TimedText:How does your memory work.webm.en.srt
1 00:00:00,320 --> 00:00:04,319 do you remember what you had for lunch
2 00:00:02,720 --> 00:00:06,080 three days ago
3 00:00:04,319 --> 00:00:08,480 or whether you got out of your bed with
4 00:00:06,080 --> 00:00:10,800 your left or your right foot
5 00:00:08,480 --> 00:00:12,570 i will tell you why you forget how does
6 00:00:10,800 --> 00:00:21,230 your memory work
7 00:00:12,570 --> 00:00:21,230 [Music]
8 00:00:21,600 --> 00:00:25,199 memories are fuzzy
9 00:00:23,199 --> 00:00:28,160 you don't remember the details for
10 00:00:25,199 --> 00:00:29,920 example where you put your phone
11 00:00:28,160 --> 00:00:33,760 however often when you learn something
12 00:00:29,920 --> 00:00:36,719 new you can remember it for a long time
13 00:00:33,760 --> 00:00:38,559 but do you really remember everything
14 00:00:36,719 --> 00:00:40,719 in this lecture i will tell you exactly
15 00:00:38,559 --> 00:00:42,320 how your memory works and how sleep
16 00:00:40,719 --> 00:00:43,760 plays a very important role in
17 00:00:42,320 --> 00:00:46,079 remembering
18 00:00:43,760 --> 00:00:48,320 let's start at the very beginning
19 00:00:46,079 --> 00:00:50,480 what is a memory
20 00:00:48,320 --> 00:00:52,719 memories are believed to be a network of
21 00:00:50,480 --> 00:00:56,239 neurons that is active in the brain but
22 00:00:52,719 --> 00:00:58,800 let's go into that in a bit more detail
23 00:00:56,239 --> 00:01:01,120 neurons are the nerve cells in our brain
24 00:00:58,800 --> 00:01:03,440 and they can send and receive
25 00:01:01,120 --> 00:01:05,199 information in your brain
26 00:01:03,440 --> 00:01:06,960 in a way they are the communicators in
27 00:01:05,199 --> 00:01:08,960 our brain
28 00:01:06,960 --> 00:01:11,600 neurons can communicate with each other
29 00:01:08,960 --> 00:01:13,280 via electrical and chemical signals they
30 00:01:11,600 --> 00:01:15,439 send to each other
31 00:01:13,280 --> 00:01:17,920 in this way they can form networks or
32 00:01:15,439 --> 00:01:20,320 patterns in the brain
33 00:01:17,920 --> 00:01:22,560 a memory is such a unique network it's
34 00:01:20,320 --> 00:01:25,039 the neurons that are active at the time
35 00:01:22,560 --> 00:01:27,759 when something happens
36 00:01:25,040 --> 00:01:29,119 when you activate this exact same set of
37 00:01:27,759 --> 00:01:32,159 neurons again
38 00:01:29,119 --> 00:01:34,000 you literally relive the memory a
39 00:01:32,159 --> 00:01:36,240 perfect memory
40 00:01:34,000 --> 00:01:39,040 and the extent of the network of neurons
41 00:01:36,240 --> 00:01:41,839 that is reactivated determines the
42 00:01:39,040 --> 00:01:44,560 detail of the memory
43 00:01:41,840 --> 00:01:46,560 the network of neurons codes for memory
44 00:01:44,560 --> 00:01:47,840 and this happens throughout the entire
45 00:01:46,560 --> 00:01:49,439 brain
46 00:01:47,840 --> 00:01:51,280 however there are certain hubs in the
47 00:01:49,439 --> 00:01:54,960 brain that are more critical for
48 00:01:51,280 --> 00:01:57,119 encoding and processing of memories
49 00:01:54,960 --> 00:01:58,880 the first hub is the hippocampus you
50 00:01:57,119 --> 00:02:00,560 have one on the left side of the brain
51 00:01:58,880 --> 00:02:03,199 and one on the right
52 00:02:00,560 --> 00:02:06,079 this is the initial storage of memories
53 00:02:03,200 --> 00:02:08,000 when you experience or learn something
54 00:02:06,079 --> 00:02:08,959 this part of the brain encodes it as a
