,420 --> 00:40:17,650 We all remember learning number lines in school, right? 865 00:40:17,650 --> 00:40:19,960 And one of the things that happens 866 00:40:19,960 --> 00:40:23,800 is that when we're doing a task like this, 867 00:40:23,800 --> 00:40:26,920 we're conceiving numbers as being on this mental number 868 00:40:26,920 --> 00:40:31,220 line, so that numbers that are close together, 869 00:40:31,220 --> 00:40:33,040 it's hard to judge which one's which. 870 00:40:33,040 --> 00:40:36,460 Numbers that are far apart, it's easier. 871 00:40:36,460 --> 00:40:40,330 And this is not so much in the consciously doing it sense, 872 00:40:40,330 --> 00:40:42,550 but this is a model for explaining 873 00:40:42,550 --> 00:40:44,990 what kinds of mistakes people tend to make in this task. 874 00:40:55,039 --> 00:40:56,830 Other ways in which this mental number line 875 00:40:56,830 --> 00:41:01,590 idea seems to kick in is that people's mental number 876 00:41:01,590 --> 00:41:04,390 line seems to compress as it gets to bigger numbers. 877 00:41:04,390 --> 00:41:07,270 So if you think of it as the numbers from one to 10 878 00:41:07,270 --> 00:41:09,130 are nicely spaced out and really far apart, 879 00:41:09,130 --> 00:41:11,410 the numbers from 80 to 90 are closer together. 880 00:41:11,410 --> 00:41:13,000 They don't seem as different. 881 00:41:13,000 --> 00:41:14,860 82 and 85 don't seem as different 882 00:41:14,860 --> 00:41:17,338 as two and five do to a lot of people. 883 00:41:17,338 --> 00:41:21,007 AUDIENCE: [INAUDIBLE] 884 00:41:21,007 --> 00:41:22,840 ABBY NOYCE: Somebody did, which is the thing 885 00:41:22,840 --> 00:41:24,610 that Zechariah reminded me of. 886 00:41:24,610 --> 00:41:27,190 Somebody did something that's equivalent of a Stroop task. 887 00:41:27,190 --> 00:41:30,280 Remember, the Stroop task was saying the color of the ink, 888 00:41:30,280 --> 00:41:33,460 rather than the name of the color that was written. 889 00:41:33,460 --> 00:41:35,740 That was kind of tricky, kind of hard. 890 00:41:35,740 --> 00:41:37,120 Somebody did something equivalent 891 00:41:37,120 --> 00:41:44,270 where they'd show two numbers, one of which is a-- 892 00:41:44,270 --> 00:41:47,050 break the chalk-- is a three, and one of which is an eight, 893 00:41:47,050 --> 00:41:49,971 and say, which number is written in the bigger type? 894 00:41:49,971 --> 00:41:51,970 Don't tell me which number is the bigger number. 895 00:41:51,970 --> 00:41:53,910 Tell me which one is written in the bigger type. 896 00:41:53,910 --> 00:41:55,409 And just like with the Stroop tasks, 897 00:41:55,409 --> 00:41:58,000 people seem to have a hard time inhibiting 898 00:41:58,000 --> 00:42:01,420 that automatic processing of number as number, 899 00:42:01,420 --> 00:42:02,970 and thinking just about the typeface. 900 00:42:02,970 --> 00:42:04,470 They're a lot slower on it than they 901 00:42:04,470 --> 00:42:07,030 are in the number as number task, 902 00:42:07,030 --> 00:42:09,280 whereas if the three was little and the eight was big, 903 00:42:09,280 --> 00:42:11,200 people would be really fast at it. 904 00:42:11,200 --> 00:42:13,435 The number size, and the type size line up. 905 00:42:16,689 --> 00:42:17,480 So we know numbers. 906 00:42:17,480 --> 00:42:18,170 We're good with numbers. 907 00:42:18,170 --> 00:42:19,920 We have a lot of information about numbers 908 00:42:19,920 --> 00:42:21,180 stuffed into our brains. 