1
00:00:06,180 --> 00:00:09,440
Speaker 1: Hey, welcome to Stuff to Blow Your Mind. Today is Saturday.

2
00:00:09,480 --> 00:00:11,200
Speaker 1: We have a vault episode for you. This is going

3
00:00:11,220 --> 00:00:13,600
Speaker 1: to be The Devil You Know Part 2, and it

4
00:00:13,640 --> 00:00:18,780
Speaker 1: originally published 8-28-2025. Let's dig into it. Welcome to Stuff

5
00:00:18,800 --> 00:00:31,460
Speaker 1: to Blow Your Mind, a production of iHeartRadio. Hey, welcome

6
00:00:31,520 --> 00:00:33,640
Speaker 1: to Step to Blow Your Mind. My name is Robert Lamb.

7
00:00:34,180 --> 00:00:36,700
Speaker 2: And I am Joe McCormick. And we're back with part

8
00:00:36,820 --> 00:00:40,280
Speaker 2: two in our series called The Devil You Know, which

9
00:00:40,460 --> 00:00:45,470
Speaker 2: is about ambiguity aversion. That's the observation that we prefer

10
00:00:45,800 --> 00:00:51,250
Speaker 2: risks with known probabilities over risks with unknown probabilities. Even

11
00:00:51,290 --> 00:00:54,170
Speaker 2: if you have no reason for thinking that the unknown

12
00:00:54,270 --> 00:00:57,960
Speaker 2: risks will be worse. This preference is captured by the

13
00:00:58,020 --> 00:01:00,700
Speaker 2: folk saying, better the devil you know than the devil

14
00:01:00,760 --> 00:01:04,660
Speaker 2: you don't. For the most part, people don't like taking

15
00:01:04,760 --> 00:01:07,380
Speaker 2: bets when they don't know what the odds are and

16
00:01:07,420 --> 00:01:10,759
Speaker 2: will pay up just to avoid having to deal with ambiguity.

17
00:01:11,620 --> 00:01:14,740
Speaker 2: And this is funny in a way because almost all

18
00:01:14,780 --> 00:01:17,620
Speaker 2: of the decisions that we make in our actual lives

19
00:01:17,860 --> 00:01:21,850
Speaker 2: are made with incomplete knowledge and some significant amount of

20
00:01:21,930 --> 00:01:25,130
Speaker 2: uncertainty about our likelihood of success. Very few things in

21
00:01:25,170 --> 00:01:29,020
Speaker 2: reality have clear objective odds, like a fair coin flip

22
00:01:29,080 --> 00:01:33,160
Speaker 2: or a dice roll. However, it's pretty well established that

23
00:01:33,300 --> 00:01:36,160
Speaker 2: most people most of the time do not like making

24
00:01:36,220 --> 00:01:39,340
Speaker 2: decisions under ambiguous conditions, and we will go to great

25
00:01:39,440 --> 00:01:44,580
Speaker 2: lengths to avoid taking risks with unknown probabilities. So in

26
00:01:44,600 --> 00:01:48,280
Speaker 2: the last episode, after we introduced the concept, we talked

27
00:01:48,380 --> 00:01:51,060
Speaker 2: a good bit about the original piece of writing that

28
00:01:51,160 --> 00:01:53,940
Speaker 2: made ambiguity aversion famous. This was a 1961 paper that

29
00:01:55,460 --> 00:02:00,000
Speaker 2: by the American economist, anti-war activist, and whistleblower Daniel Ellsberg.

30
00:02:00,360 --> 00:02:04,360
Speaker 2: The paper was called Risk, Ambiguity, and the Savage Axioms,

31
00:02:04,460 --> 00:02:09,090
Speaker 2: published in the Quarterly Journal of Economics. And to briefly recap,

32
00:02:09,150 --> 00:02:13,350
Speaker 2: this paper proposed the concept of ambiguity aversion using a

33
00:02:13,430 --> 00:02:17,350
Speaker 2: number of thought experiments where you could take different bets

34
00:02:17,470 --> 00:02:20,730
Speaker 2: based on which color ball you would draw by chance

35
00:02:20,850 --> 00:02:24,149
Speaker 2: out of an urn, Ellsberg's point was that people would

36
00:02:24,230 --> 00:02:27,090
Speaker 2: tend to prefer bets with clear odds of winning, say

37
00:02:27,190 --> 00:02:31,359
Speaker 2: a 1 in 3 chance, over bets with unknown odds,

38
00:02:31,460 --> 00:02:34,880
Speaker 2: where you could have anywhere between a 0% chance or