55 00:02:08,000 --> 00:02:11,520 memory
56 00:02:08,959 --> 00:02:14,160 the activation of certain urines
57 00:02:11,520 --> 00:02:16,160 and stores it exactly like this
58 00:02:14,160 --> 00:02:18,160 the main reason why we know that the
59 00:02:16,160 --> 00:02:20,000 hippocampus is important is because of
60 00:02:18,160 --> 00:02:22,560 henry malayasan
61 00:02:20,000 --> 00:02:24,080 in the 50s he had a very bad form of
62 00:02:22,560 --> 00:02:25,840 epilepsy
63 00:02:24,080 --> 00:02:28,000 epilepsy tends to occur in the
64 00:02:25,840 --> 00:02:30,080 hippocampus so they took out both
65 00:02:28,000 --> 00:02:32,000 hippocampi
66 00:02:30,080 --> 00:02:33,920 afterwards he could never remember new
67 00:02:32,000 --> 00:02:36,000 things again but he could remember
68 00:02:33,920 --> 00:02:38,879 really old things that were already
69 00:02:36,000 --> 00:02:40,239 stored in his long-term memory
70 00:02:38,879 --> 00:02:42,480 you can compare this to the main
71 00:02:40,239 --> 00:02:44,879 character in the movie memento or if you
72 00:02:42,480 --> 00:02:46,480 haven't seen that one to dory in finding
73 00:02:44,879 --> 00:02:49,120 nemo
74 00:02:46,480 --> 00:02:51,760 this example shows that new experiences
75 00:02:49,120 --> 00:02:54,400 you have daily and old memories are
76 00:02:51,760 --> 00:02:57,760 stored in different parts of the brain
77 00:02:54,400 --> 00:02:59,760 this is the second memory hub the cortex
78 00:02:57,760 --> 00:03:01,518 here memories are processed and
79 00:02:59,760 --> 00:03:03,440 consolidated
80 00:03:01,519 --> 00:03:06,400 when you think back to something you've
81 00:03:03,440 --> 00:03:07,760 learned do you also exactly remember how
82 00:03:06,400 --> 00:03:10,159 you learned it
83 00:03:07,760 --> 00:03:12,159 probably not even so you can perfectly
84 00:03:10,159 --> 00:03:14,640 recall the knowledge
85 00:03:12,159 --> 00:03:17,519 the memory for learning effect isn't the
86 00:03:14,640 --> 00:03:20,319 same as the fact that you can recall
87 00:03:17,519 --> 00:03:23,200 the fact memory is consolidated
88 00:03:20,319 --> 00:03:25,518 you know the capital of france is paris
89 00:03:23,200 --> 00:03:28,079 which is stored in your cortex but you
90 00:03:25,519 --> 00:03:30,400 don't recall the event where you learned
91 00:03:28,080 --> 00:03:32,319 this information
92 00:03:30,400 --> 00:03:34,640 the hippocampus stores information as
93 00:03:32,319 --> 00:03:36,958 event memories or sequences
94 00:03:34,640 --> 00:03:38,480 then if you consolidate to the cortex
95 00:03:36,959 --> 00:03:40,000 you abstract the information that is
96 00:03:38,480 --> 00:03:42,319 most critical
97 00:03:40,000 --> 00:03:44,480 we can show this in an example
98 00:03:42,319 --> 00:03:46,480 when you first visit a city and walk to
99 00:03:44,480 --> 00:03:49,280 your destination you see different
100 00:03:46,480 --> 00:03:51,119 buildings and landmarks around you
101 00:03:49,280 --> 00:03:52,720 this walk is stored in the hippocampus
102 00:03:51,120 --> 00:03:54,720 as an event
103 00:03:52,720 --> 00:03:58,000 part of this information is consolidated
104 00:03:54,720 --> 00:03:59,920 to the cortex as part of a spatial map
105 00:03:58,000 --> 00:04:02,000 then when you take a new route this
106 00:03:59,920 --> 00:04:04,480 information is abstracted and added to
107 00:04:02,000 --> 00:04:06,480 the map as well
