Can you outsmart this logical fallacy? - Alex Gendler

1,988,606 views ใƒป 2019-11-25

TED-Ed


์•„๋ž˜ ์˜๋ฌธ์ž๋ง‰์„ ๋”๋ธ”ํด๋ฆญํ•˜์‹œ๋ฉด ์˜์ƒ์ด ์žฌ์ƒ๋ฉ๋‹ˆ๋‹ค.

๋ฒˆ์—ญ: Jumee Lim ๊ฒ€ํ† : Jungmin Hwang
00:06
Meet Lucy.
0
6810
1190
๋ฃจ์‹œ๋ฅผ ๋งŒ๋‚˜๋ณด์„ธ์š”.
00:08
She was a math major in college,
1
8000
1980
๊ทธ๋…€๋Š” ๋Œ€ํ•™์—์„œ ์ˆ˜ํ•™์„ ์ „๊ณตํ–ˆ๊ณ 
00:09
and aced all her courses in probability and statistics.
2
9980
4130
ํ™•๋ฅ ๊ณผ ํ†ต๊ณ„ํ•™์„ ๋ชจ๋‘ ์„ญ๋ ตํ–ˆ์Šต๋‹ˆ๋‹ค.
00:14
Which do you think is more likely: that Lucy is a portrait artist,
3
14110
4352
๋ฃจ์‹œ๊ฐ€ ์ดˆ์ƒ์˜ˆ์ˆ ๊ฐ€๋ผ๋Š” ๊ฒƒ๊ณผ ํฌ์ปค๋„ ํ•˜๋Š” ์ดˆ์ƒ์˜ˆ์ˆ ๊ฐ€๋ผ๋Š” ๊ฒƒ ์ค‘
00:18
or that Lucy is a portrait artist who also plays poker?
4
18462
5028
์–ด๋Š ์ชฝ์ด ๋” ์‚ฌ์‹ค ๊ฐ™๋‹ค๊ณ  ์ƒ๊ฐํ•˜์‹œ๋‚˜์š”?
00:23
In studies of similar questions, up to 80 percent of participants
5
23490
4016
์œ ์‚ฌํ•œ ์งˆ๋ฌธ์„ ๋‹ค๋ฃฌ ์—ฐ๊ตฌ์—์„œ ์ฐธ๊ฐ€์ž ์ค‘ ์ตœ๋Œ€ 80%๊ฐ€
00:27
chose the equivalent of the second statement:
6
27506
2504
๋ฃจ์‹œ๊ฐ€ ํฌ์ปคํ•˜๋Š” ์ดˆ์ƒ์˜ˆ์ˆ ๊ฐ€๋ผ๋Š”
00:30
that Lucy is a portrait artist who also plays poker.
7
30010
3636
๋‘ ๋ฒˆ์งธ ๋ณด๊ธฐ์— ํ•ด๋‹นํ•˜๋Š” ๊ฒƒ์„ ๋‹ต์œผ๋กœ ์„ ํƒํ–ˆ์Šต๋‹ˆ๋‹ค.
00:33
After all, nothing we know about Lucy suggests an affinity for art,
8
33646
4722
๋ฃจ์‹œ๊ฐ€ ์˜ˆ์ˆ ์„ ์ข‹์•„ํ•˜๋Š”์ง€๋Š” ์•Œ ์ˆ˜ ์—†์ง€๋งŒ,
00:38
but statistics and probability are useful in poker.
9
38368
3812
ํ†ต๊ณ„์™€ ํ™•๋ฅ ์€ ํฌ์ปค์— ์œ ์šฉํ•˜๊ฒŒ ์“ฐ์ด๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
00:42
And yet, this is the wrong answer.
10
42180
2558
๊ทธ๋Ÿฌ๋‚˜ ์ด๋Š” ์˜ค๋‹ต์ž…๋‹ˆ๋‹ค.
00:44
Look at the options again.
11
44738
1790
์„ ํƒ์ง€๋ฅผ ๋‹ค์‹œ ๋ณด์‹œ์ฃ .
00:46
How do we know the first statement is more likely to be true?