39
00:02:34,919 --> 00:02:38,890
Speaker 2: a 2 out of 3 chance. Ellsberg's intuition here has

40
00:02:38,950 --> 00:02:43,970
Speaker 2: been broadly supported by real-world experiments. There are a few exceptions,

41
00:02:44,050 --> 00:02:47,170
Speaker 2: but most of the time, most people would rather take

42
00:02:47,230 --> 00:02:50,359
Speaker 2: bets with clear odds. even in ways that end up

43
00:02:50,440 --> 00:02:55,500
Speaker 2: implying self-contradictory assumptions about the unknown odds. This was an

44
00:02:55,540 --> 00:02:59,190
Speaker 2: important discovery in the economic field known as decision theory

45
00:02:59,250 --> 00:03:03,710
Speaker 2: because it violated a framework known as Savage's Axioms, which

46
00:03:03,750 --> 00:03:07,109
Speaker 2: were widely used to model how people make decisions in

47
00:03:07,169 --> 00:03:12,750
Speaker 2: situations of uncertainty or ambiguity. And this self-contradictory behavior is

48
00:03:12,810 --> 00:03:16,700
Speaker 2: now known as the Ellsberg Paradox. And then finally, in

49
00:03:16,740 --> 00:03:19,940
Speaker 2: the last episode, after talking about the theoretical origins of

50
00:03:20,040 --> 00:03:24,730
Speaker 2: ambiguity aversion, we also talked about this phenomenon in practice

51
00:03:24,790 --> 00:03:27,530
Speaker 2: in the real world. And our main example here was

52
00:03:27,610 --> 00:03:32,070
Speaker 2: the negotiation strategy famously associated with the Nixon White House

53
00:03:32,590 --> 00:03:37,180
Speaker 2: known as madman theory, a strategy where you intentionally cultivate

54
00:03:37,640 --> 00:03:42,900
Speaker 2: a reputation for volatility and unpredictability. In essence, you make

55
00:03:42,940 --> 00:03:46,180
Speaker 2: your peers and your counterparties worry that you are a

56
00:03:46,200 --> 00:03:50,600
Speaker 2: madman who is capable of anything. And the, the idea here, we,

57
00:03:50,660 --> 00:03:52,600
Speaker 2: we ended up talking about a lot of reasons for

58
00:03:52,640 --> 00:03:55,120
Speaker 2: thinking that this is actually not a good strategy, but

59
00:03:55,140 --> 00:03:58,180
Speaker 2: the idea here is that you will exploit their natural

60
00:03:58,300 --> 00:04:01,660
Speaker 2: ambiguity aversion to get them to make concessions. They would

61
00:04:01,720 --> 00:04:04,520
Speaker 2: not make otherwise. And then we did get into some

62
00:04:04,600 --> 00:04:09,880
Speaker 2: international relations, uh, research on madman theory, uh, reasons why

63
00:04:09,940 --> 00:04:12,500
Speaker 2: it might not actually be a good strategy, especially for

64
00:04:12,540 --> 00:04:15,640
Speaker 2: achieving long-term goals. In fact, I think in the long term,

65
00:04:16,120 --> 00:04:18,859
Speaker 2: most experts seem to agree that it is harmful to

66
00:04:18,960 --> 00:04:22,870
Speaker 2: everyone involved, including the practitioner. But again, the reason it

67
00:04:22,970 --> 00:04:26,730
Speaker 2: may sometimes work for achieving short-term wins is by taking

68
00:04:26,790 --> 00:04:29,890
Speaker 2: advantage of this fear people have of ambiguity. The fact

69
00:04:29,950 --> 00:04:32,110
Speaker 2: that a lot of people would rather just take a

70
00:04:32,150 --> 00:04:36,200
Speaker 2: bad deal than keep negotiating with somebody who is highly unpredictable.

71
00:04:37,070 --> 00:04:39,469
Speaker 1: You know, I was thinking after our recording that another

72
00:04:40,270 --> 00:04:47,140
Speaker 1: often cited bit of supposed wisdom along these lines is

73
00:04:47,339 --> 00:04:51,820
Speaker 1: the idea, and generally you'll see this referenced out of context.

74
00:04:52,020 --> 00:04:55,800
Speaker 1: So someone will not be talking about actually going to prison.