108 00:04:04,480 --> 00:04:08,159 now when you hear about a new location
109 00:04:06,480 --> 00:04:10,959 let's say a restaurant that just opened
110 00:04:08,159 --> 00:04:13,920 up in the same city you can easily find
111 00:04:10,959 --> 00:04:15,120 the location remember it for a long time
112 00:04:13,920 --> 00:04:16,798 why
113 00:04:15,120 --> 00:04:19,280 because you already have a map in your
114 00:04:16,798 --> 00:04:21,358 cortex you can refer to
115 00:04:19,279 --> 00:04:23,039 we don't only use our memory systems for
116 00:04:21,358 --> 00:04:26,000 spatial information like maps and
117 00:04:23,040 --> 00:04:28,720 locations but we can also create general
118 00:04:26,000 --> 00:04:30,960 knowledge networks for example in my
119 00:04:28,720 --> 00:04:33,040 life i've encountered different birds
120 00:04:30,960 --> 00:04:35,680 these events are consolidated or
121 00:04:33,040 --> 00:04:38,000 abstracted to my cortex
122 00:04:35,680 --> 00:04:39,280 the memory is then in a downsized
123 00:04:38,000 --> 00:04:40,880 version
124 00:04:39,280 --> 00:04:42,960 when consolidating memories from the
125 00:04:40,880 --> 00:04:44,960 hippocampus to the cortex
126 00:04:42,960 --> 00:04:47,280 my brain extracts overlapping
127 00:04:44,960 --> 00:04:49,599 information about the concept of birds
128 00:04:47,280 --> 00:04:51,440 they can fly they have wings they have
129 00:04:49,600 --> 00:04:53,440 feathers
130 00:04:51,440 --> 00:04:56,160 the advantage of these abstract memories
131 00:04:53,440 --> 00:04:58,560 or semantic memories is that when you
132 00:04:56,160 --> 00:04:59,600 encounter a new bird that looks slightly
133 00:04:58,560 --> 00:05:01,680 different
134 00:04:59,600 --> 00:05:02,880 you generalize and still recognize that
135 00:05:01,680 --> 00:05:04,800 a bird
136 00:05:02,880 --> 00:05:06,320 and you can rapidly update your
137 00:05:04,800 --> 00:05:08,479 knowledge network
138 00:05:06,320 --> 00:05:11,360 if your semantic memory is very large
139 00:05:08,479 --> 00:05:14,240 and extensive you can update it even
140 00:05:11,360 --> 00:05:16,240 more rapidly and that's what you get
141 00:05:14,240 --> 00:05:18,160 when we call schema
142 00:05:16,240 --> 00:05:22,560 it's a bit like a mind map with all
143 00:05:18,160 --> 00:05:24,560 information connected to the one concept
144 00:05:22,560 --> 00:05:25,759 we don't have memories for the sake of
145 00:05:24,560 --> 00:05:27,520 nostalgia
146 00:05:25,759 --> 00:05:29,039 for reminiscing and perfectly
147 00:05:27,520 --> 00:05:30,880 remembering the events stored in the
148 00:05:29,039 --> 00:05:33,039 hippocampus
149 00:05:30,880 --> 00:05:34,880 the purpose of our memory is to create a
150 00:05:33,039 --> 00:05:36,560 model of the world
151 00:05:34,880 --> 00:05:38,479 so you can deal with it
152 00:05:36,560 --> 00:05:40,800 so memory is not about remembering the
153 00:05:38,479 --> 00:05:43,039 past it's about being prepared for the
154 00:05:40,800 --> 00:05:44,800 future
155 00:05:43,039 --> 00:05:47,039 it's a good thing you don't recall every
156 00:05:44,800 --> 00:05:48,880 bit of detail the more details you
157 00:05:47,039 --> 00:05:51,120 remember the less you can adapt to
158 00:05:48,880 --> 00:05:53,039 things that are different
159 00:05:51,120 --> 00:05:55,360 in real life it's very unlikely you're
160 00:05:53,039 --> 00:05:56,560 going to encounter the same thing twice
161 00:05:55,360 --> 00:05:58,319 that's why you want to have the