909 00:42:25,330 --> 00:42:28,420 Let's talk for a minute about these representations 910 00:42:28,420 --> 00:42:31,040 of numbers. 911 00:42:31,040 --> 00:42:34,000 We all deal with numbers primarily as numerals. 912 00:42:34,000 --> 00:42:37,490 We've got words for them as well. 913 00:42:37,490 --> 00:42:42,160 And one of the things about numerals versus older writing 914 00:42:42,160 --> 00:42:43,646 systems, like Roman numerals, one 915 00:42:43,646 --> 00:42:46,360 of the things about Arabic numbers 916 00:42:46,360 --> 00:42:49,420 is that it lets you take advantage 917 00:42:49,420 --> 00:42:51,790 of your linguistic skills, your ability 918 00:42:51,790 --> 00:42:56,290 to learn a set of steps and use that 919 00:42:56,290 --> 00:42:58,750 to do things to these numbers. 920 00:42:58,750 --> 00:43:02,800 Multiply by 10 is add a 0, whereas in Roman numerals, 921 00:43:02,800 --> 00:43:05,530 you've got to switch everything over into different characters. 922 00:43:05,530 --> 00:43:07,990 So it's a different kind of processing when 923 00:43:07,990 --> 00:43:10,480 you're manipulating numbers. 924 00:43:10,480 --> 00:43:14,350 You can build these simple algorithms for steps 925 00:43:14,350 --> 00:43:16,810 for adding or multiplying that we all learned 926 00:43:16,810 --> 00:43:18,040 in like third grade, right? 927 00:43:20,800 --> 00:43:23,950 And unlike older number systems, you 928 00:43:23,950 --> 00:43:26,710 didn't have to do the calculation 929 00:43:26,710 --> 00:43:27,730 and then rewrite it all. 930 00:43:27,730 --> 00:43:29,380 It was just right there. 931 00:43:29,380 --> 00:43:32,030 So numerals are important. 932 00:43:32,030 --> 00:43:33,760 They're one of the big things that 933 00:43:33,760 --> 00:43:38,920 made arithmetic at least be accessible to the masses, 934 00:43:38,920 --> 00:43:41,110 higher level arithmetic. 935 00:43:41,110 --> 00:43:44,620 And they're, in some ways, simply a way 936 00:43:44,620 --> 00:43:46,300 of writing down number names. 937 00:43:46,300 --> 00:43:50,170 We all know that if I say 1,245, then 938 00:43:50,170 --> 00:43:54,430 you write it as 1, 2, 4, or 5. 939 00:43:54,430 --> 00:43:56,720 It's just a way of writing down the set of words. 940 00:43:56,720 --> 00:43:57,880 It's a shorthand. 941 00:43:57,880 --> 00:43:59,380 But on the other hand, numerals seem 942 00:43:59,380 --> 00:44:01,796 to be handled differently from the parts of the brain that 943 00:44:01,796 --> 00:44:05,350 handle language processing. 944 00:44:05,350 --> 00:44:08,590 And one example is a patient who had fairly early onset 945 00:44:08,590 --> 00:44:14,350 Alzheimer's who could read words, 946 00:44:14,350 --> 00:44:16,780 and who could read the written out names of numbers, even 947 00:44:16,780 --> 00:44:24,060 multi-digit numbers, was 50% accurate reading off 948 00:44:24,060 --> 00:44:25,120 a single-numeral names. 949 00:44:25,120 --> 00:44:28,660 Could read a one, or a two, or a three about half of the time. 950 00:44:28,660 --> 00:44:30,700 And not at all accurate at reading 951 00:44:30,700 --> 00:44:34,030 multi-digit numeral representations of numbers. 952 00:44:34,030 --> 00:44:36,970 So would be able to read 1,245. 953 00:44:36,970 --> 00:44:39,310 Would not be able to read this representation 954 00:44:39,310 --> 00:44:40,630 of the same number. 955 00:44:40,630 --> 00:44:44,200 So that's one piece of evidence that the word representation 956 00:44:44,200 --> 00:44:46,060 of this and the numeral representation 957 00:44:46,060 --> 00:44:48,880 are being handled by, in some ways, 958 00:44:48,880 --> 00:44:51,050 different processes in the brain. 