12
46528
3740
์ฒซ ๋ฒˆ์งธ ์ง„์ˆ ์ด ์‚ฌ์‹ค์ผ ํ™•๋ฅ ์ด ๋” ๋†’๋‹ค๊ณ  ์–ด๋–ป๊ฒŒ ์•Œ ์ˆ˜ ์žˆ์„๊นŒ์š”?
00:50
Because itโ€™s a less specific version of the second statement.
13
50268
4180
๋‘ ๋ฒˆ์งธ ์ง„์ˆ ๋ณด๋‹ค ๋” ํฌ๊ด„์ ์ด๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
00:54
Saying that Lucy is a portrait artist doesnโ€™t make any claims
14
54448
3670
๋ฃจ์‹œ๊ฐ€ ์ดˆ์ƒ์˜ˆ์ˆ ๊ฐ€๋ผ๋Š” ์ง„์ˆ ์€ ๊ทธ๋…€๊ฐ€ ์ถ”๊ฐ€์ ์œผ๋กœ ํ•  ์ˆ˜ ์žˆ๋Š” ๊ฒƒ๊ณผ
00:58
about what else she might or might not do.
15
58118
3500
ํ•  ์ˆ˜ ์—†๋Š” ๊ฒƒ์ด ๋ฌด์—‡์ธ์ง€ ์•Œ๋ ค ์ฃผ์ง€ ์•Š์Šต๋‹ˆ๋‹ค.
01:01
But even though itโ€™s far easier to imagine her playing poker than making art
16
61618
4890
๊ทธ๋…€๊ฐ€ ์˜ˆ์ˆ ํ•˜๋Š” ๋ชจ์Šต๋ณด๋‹ค ํฌ์ปคํ•˜๋Š” ์žฅ๋ฉด์„ ์ƒ์ƒํ•˜๋Š” ๊ฒƒ์ด ๋” ์‰ฝ๋‹ค ํ•ด๋„
01:06
based on the background information,
17
66508
1880
๋ฐฐ๊ฒฝ ์ •๋ณด๋ฅผ ๋ฐ”ํƒ•์œผ๋กœ ํ•œ๋‹ค๋ฉด,
01:08
the second statement is only true if she does both of these things.
18
68388
4685
๋‘ ๋ฒˆ์งธ ์ง„์ˆ ์€ ๊ทธ๋…€๊ฐ€ ๋‘ ๊ฐ€์ง€๋ฅผ ๋ชจ๋‘ ๋งŒ์กฑํ•˜๋Š” ๊ฒฝ์šฐ์—๋งŒ ์‚ฌ์‹ค์ด ๋ฉ๋‹ˆ๋‹ค.
01:13
However counterintuitive it seems to imagine Lucy as an artist,
19
73073
4201
๋ฃจ์‹œ๊ฐ€ ์˜ˆ์ˆ ๊ฐ€๋ผ๋Š” ๊ฒƒ์„ ์ง๊ด€์ ์œผ๋กœ ๋ฐ›์•„๋“ค์ด๊ธฐ ํž˜๋“ค๋”๋ผ๋„,
01:17
the second scenario adds another condition on top of that, making it less likely.
20
77274
5953
๋‘ ๋ฒˆ์งธ ์ง„์ˆ ์€ ์—ฌ๊ธฐ์— ํ•œ ์กฐ๊ฑด์ด ๋”ํ•ด์กŒ๊ธฐ ๋•Œ๋ฌธ์— ์‚ฌ์‹ค์ผ ํ™•๋ฅ ์ด ๋” ๋‚ฎ์Šต๋‹ˆ๋‹ค.
01:23
For any possible set of events, the likelihood of A occurring
21
83227
4529
๋ฐœ์ƒ ๊ฐ€๋Šฅํ•œ ์‚ฌ๊ฑด์˜ ์ง‘ํ•ฉ์— ๋Œ€ํ•ด A๊ฐ€ ๋ฐœ์ƒํ•  ๊ฐ€๋Šฅ์„ฑ์€
01:27
will always be greater than the likelihood of A and B both occurring.