162 00:05:56,560 --> 00:06:00,800 knowledge to stop for example at a red
163 00:05:58,319 --> 00:06:03,120 light and it should not be a specific
164 00:06:00,800 --> 00:06:04,880 red light but you want any red light no
165 00:06:03,120 --> 00:06:06,560 matter what it looks like or how the
166 00:06:04,880 --> 00:06:09,600 streets are built
167 00:06:06,560 --> 00:06:12,639 so this case a non-perfect memory is
168 00:06:09,600 --> 00:06:15,120 perfectly suited
169 00:06:12,639 --> 00:06:17,039 now when we talk about memory we also
170 00:06:15,120 --> 00:06:18,880 have to talk about sleep
171 00:06:17,039 --> 00:06:21,039 memories are events initially stored in
172 00:06:18,880 --> 00:06:22,639 the hippocampus and later consolidated
173 00:06:21,039 --> 00:06:25,120 to the cortex
174 00:06:22,639 --> 00:06:27,680 sleep is exactly the moment when you
175 00:06:25,120 --> 00:06:29,360 update your memory systems
176 00:06:27,680 --> 00:06:31,039 compared to desktop
177 00:06:29,360 --> 00:06:32,880 throughout the day you store all the
178 00:06:31,039 --> 00:06:35,840 information and files on your desktop no
179 00:06:32,880 --> 00:06:37,280 matter what at the end of the day
180 00:06:35,840 --> 00:06:39,198 you sift and sort through the
181 00:06:37,280 --> 00:06:40,960 information decide what you want to
182 00:06:39,199 --> 00:06:42,639 throw in the trash bin and what
183 00:06:40,960 --> 00:06:45,280 information you want to kind of keep and
184 00:06:42,639 --> 00:06:46,880 put safely away in a folder
185 00:06:45,280 --> 00:06:50,638 then afterwards you can reformat the
186 00:06:46,880 --> 00:06:52,000 desktop so it's ready for a new day
187 00:06:50,639 --> 00:06:53,919 after all you don't want to remember
188 00:06:52,000 --> 00:06:55,360 what sandwich you ate yesterday but you
189 00:06:53,919 --> 00:06:57,840 might want to store the knowledge you
190 00:06:55,360 --> 00:06:59,759 learn by watching this video in your
191 00:06:57,840 --> 00:07:02,000 long-term memory
192 00:06:59,759 --> 00:07:03,599 so the sifting and sorting or the
193 00:07:02,000 --> 00:07:06,080 consolidation
194 00:07:03,599 --> 00:07:08,240 typically happens during sleep
195 00:07:06,080 --> 00:07:11,120 it's a quiet period of rest when there
196 00:07:08,240 --> 00:07:13,280 is no new outside information coming in
197 00:07:11,120 --> 00:07:14,479 this might be one of the main reasons we
198 00:07:13,280 --> 00:07:16,719 sleep
199 00:07:14,479 --> 00:07:18,400 after all we spent roughly 30 percent of
200 00:07:16,720 --> 00:07:20,400 our lifetime sleeping
201 00:07:18,400 --> 00:07:23,280 while we don't know the exact reason for
202 00:07:20,400 --> 00:07:24,960 this this could be part of it
203 00:07:23,280 --> 00:07:26,638 just imagine what it'd be like if i
204 00:07:24,960 --> 00:07:27,680 would be consolidating memories while
205 00:07:26,639 --> 00:07:29,840 awake
206 00:07:27,680 --> 00:07:32,080 while i'm talking to you my brain would
207 00:07:29,840 --> 00:07:34,880 then at the very same time reactivate
208 00:07:32,080 --> 00:07:36,639 memories from earlier during the day
209 00:07:34,880 --> 00:07:39,440 so while i'm trying to pay attention and
210 00:07:36,639 --> 00:07:41,440 store new events i'm also replaying a
211 00:07:39,440 --> 00:07:42,400 memory of my son playing early this
212 00:07:41,440 --> 00:07:44,080 morning
213 00:07:42,400 --> 00:07:46,239 that would really distract me and