959 00:44:51,050 --> 00:44:55,340 One of them is intact, one of them is not. 960 00:44:55,340 --> 00:44:58,480 And there's evidence of the opposite case, too. 961 00:44:58,480 --> 00:45:01,900 There's a woman who had surgery on her frontal lobes, again, 962 00:45:01,900 --> 00:45:04,810 a tumor removal surgery, and lost 963 00:45:04,810 --> 00:45:06,400 most of her linguistic ability. 964 00:45:06,400 --> 00:45:09,070 She can speak, but she can't read, can't write, 965 00:45:09,070 --> 00:45:11,260 can't deal with written language, 966 00:45:11,260 --> 00:45:14,320 can still deal with numerals, and can still do arithmetic, 967 00:45:14,320 --> 00:45:16,270 as long as it just has just the numerals 968 00:45:16,270 --> 00:45:17,540 lined up on top of each other. 969 00:45:17,540 --> 00:45:21,685 So if you think of it as 1, 2, 4, 5, plus-- 970 00:45:24,250 --> 00:45:26,220 she could do that. 971 00:45:26,220 --> 00:45:27,470 4 is 13. 972 00:45:27,470 --> 00:45:29,500 1 and 2 is-- 973 00:45:29,500 --> 00:45:31,810 that kind of arithmetic, she could still do, 974 00:45:31,810 --> 00:45:36,020 even though she couldn't read or write words at all. 975 00:45:36,020 --> 00:45:38,440 So that's what's called a double dissociation. 976 00:45:38,440 --> 00:45:41,770 You found a patient who has ability A, and not ability B, 977 00:45:41,770 --> 00:45:45,010 and another patient who has ability B, and not ability A. 978 00:45:45,010 --> 00:45:47,260 And so in the cognitive neuroscience world, 979 00:45:47,260 --> 00:45:50,740 a double dissociation like that is the gold standard 980 00:45:50,740 --> 00:45:53,020 for saying that these two abilities are 981 00:45:53,020 --> 00:45:54,280 independent in the brain. 982 00:45:54,280 --> 00:45:57,010 That ability A doesn't depend on ability B, 983 00:45:57,010 --> 00:45:58,810 and ability B doesn't depend on ability A. 984 00:45:58,810 --> 00:46:01,430 That they're handled by separate processing modules. 985 00:46:05,840 --> 00:46:06,802 AUDIENCE: Synapse. 986 00:46:06,802 --> 00:46:08,510 ABBY NOYCE: Quick review of [INAUDIBLE].. 987 00:46:08,510 --> 00:46:09,385 Yes, it is a synapse. 988 00:46:09,385 --> 00:46:10,499 Excellent! 989 00:46:10,499 --> 00:46:12,290 What kind of synapse is it if we're talking 990 00:46:12,290 --> 00:46:15,570 about long-term potentiation? 991 00:46:15,570 --> 00:46:19,460 What neurotransmitter is going to be here? 992 00:46:19,460 --> 00:46:22,620 The main excitatory transmitter in the brain. 993 00:46:22,620 --> 00:46:23,870 So it's going to be glutamate. 994 00:46:28,217 --> 00:46:32,090 AUDIENCE: [INAUDIBLE] 995 00:46:32,090 --> 00:46:35,660 ABBY NOYCE: So we know since we harped on it for the first week 996 00:46:35,660 --> 00:46:38,240 of class, when an action potential comes down the axon 997 00:46:38,240 --> 00:46:40,160 terminal, it causes neurotransmitter 998 00:46:40,160 --> 00:46:43,760 to be released into the gap between them into the synapse. 999 00:46:43,760 --> 00:46:48,200 And it triggers receptors on the postsynaptic cell. 1000 00:46:48,200 --> 00:46:50,870 For glutamate, there's two kinds of receptors 1001 00:46:50,870 --> 00:46:53,310 that we care about. 1002 00:46:53,310 --> 00:46:55,375 One of them is called the AMPA receptor. 1003 00:46:58,710 --> 00:47:02,890 [INAUDIBLE] the NMDA receptor. 1004 00:47:02,890 --> 00:47:08,290 N-methyl-D-aspartate something. 1005 00:47:08,290 --> 00:47:13,240 And the AMPA receptors aren't picky. 