22
87756
5722
ํ•ญ์ƒ A์™€ B๊ฐ€ ๋™์‹œ์— ๋ฐœ์ƒํ•  ํ™•๋ฅ ๋ณด๋‹ค ํฝ๋‹ˆ๋‹ค.
01:33
If we took a random sample of a million people who majored in math,
23
93478
3890
์ˆ˜ํ•™์„ ์ „๊ณตํ•œ ๋ฐฑ๋งŒ ๋ช…์˜ ์‚ฌ๋žŒ๋“ค์„ ๋ฌด์ž‘์œ„๋กœ ๋ฝ‘์œผ๋ฉด
01:37
the subset who are portrait artists might be relatively small.
24
97368
4152
์ดˆ์ƒ์˜ˆ์ˆ ๊ฐ€์ธ ๋ถ€๋ถ„ ์ง‘ํ•ฉ์€ ์ƒ๋Œ€์ ์œผ๋กœ ์ž‘์„ ์ˆ˜ ์žˆ์ง€๋งŒ
01:41
But it will necessarily be bigger
25
101520
2210
์ดˆ์ƒ์˜ˆ์ˆ ๊ฐ€์ด๊ณ  ํฌ์ปค๋„ ํ•˜๋Š”
01:43
than the subset who are portrait artists and play poker.
26
103730
3660
๋ถ€๋ถ„ ์ง‘ํ•ฉ๋ณด๋‹ค๋Š” ๋ฐ˜๋“œ์‹œ ๋” ํด ๊ฒƒ์ž…๋‹ˆ๋‹ค.
01:47
Anyone who belongs to the second group will also belong to the firstโ€“
27
107390
3610
๋‘ ๋ฒˆ์งธ ์ง‘ํ•ฉ์— ์†ํ•œ ์ง‘๋‹จ์€ ์ฒซ ๋ฒˆ์งธ ์ง‘ํ•ฉ์—๋„ ์†ํ•˜์ง€๋งŒ
01:51
but not vice versa.
28
111000
1490
๊ทธ ๋ฐ˜๋Œ€๋Š” ์•„๋‹™๋‹ˆ๋‹ค.
01:52
The more conditions there are, the less likely an event becomes.
29
112490
5110
์กฐ๊ฑด์ด ๋งŽ์„์ˆ˜๋ก ์‚ฌ๊ฑด์ด ๋ฐœ์ƒํ•  ๊ฐ€๋Šฅ์„ฑ์ด ์ค„์–ด๋“ญ๋‹ˆ๋‹ค.
01:57
So why do statements with more conditions sometimes seem more believable?
30
117600
4619
๊ทธ๋ ‡๋‹ค๋ฉด ์™œ ๋” ๋งŽ์€ ์กฐ๊ฑด์„ ๊ฐ€์ง„ ์ง„์ˆ ์ด ๋” ๋ฏฟ์„ ๋งŒํ• ๊นŒ์š”?
02:02
This is a phenomenon known as the conjunction fallacy.
31
122219
3320
์ด๋Š” '๊ฒฐํ•ฉ ์˜ค๋ฅ˜'๋ผ๊ณ  ์•Œ๋ ค์ง„ ํ˜„์ƒ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
02:05
When weโ€™re asked to make quick decisions, we tend to look for shortcuts.
32
125539
3800
์šฐ๋ฆฌ๋Š” ๋น ๋ฅธ ๊ฒฐ์ •์„ ๋‚ด๋ ค์•ผ ํ•  ๋•Œ ์ง€๋ฆ„๊ธธ์„ ์ฐพ๋Š” ๊ฒฝํ–ฅ์ด ์žˆ์Šต๋‹ˆ๋‹ค.
02:09
In this case, we look for what seems plausible
33
129339
3120
์ด๋Ÿฐ ๊ฒฝ์šฐ, ํ†ต๊ณ„์ ์œผ๋กœ ๊ฐ€์žฅ ๊ฐ€๋Šฅ์„ฑ์ด ํฐ ๊ฒƒ๋ณด๋‹ค๋Š”
02:12
rather than what is statistically most probable.