214 00:07:44,080 --> 00:07:47,919 interfere with new memories i'm trying
215 00:07:46,240 --> 00:07:49,680 to create
216 00:07:47,919 --> 00:07:51,840 during our sleep we go through different
217 00:07:49,680 --> 00:07:54,160 sleep stages in general we can make a
218 00:07:51,840 --> 00:07:56,000 distinction between rem sleep and
219 00:07:54,160 --> 00:07:58,000 non-rem sleep
220 00:07:56,000 --> 00:08:00,720 rem sleep is noticeable because of the
221 00:07:58,000 --> 00:08:02,479 rapid eye movements you're making
222 00:08:00,720 --> 00:08:04,960 it's also the sleep stage you can
223 00:08:02,479 --> 00:08:07,840 remember most dreams from these are
224 00:08:04,960 --> 00:08:10,318 weird and emotional dreams
225 00:08:07,840 --> 00:08:12,719 everything else is non-rem sleep
226 00:08:10,319 --> 00:08:14,400 with different substations such as light
227 00:08:12,720 --> 00:08:16,319 and deep sleep
228 00:08:14,400 --> 00:08:18,799 the processing happens during this
229 00:08:16,319 --> 00:08:20,960 non-rem sleep
230 00:08:18,800 --> 00:08:22,319 in the non-rem sleep we can see activity
231 00:08:20,960 --> 00:08:24,799 in the cortex
232 00:08:22,319 --> 00:08:27,360 and very fast and short oscillations of
233 00:08:24,800 --> 00:08:28,879 the hippocampus the two brain areas are
234 00:08:27,360 --> 00:08:30,160 communicating
235 00:08:28,879 --> 00:08:32,479 and this is when we believe the
236 00:08:30,160 --> 00:08:34,640 hippocampus reactivates the neurons of a
237 00:08:32,479 --> 00:08:37,200 memory so the exact same network of
238 00:08:34,640 --> 00:08:39,279 neurons active during an event
239 00:08:37,200 --> 00:08:41,680 the cortex can then update your
240 00:08:39,279 --> 00:08:44,000 long-term memory system
241 00:08:41,679 --> 00:08:46,000 this also explains how you can get
242 00:08:44,000 --> 00:08:49,519 better in a motor skill like dancing a
243 00:08:46,000 --> 00:08:51,360 routine or playing a guitar overnight
244 00:08:49,519 --> 00:08:52,959 it's not you don't literally learn it so
245 00:08:51,360 --> 00:08:54,720 it's not as if you can place a book
246 00:08:52,959 --> 00:08:57,439 under your pillow and learn
247 00:08:54,720 --> 00:08:59,040 but sleep is very important
248 00:08:57,440 --> 00:09:01,680 during the day time you practice a
249 00:08:59,040 --> 00:09:03,519 certain routine movement of your body or
250 00:09:01,680 --> 00:09:04,880 the fingers on your guitar
251 00:09:03,519 --> 00:09:06,480 this routine is stored in the
252 00:09:04,880 --> 00:09:08,720 hippocampus
253 00:09:06,480 --> 00:09:11,600 then overnight you consolidate the
254 00:09:08,720 --> 00:09:14,480 crucial information to your cortex
255 00:09:11,600 --> 00:09:17,200 so when you try again the next day you
256 00:09:14,480 --> 00:09:18,880 are better at accessing the information
257 00:09:17,200 --> 00:09:20,560 and using it
258 00:09:18,880 --> 00:09:22,880 so it might feel like you've gotten
259 00:09:20,560 --> 00:09:26,000 better overnight
260 00:09:22,880 --> 00:09:28,800 we know this works but learning how
261 00:09:26,000 --> 00:09:30,720 it works is still a challenge
262 00:09:28,800 --> 00:09:33,279 a really good way of learning more about
263 00:09:30,720 --> 00:09:34,480 memory and sleep is to let rats run
264 00:09:33,279 --> 00:09:36,880 through a maze
265 00:09:34,480 --> 00:09:38,720 in my lab we have a maze where we let
266 00:09:36,880 --> 00:09:41,279 rat find some food
267 00:09:38,720 --> 00:09:42,720 in this video the rat already knows the