1006 00:47:13,240 --> 00:47:20,140 So an AMPA receptor, when the glutamate binds to it, 1007 00:47:20,140 --> 00:47:21,650 it opens. 1008 00:47:21,650 --> 00:47:25,130 And AMPA receptors are sodium channels. 1009 00:47:25,130 --> 00:47:28,450 That's a plus, not a t. 1010 00:47:28,450 --> 00:47:31,350 So when glutamate is released into the synapse, 1011 00:47:31,350 --> 00:47:34,600 AMPA receptors open, sodium comes in. 1012 00:47:34,600 --> 00:47:35,150 This is good. 1013 00:47:35,150 --> 00:47:36,520 So this is excitatory. 1014 00:47:36,520 --> 00:47:38,240 It's a positive ion coming into the-- 1015 00:47:38,240 --> 00:47:40,073 AUDIENCE: Don't you call them glutaminergic? 1016 00:47:40,073 --> 00:47:42,220 ABBY NOYCE: Glutamatergic, yes. 1017 00:47:42,220 --> 00:47:44,260 So a glutamate synapse. 1018 00:47:44,260 --> 00:47:49,750 Kind of like cholinergic, or dopaminergic. 1019 00:47:49,750 --> 00:47:50,710 So sodium comes in. 1020 00:47:50,710 --> 00:47:52,180 So this is an excitatory synapse. 1021 00:47:52,180 --> 00:47:54,120 It's a positive ion coming into the cell. 1022 00:47:54,120 --> 00:47:58,421 It's going to make the membrane of this cell less polarized. 1023 00:48:02,660 --> 00:48:09,080 NMDA receptors are pickier because in their normal state, 1024 00:48:09,080 --> 00:48:13,250 they've got this magnesium ion hanging out, 1025 00:48:13,250 --> 00:48:14,570 blocking that channel. 1026 00:48:14,570 --> 00:48:18,860 Magnesium ion is attracted to the membrane potential, 1027 00:48:18,860 --> 00:48:22,010 that because this cell is more negative inside than outside, 1028 00:48:22,010 --> 00:48:27,200 this magnesium ion is stuck into that channel there. 1029 00:48:27,200 --> 00:48:30,760 So even after the glutamate binds to this NMDA receptor, 1030 00:48:30,760 --> 00:48:37,550 this magnesium ion stays put, unless this cell 1031 00:48:37,550 --> 00:48:41,870 fires so many times that the membrane here 1032 00:48:41,870 --> 00:48:49,310 depolarizers all the way to about minus 35 millivolts. 1033 00:48:49,310 --> 00:48:51,530 So remember that resting potential is minus 70. 1034 00:48:54,270 --> 00:48:57,050 So we're going to about minus 35 millivolts. 1035 00:48:57,050 --> 00:49:01,460 And at that point, the electrostatic pressure 1036 00:49:01,460 --> 00:49:05,300 on this magnesium ion decreases because this cell 1037 00:49:05,300 --> 00:49:08,430 is no longer negative enough inside to hang on to it. 1038 00:49:08,430 --> 00:49:12,650 Magnesium ion pops off, floats away out into the extracellular 1039 00:49:12,650 --> 00:49:16,820 space, and then both sodium-- 1040 00:49:16,820 --> 00:49:18,170 that's a plus, not a t. 1041 00:49:18,170 --> 00:49:20,870 I Keep having this problem. 1042 00:49:20,870 --> 00:49:28,460 And calcium both can go through the NMDA channels. 1043 00:49:28,460 --> 00:49:32,570 NMDA are channels for both sodium, which is excitatory, 1044 00:49:32,570 --> 00:49:34,250 and calcium. 1045 00:49:34,250 --> 00:49:36,050 And calcium does cool stuff. 1046 00:49:36,050 --> 00:49:40,010 Calcium acts as a second messenger. 1047 00:49:40,010 --> 00:49:43,610 Goes inside the cell, and changes a bunch of things 1048 00:49:43,610 --> 00:49:47,576 about the cell's behavior. 1049 00:49:47,576 --> 00:49:48,950 So one of the things calcium does 1050 00:49:48,950 --> 00:49:52,330 is it activates a bunch of enzymes called protein kinases. 1051 00:49:57,950 --> 00:50:01,670 Kinases are enzymes that phosphorylate other proteins. 