34
132459
3010
๊ทธ๋Ÿด๋“ฏํ•ด ๋ณด์ด๋Š” ์ง„์ˆ ์„ ์ฐพ๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
02:15
On its own, Lucy being an artist doesnโ€™t match the expectations
35
135469
4431
๋ฃจ์‹œ๊ฐ€ ์˜ˆ์ˆ ๊ฐ€๋ž€ ์‚ฌ์‹ค์€ ๊ทธ๋…€๊ฐ€ ์ˆ˜ํ•™์„ ์ „๊ณตํ–ˆ๋‹ค๋Š” ์‚ฌ์‹ค์—์„œ ์ƒ๊ธด
02:19
formed by the preceding information.
36
139900
2170
๊ธฐ๋Œ€์™€ ์ผ์น˜ํ•˜์ง€ ์•Š์Šต๋‹ˆ๋‹ค.
02:22
The additional detail about her playing poker
37
142070
2740
ํฌ์ปค ๊ฒŒ์ž„์— ๋Œ€ํ•œ ์ถ”๊ฐ€ ์ •๋ณด๋Š”
02:24
gives us a narrative that resonates with our intuitionsโ€”
38
144810
3360
์šฐ๋ฆฌ์˜ ์ง๊ด€๊ณผ ์ผ์น˜ํ•˜๋Š” ์ด์•ผ๊ธฐ๋ฅผ ์ œ๊ณตํ•ฉ๋‹ˆ๋‹ค.
02:28
it makes it seem more plausible.
39
148170
2170
์ด ์„ ํƒ์ง€๊ฐ€ ๋”์šฑ ๊ทธ๋Ÿด๋“ฏํ•˜๊ฒŒ ๋“ค๋ฆฌ๋Š” ๊ฒƒ์ด์ฃ .
02:30
And we choose the option that seems more representative of the overall picture,
40
150340
4341
์šฐ๋ฆฌ๋Š” ์‹ค์ œ ํ™•๋ฅ ๊ณผ ๊ด€๊ณ„์—†์ด ์ „์ฒด์ ์ธ ์ƒํ™ฉ์„ ์ž˜ ๋Œ€ํ‘œํ•˜๋Š”
02:34
regardless of its actual probability.
41
154681
2950
์„ ํƒ์ง€๋ฅผ ํƒํ•˜๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
02:37
This effect has been observed across multiple studies,
42
157631
3590
์ด ํšจ๊ณผ๋Š” ์—ฌ๋Ÿฌ ์—ฐ๊ตฌ์—์„œ ๊ด€์ฐฐ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.
02:41
including ones with participants who understood statistics wellโ€“
43
161221
3680
ํ†ต๊ณ„๋ฅผ ์ž˜ ์ดํ•ดํ•˜๋Š” ์ฐธ๊ฐ€์ž๋“ค๋กœ ๊ตฌ์„ฑ๋œ ์—ฐ๊ตฌ๋„ ๋งˆ์ฐฌ๊ฐ€์ง€์˜€์Šต๋‹ˆ๋‹ค.
02:44
from students betting on sequences of dice rolls,
44
164901
2714
์ฃผ์‚ฌ์œ„ ๊ฒฐ๊ณผ์˜ ์ˆœ์„œ์— ๋‚ด๊ธฐ๋ฅผ ํ•œ ํ•™์ƒ๋“ค๋ถ€ํ„ฐ
02:47
to foreign policy experts predicting the likelihood of a diplomatic crisis.
45
167615
5603
์™ธ๊ต ์œ„๊ธฐ์˜ ๊ฐ€๋Šฅ์„ฑ์„ ์˜ˆ์ธกํ•˜๋Š” ๋Œ€์™ธ ์ •์ฑ… ์ „๋ฌธ๊ฐ€๊นŒ์ง€ ๋ง์ด์ฃ .
02:53
The conjunction fallacy isnโ€™t just a problem in hypothetical situations.