268 00:09:41,279 --> 00:09:45,439 food location
269 00:09:42,720 --> 00:09:46,640 he gets put into the maze briefly thinks
270 00:09:45,440 --> 00:09:48,720 where am i
271 00:09:46,640 --> 00:09:50,319 and once he recognizes where he is he'll
272 00:09:48,720 --> 00:09:53,279 start running directly to the goal
273 00:09:50,320 --> 00:09:55,839 location in the most efficient manner
274 00:09:53,279 --> 00:09:58,480 by running different trials he created a
275 00:09:55,839 --> 00:10:01,040 spatial map just like in the example of
276 00:09:58,480 --> 00:10:03,519 walking through a new city
277 00:10:01,040 --> 00:10:05,839 this is really impressive even more
278 00:10:03,519 --> 00:10:08,000 because the maze see-through so there's
279 00:10:05,839 --> 00:10:09,839 a bit of a mirroring effect making it
280 00:10:08,000 --> 00:10:11,920 more difficult
281 00:10:09,839 --> 00:10:15,440 the first time we were even surprised at
282 00:10:11,920 --> 00:10:17,599 how good rats can find their way
283 00:10:15,440 --> 00:10:19,440 but why do we do these tests where we
284 00:10:17,600 --> 00:10:20,959 let them run the maze and then sleep for
285 00:10:19,440 --> 00:10:23,120 four hours
286 00:10:20,959 --> 00:10:24,079 we know a lot about memory processing
287 00:10:23,120 --> 00:10:27,200 and sleep
288 00:10:24,079 --> 00:10:29,279 but there is so much more we don't know
289 00:10:27,200 --> 00:10:32,000 and researching it provides quite some
290 00:10:29,279 --> 00:10:34,160 problems the hippocampus which is
291 00:10:32,000 --> 00:10:37,040 important for memory it's deep in the
292 00:10:34,160 --> 00:10:38,480 brain so deep it's nearly impossible to
293 00:10:37,040 --> 00:10:39,839 place electrodes to measure brain
294 00:10:38,480 --> 00:10:42,160 activity
295 00:10:39,839 --> 00:10:44,880 and that's exactly what we need we need
296 00:10:42,160 --> 00:10:47,040 to be able to tell what every single
297 00:10:44,880 --> 00:10:49,279 neuron is doing
298 00:10:47,040 --> 00:10:51,360 in rats we can get deep enough to place
299 00:10:49,279 --> 00:10:52,959 those electrodes but don't worry it
300 00:10:51,360 --> 00:10:55,200 doesn't harm them
301 00:10:52,959 --> 00:10:57,518 but through the electrodes i can exactly
302 00:10:55,200 --> 00:10:59,920 measure the activation of neurons during
303 00:10:57,519 --> 00:11:02,720 the run and then see the exact same
304 00:10:59,920 --> 00:11:05,040 neurons reactivated during sleep
305 00:11:02,720 --> 00:11:07,519 this shows that indeed the reactivation
306 00:11:05,040 --> 00:11:09,839 is key to processing of memories during
307 00:11:07,519 --> 00:11:09,839 sleep
308 00:11:10,000 --> 00:11:14,480 another reason i work with rats is
309 00:11:11,680 --> 00:11:16,959 because i want to research if and when
310 00:11:14,480 --> 00:11:19,120 memories separate from the hippocampus
311 00:11:16,959 --> 00:11:20,959 if a memory of the maize and the food
312 00:11:19,120 --> 00:11:24,079 location is consolidated from that
313 00:11:20,959 --> 00:11:26,079 initial storage to the long-term storage
314 00:11:24,079 --> 00:11:27,599 they don't need the hippocampus to find
315 00:11:26,079 --> 00:11:29,599 their way
316 00:11:27,600 --> 00:11:31,440 to be able to test this we need to be
317 00:11:29,600 --> 00:11:33,839 able to temporarily inactivate the
318 00:11:31,440 --> 00:11:35,680 hippocampus and in rodents we can do
319 00:11:33,839 --> 00:11:37,279 that with either light or drugs that are