1052 00:50:01,670 --> 00:50:05,000 They reach out, they grab a phosphate group, 1053 00:50:05,000 --> 00:50:09,710 and a protein, and they glom, stick them together. 1054 00:50:09,710 --> 00:50:12,200 In these cases, there's a bunch of different kinases. 1055 00:50:12,200 --> 00:50:19,030 There's protein kinase A, protein kinase C, 1056 00:50:19,030 --> 00:50:22,660 and one called calcium calmodulin kinase. 1057 00:50:25,820 --> 00:50:28,010 So calcium is activating all of these. 1058 00:50:30,640 --> 00:50:37,870 And this calcium calmodulin kinase, CaMK, 1059 00:50:37,870 --> 00:50:40,480 has some direct effects on the behavior 1060 00:50:40,480 --> 00:50:43,780 of receptors at this cell. 1061 00:50:43,780 --> 00:50:49,480 First of all, the calcium calmodulin kinase, the CaMK, 1062 00:50:49,480 --> 00:50:51,160 remember, these phosphorylate things. 1063 00:50:51,160 --> 00:50:52,250 It goes up. 1064 00:50:52,250 --> 00:50:54,700 It will go to these AMPA receptors. 1065 00:50:54,700 --> 00:50:57,430 It sticks a phosphate group onto them. 1066 00:50:57,430 --> 00:51:01,190 And by changing the molecular structure of the receptor, 1067 00:51:01,190 --> 00:51:04,750 it causes it, when glutamate binds to it and it opens, 1068 00:51:04,750 --> 00:51:08,920 when sodium can flow in, it causes it to stay open longer. 1069 00:51:08,920 --> 00:51:11,140 So every time these AMPA receptors 1070 00:51:11,140 --> 00:51:14,500 open up after they've been phosphorylated, 1071 00:51:14,500 --> 00:51:17,710 more sodium gets in. 1072 00:51:17,710 --> 00:51:21,720 So each AMPA receptor has a bigger excitatory effect 1073 00:51:21,720 --> 00:51:23,720 than it did before that phosphorylation happens. 1074 00:51:27,704 --> 00:51:30,200 Who's lost? 1075 00:51:30,200 --> 00:51:31,193 Questions? 1076 00:51:31,193 --> 00:51:33,026 Things that you would like to have repeated? 1077 00:51:33,026 --> 00:51:36,428 AUDIENCE: I heard something that [INAUDIBLE].. 1078 00:51:41,780 --> 00:51:42,860 ABBY NOYCE: That's true. 1079 00:51:42,860 --> 00:51:45,460 It's probably not directly correlated to this. 1080 00:51:45,460 --> 00:51:47,210 It probably has more to do with the extent 1081 00:51:47,210 --> 00:51:50,360 that epinephrine and norepinephrine affect 1082 00:51:50,360 --> 00:51:52,400 your attentional systems so that you're 1083 00:51:52,400 --> 00:51:54,200 focusing lots of processing resources 1084 00:51:54,200 --> 00:51:58,874 on whatever is coming in so that things that occurred when 1085 00:51:58,874 --> 00:52:01,040 you're having some kind of strong emotional response 1086 00:52:01,040 --> 00:52:04,026 to something, for example, tend to be recalled very vividly. 1087 00:52:07,760 --> 00:52:10,000 How old are you guys on 9/11? 1088 00:52:10,000 --> 00:52:10,730 Middle school? 1089 00:52:10,730 --> 00:52:11,660 Elementary school? 1090 00:52:11,660 --> 00:52:12,910 That was a while ago. 1091 00:52:12,910 --> 00:52:14,890 Yeah, see, OK. 1092 00:52:14,890 --> 00:52:16,210 I was a senior in high school. 1093 00:52:16,210 --> 00:52:18,910 And I know exactly where I was, and how 1094 00:52:18,910 --> 00:52:21,800 I found out that this thing had happened, and boom. 1095 00:52:21,800 --> 00:52:25,330 It's this very vivid memory, because it 1096 00:52:25,330 --> 00:52:29,030 was very emotional, very, oh my gosh, what's going on, 1097 00:52:29,030 --> 00:52:31,230 the world is coming to an end kind of day. 1098 00:52:34,026 --> 00:52:35,430 AUDIENCE: [INAUDIBLE] 1099 00:52:35,430 --> 00:52:39,540 ABBY NOYCE: 2001. 