46
173218
4338
๊ฒฐํ•ฉ ์˜ค๋ฅ˜๋Š” ๊ฐ€์ƒ์˜ ์ƒํ™ฉ์—์„œ๋งŒ ๋ฌธ์ œ๊ฐ€ ๋˜๋Š” ๊ฒƒ์€ ์•„๋‹™๋‹ˆ๋‹ค.
02:57
Conspiracy theories and false news stories
47
177556
3130
์Œ๋ชจ๋ก ๊ณผ ๊ฑฐ์ง“ ๋‰ด์Šค๋Š”
03:00
often rely on a version of the conjunction fallacy to seem credibleโ€“
48
180686
4588
์‹ ๋ขฐํ•  ์ˆ˜ ์žˆ์–ด ๋ณด์ด๊ธฐ ์œ„ํ•ด ์ข…์ข… ๊ฒฐํ•ฉ ์˜ค๋ฅ˜์— ์˜์กดํ•ฉ๋‹ˆ๋‹ค.
03:05
the more resonant details are added to an outlandish story,
49
185274
3630
๊ณต๊ฐ ๊ฐ€๋Š” ๋””ํ…Œ์ผ๋“ค์ด ์ž๊ทน์ ์ธ ์ด์•ผ๊ธฐ์— ์ถ”๊ฐ€๋ ์ˆ˜๋ก
03:08
the more plausible it begins to seem.
50
188904
2850
๋” ๊ทธ๋Ÿด๋“ฏํ•ด ๋ณด์ด๊ธฐ ์‹œ์ž‘ํ•ฉ๋‹ˆ๋‹ค.
03:11
But ultimately, the likelihood a story is true
51
191754
2890
๊ทธ๋Ÿฌ๋‚˜ ๊ถ๊ทน์ ์œผ๋กœ ์–ด๋–ค ์ด์•ผ๊ธฐ๊ฐ€ ์‚ฌ์‹ค์ผ ํ™•๋ฅ ์€
03:14
can never be greater than the probability that its least likely component is true.
52
194644
5150
๊ฐ€์žฅ ๊ฐ€๋Šฅ์„ฑ์ด ํฌ๋ฐ•ํ•œ ์š”์†Œ๊ฐ€ ์‚ฌ์‹ค์ผ ํ™•๋ฅ ๋ณด๋‹ค ๊ฒฐ์ฝ” ํด ์ˆ˜ ์—†์Šต๋‹ˆ๋‹ค.
์ด ์›น์‚ฌ์ดํŠธ ์ •๋ณด

์ด ์‚ฌ์ดํŠธ๋Š” ์˜์–ด ํ•™์Šต์— ์œ ์šฉํ•œ YouTube ๋™์˜์ƒ์„ ์†Œ๊ฐœํ•ฉ๋‹ˆ๋‹ค. ์ „ ์„ธ๊ณ„ ์ตœ๊ณ ์˜ ์„ ์ƒ๋‹˜๋“ค์ด ๊ฐ€๋ฅด์น˜๋Š” ์˜์–ด ์ˆ˜์—…์„ ๋ณด๊ฒŒ ๋  ๊ฒƒ์ž…๋‹ˆ๋‹ค. ๊ฐ ๋™์˜์ƒ ํŽ˜์ด์ง€์— ํ‘œ์‹œ๋˜๋Š” ์˜์–ด ์ž๋ง‰์„ ๋”๋ธ” ํด๋ฆญํ•˜๋ฉด ๊ทธ๊ณณ์—์„œ ๋™์˜์ƒ์ด ์žฌ์ƒ๋ฉ๋‹ˆ๋‹ค. ๋น„๋””์˜ค ์žฌ์ƒ์— ๋งž์ถฐ ์ž๋ง‰์ด ์Šคํฌ๋กค๋ฉ๋‹ˆ๋‹ค. ์˜๊ฒฌ์ด๋‚˜ ์š”์ฒญ์ด ์žˆ๋Š” ๊ฒฝ์šฐ ์ด ๋ฌธ์˜ ์–‘์‹์„ ์‚ฌ์šฉํ•˜์—ฌ ๋ฌธ์˜ํ•˜์‹ญ์‹œ์˜ค.

https://forms.gle/WvT1wiN1qDtmnspy7