320 00:11:35,680 --> 00:11:39,120 locally given
321 00:11:37,279 --> 00:11:41,519 so you could turn off the hippocampus
322 00:11:39,120 --> 00:11:43,279 for a really short time
323 00:11:41,519 --> 00:11:46,959 exactly when the rat is tries to
324 00:11:43,279 --> 00:11:46,959 remember the location of the food
325 00:11:47,200 --> 00:11:50,959 why do we do these tests and learn how
326 00:11:48,880 --> 00:11:52,880 our memory systems work
327 00:11:50,959 --> 00:11:54,959 our memory is critical just imagine what
328 00:11:52,880 --> 00:11:57,200 your life would be without them
329 00:11:54,959 --> 00:11:59,040 a disease that is commonly and justly
330 00:11:57,200 --> 00:12:00,480 associated with memory loss is
331 00:11:59,040 --> 00:12:02,319 alzheimer's
332 00:12:00,480 --> 00:12:03,839 one of the first mechanisms to be
333 00:12:02,320 --> 00:12:05,200 affected in the conservation of new
334 00:12:03,839 --> 00:12:07,040 memories
335 00:12:05,200 --> 00:12:09,680 so if we learn more about how memory
336 00:12:07,040 --> 00:12:11,680 works exactly also during sleep
337 00:12:09,680 --> 00:12:12,880 it can hopefully help patients in the
338 00:12:11,680 --> 00:12:15,040 future
339 00:12:12,880 --> 00:12:17,120 another reason why we want to learn more
340 00:12:15,040 --> 00:12:19,519 about how our memories work apart from
341 00:12:17,120 --> 00:12:22,800 just understanding and pure knowledge
342 00:12:19,519 --> 00:12:25,120 is to build better computers
343 00:12:22,800 --> 00:12:27,439 computers are very good at remembering
344 00:12:25,120 --> 00:12:29,279 every last detail exactly what our
345 00:12:27,440 --> 00:12:31,279 brains can't do
346 00:12:29,279 --> 00:12:33,760 but where we can consolidate and
347 00:12:31,279 --> 00:12:35,519 generalize to learn computers don't tend
348 00:12:33,760 --> 00:12:37,920 to be very creative
349 00:12:35,519 --> 00:12:40,880 so by understanding the processes in our
350 00:12:37,920 --> 00:12:44,399 brains we can build better computers
351 00:12:40,880 --> 00:12:45,839 that are better suited for learning
352 00:12:44,399 --> 00:12:46,800 we started this lecture with the
353 00:12:45,839 --> 00:12:49,120 question
354 00:12:46,800 --> 00:12:50,880 how does your memory work
355 00:12:49,120 --> 00:12:53,040 what we saw is that a memory is a
356 00:12:50,880 --> 00:12:54,880 network of neurons stored as an event in
357 00:12:53,040 --> 00:12:57,439 the hippocampus
358 00:12:54,880 --> 00:13:00,880 when remembering the exact event the
359 00:12:57,440 --> 00:13:03,200 same set of neurons is activated
360 00:13:00,880 --> 00:13:04,720 during our sleep we consolidate memories
361 00:13:03,200 --> 00:13:06,720 to the cortex
362 00:13:04,720 --> 00:13:08,880 in a way we filter out the important
363 00:13:06,720 --> 00:13:11,120 information and neatly stored for
364 00:13:08,880 --> 00:13:13,600 example in a knowledge network which is
365 00:13:11,120 --> 00:13:16,000 placed in the cortex
366 00:13:13,600 --> 00:13:18,320 there is still a lot we don't know but i
367 00:13:16,000 --> 00:13:20,880 hope that one day we figure out exactly
368 00:13:18,320 --> 00:13:23,200 how our memory works and how sleep is
369 00:13:20,880 --> 00:13:25,439 crucial in the process
370 00:13:23,200 --> 00:13:29,340 as always i would like to thank my rats
371 00:13:25,440 --> 00:13:40,229 for sleeping and you for staying awake
372 00:13:29,340 --> 00:13:40,229 [Music]
373 00:13:43,360 --> 00:13:45,440 you