1100 00:52:39,540 --> 00:52:40,040 Yeah. 1101 00:52:40,040 --> 00:52:41,990 So you guys are probably young enough that this didn't happen. 1102 00:52:41,990 --> 00:52:44,376 But I know that for like me and my immediate peer group, 1103 00:52:44,376 --> 00:52:46,250 this was one of these really defining moments 1104 00:52:46,250 --> 00:52:47,800 of our adolescence. 1105 00:52:47,800 --> 00:52:52,250 That bam, this thing happened and the world changed. 1106 00:52:52,250 --> 00:52:54,210 AUDIENCE: [INAUDIBLE] 1107 00:52:54,210 --> 00:52:57,330 ABBY NOYCE: [LAUGHS] It was September. 1108 00:52:57,330 --> 00:52:59,120 AUDIENCE: [INAUDIBLE] 1109 00:52:59,120 --> 00:53:00,390 ABBY NOYCE: OK, anyway. 1110 00:53:00,390 --> 00:53:02,006 So calcium calmodulin kinase. 1111 00:53:02,006 --> 00:53:02,880 AUDIENCE: [INAUDIBLE] 1112 00:53:02,880 --> 00:53:06,200 ABBY NOYCE: Back to the chemistry! 1113 00:53:06,200 --> 00:53:09,720 Calcium calmodulin kinase phosphorylates these AMPA 1114 00:53:09,720 --> 00:53:10,440 receptors. 1115 00:53:10,440 --> 00:53:14,760 Also, there might be other AMPA receptors hanging out 1116 00:53:14,760 --> 00:53:17,160 in the spine of the dendrite further up 1117 00:53:17,160 --> 00:53:19,030 that aren't on the membrane yet. 1118 00:53:19,030 --> 00:53:24,030 And CaMK basically goes and gets these, and brings them up 1119 00:53:24,030 --> 00:53:25,860 to the synaptic cleft. 1120 00:53:25,860 --> 00:53:29,490 So it encourages these receptors that 1121 00:53:29,490 --> 00:53:30,960 were not previously on the membrane 1122 00:53:30,960 --> 00:53:33,240 to move to the membrane, so that there 1123 00:53:33,240 --> 00:53:35,160 are more receptors on the membrane, which 1124 00:53:35,160 --> 00:53:37,809 makes this synapse then more sensitive to glutamate. 1125 00:53:37,809 --> 00:53:40,350 Of the glutamate that's released into the synapse, more of it 1126 00:53:40,350 --> 00:53:41,308 will bind to receptors. 1127 00:53:41,308 --> 00:53:43,075 It will have a bigger effect, even 1128 00:53:43,075 --> 00:53:45,450 without changing the amount of glutamate that's released. 1129 00:53:56,600 --> 00:54:00,500 The other thing that all of these kinases 1130 00:54:00,500 --> 00:54:08,580 will do, all three, is they'll find this protein called 1131 00:54:08,580 --> 00:54:11,930 CREB, which is-- 1132 00:54:11,930 --> 00:54:13,120 oh, shoot. 1133 00:54:13,120 --> 00:54:16,240 I knew it on Tuesday. 1134 00:54:16,240 --> 00:54:17,650 AUDIENCE: Can you just put CREB? 1135 00:54:17,650 --> 00:54:19,530 ABBY NOYCE: You can just put CREB, yes. 1136 00:54:19,530 --> 00:54:24,030 And CREB binds to specific regions on the DNA 1137 00:54:24,030 --> 00:54:26,470 and affects the transcription of those genes. 1138 00:54:26,470 --> 00:54:29,730 So it causes those particular proteins 1139 00:54:29,730 --> 00:54:32,910 that are being coded for by those regions of DNA. 1140 00:54:32,910 --> 00:54:35,820 It changes how they are expressed, whether more of them 1141 00:54:35,820 --> 00:54:38,730 are synthesized or fewer of them are synthesized, 1142 00:54:38,730 --> 00:54:40,950 which is why you'll see people saying that LTP, 1143 00:54:40,950 --> 00:54:43,890 long-term potentiation, has a short stage, which 1144 00:54:43,890 --> 00:54:45,690 is this stuff with the AMPA receptors 1145 00:54:45,690 --> 00:54:48,210 being phosphorylated and more AMPA receptors being moved 1146 00:54:48,210 --> 00:54:53,850 to the tip, and then a longer stage that has to do with gene 1147 00:54:53,850 --> 00:54:58,020 expression and protein synthesis, which takes 1148 00:54:58,020 --> 00:54:59,440 a few hours to really kick in. 1149 00:55:02,428 --> 00:55:07,470 AUDIENCE: [INAUDIBLE] 1150 00:55:07,470 --> 00:55:10,080 ABBY NOYCE: AMPA receptors and NMDA receptors 1151 00:55:10,080 --> 00:55:12,090 are both glutamatergic receptors. 1152 00:55:12,090 --> 00:55:13,751 They both respond to glutamate. 1153 00:55:13,751 --> 00:55:16,010 AUDIENCE: [INAUDIBLE] 1154 00:55:16,010 --> 00:55:17,380 ABBY NOYCE: Yeah, AMPA. 1155 00:55:17,380 --> 00:55:20,256 AUDIENCE: OK. 1156 00:55:20,256 --> 00:55:21,880 ABBY NOYCE: And AMPA receptors are just 1157 00:55:21,880 --> 00:55:23,490 your classic ionotropic receptor. 1158 00:55:23,490 --> 00:55:25,540 The neurotransmitter binds to them, they open up. 1159 00:55:25,540 --> 00:55:28,750 A particular kind of ion gets to go through their channel. 1160 00:55:28,750 --> 00:55:33,042 NMDA receptors are special because even 1161 00:55:33,042 --> 00:55:34,750 after the neurotransmitter binds to them, 1162 00:55:34,750 --> 00:55:38,410 they've still got that magnesium ion blocking them 1163 00:55:38,410 --> 00:55:40,660 until the cell depolarizes past a certain point, 1164 00:55:40,660 --> 00:55:42,986 so only when lots of AMPA receptors 1165 00:55:42,986 --> 00:55:44,860 have already opened up, or the AMPA receptors 1166 00:55:44,860 --> 00:55:46,568 have already been opened a bunch of times 1167 00:55:46,568 --> 00:55:48,190 to let all that sodium in. 1168 00:55:48,190 --> 00:55:49,840 So NMDA receptors are picky. 1169 00:55:49,840 --> 00:55:51,639 They only open up when there's been lots 1170 00:55:51,639 --> 00:55:52,930 of activity coming to the cell. 1171 00:55:56,350 --> 00:55:58,080 Does that make sense? 1172 00:55:58,080 --> 00:56:01,410 So remember, this is instantiating 1173 00:56:01,410 --> 00:56:05,080 this idea of neurons that fire together wire together. 1174 00:56:05,080 --> 00:56:10,230 So when the presynaptic neuron is 1175 00:56:10,230 --> 00:56:11,820 firing to the postsynaptic neuron, 1176 00:56:11,820 --> 00:56:14,650 and the postsynaptic neuron is really depolarized, 1177 00:56:14,650 --> 00:56:16,920 which means that it in turn is probably firing, 1178 00:56:16,920 --> 00:56:19,500 then that synapse gets strengthened. 1179 00:56:19,500 --> 00:56:23,370 But if the presynaptic neuron is firing 1180 00:56:23,370 --> 00:56:26,289 and the postsynaptic neuron only depolarizes a little bit, 1181 00:56:26,289 --> 00:56:27,330 then it won't strengthen. 1182 00:56:27,330 --> 00:56:30,720 It depends on when both neurons in the chain 1183 00:56:30,720 --> 00:56:34,710 are firing is where you'll see this strengthening 1184 00:56:34,710 --> 00:56:40,310 of the synapse, making it more receptive to that glutamate, 1185 00:56:40,310 --> 00:56:42,300 so that when the glutamate is released, then 1186 00:56:42,300 --> 00:56:46,980 it has a bigger effect on the postsynaptic neuron by putting 1187 00:56:46,980 --> 00:56:51,180 out more receptors, by strengthening these receptors, 1188 00:56:51,180 --> 00:56:52,520 by phosphorylating them. 1189 00:56:52,520 --> 00:56:54,510 That's what that CaM kinase is doing, 1190 00:56:54,510 --> 00:56:57,450 so that they stay open longer, so more sodium gets in. 1191 00:56:57,450 --> 00:56:59,370 All of these are changes that are allowing 1192 00:56:59,370 --> 00:57:01,700 that postsynaptic cell to be more sensitive, really, 1193 00:57:01,700 --> 00:57:02,700 to the presynaptic cell. 1194 00:57:06,900 --> 00:57:08,750 Are we good?