75
00:04:55,940 --> 00:04:57,720
Speaker 1: They'll be talking about your first day at a new

76
00:04:57,760 --> 00:05:00,120
Speaker 1: school or something. And they'll jokingly say, well, you know

77
00:05:00,140 --> 00:05:01,440
Speaker 1: what you're supposed to do? You're supposed to go up

78
00:05:01,490 --> 00:05:03,650
Speaker 1: to the biggest person around and hit them with a

79
00:05:03,670 --> 00:05:06,490
Speaker 1: steel chair. And then everyone will be like, whoa, that

80
00:05:06,550 --> 00:05:10,650
Speaker 1: guy's crazy. We better not mess with him. It seems

81
00:05:10,690 --> 00:05:15,229
Speaker 1: to follow similar logic and also seem equally, if not more,

82
00:05:15,310 --> 00:05:18,130
Speaker 1: flawed in its approach. It does not sound like a

83
00:05:18,170 --> 00:05:22,390
Speaker 1: tactic that would actually generate long-term benefits.

84
00:05:22,970 --> 00:05:25,870
Speaker 2: It's one of those pieces of advice that is probably...

85
00:05:26,040 --> 00:05:29,500
Speaker 2: more often repeated because it's like memorable and funny and

86
00:05:29,540 --> 00:05:32,060
Speaker 2: makes for a good story than because that many people

87
00:05:32,100 --> 00:05:34,520
Speaker 2: would think it's actually good advice, especially if you have

88
00:05:34,580 --> 00:05:36,920
Speaker 2: relevant experience. I don't know. I mean, I don't know

89
00:05:36,960 --> 00:05:40,100
Speaker 2: a lot about like prison strategy, but that does seem

90
00:05:40,120 --> 00:05:43,140
Speaker 2: like you could have some real downsides. I, you know,

91
00:05:43,180 --> 00:05:46,330
Speaker 2: it makes me think of the, you know, the, the, the, the,

92
00:05:46,410 --> 00:05:49,290
Speaker 2: another bit of wisdom, another nugget that's often dished out

93
00:05:49,330 --> 00:05:54,110
Speaker 2: in that is, getting into the philosophy of Nietzsche, you know,

94
00:05:54,150 --> 00:05:56,060
Speaker 2: that which doesn't kill us makes us stronger.

95
00:05:56,560 --> 00:05:58,420
Speaker 1: But I forget who it was who did a spin

96
00:05:58,480 --> 00:06:00,400
Speaker 1: on this and said, well, that which doesn't kill us

97
00:06:00,760 --> 00:06:05,000
Speaker 1: nearly kills us. And that should be a reason to

98
00:06:05,040 --> 00:06:05,659
Speaker 1: give us pause.

99
00:06:06,100 --> 00:06:08,800
Speaker 2: That's funny because that Nietzsche saying is also one. I

100
00:06:08,860 --> 00:06:10,260
Speaker 2: know this has come up on the show before. I

101
00:06:10,279 --> 00:06:13,500
Speaker 2: know like in a huge number of cases, that's not true. Yeah.

102
00:06:13,540 --> 00:06:17,260
Speaker 2: Like sometimes it makes us stronger. Like we do sometimes

103
00:06:17,400 --> 00:06:20,920
Speaker 2: learn and grow and gain strength from adversity, but that's

104
00:06:20,960 --> 00:06:24,000
Speaker 2: like some subset of adversities. A lot of adversities leave

105
00:06:24,040 --> 00:06:26,420
Speaker 2: us much weaker and deeply shaken.

106
00:06:26,440 --> 00:06:30,800
Speaker 1: Yeah, shaken or less trusting or just completely weakened. An

107
00:06:30,820 --> 00:06:33,479
Speaker 1: illness may weaken us. We survive it, but it's weakened

108
00:06:33,520 --> 00:06:35,460
Speaker 1: our system, and then we're more susceptible to the next

109
00:06:35,500 --> 00:06:38,120
Speaker 1: thing and so forth. So, yeah, I mean, it's great

110
00:06:38,140 --> 00:06:40,669
Speaker 1: to take solace in some of these sayings, but you

111
00:06:40,690 --> 00:06:41,990
Speaker 1: can only take them so far, really.

112
00:06:42,440 --> 00:06:44,580
Speaker 2: So anyway, we're back today to talk about some more

113
00:06:44,720 --> 00:06:48,200
Speaker 2: angles on ambiguity aversion. And one of the first things

114
00:06:48,240 --> 00:06:51,720
Speaker 2: I want to get into is that in the last episode,

115
00:06:52,779 --> 00:06:55,260
Speaker 2: we did talk a bit about the distinction that Daniel

116
00:06:55,300 --> 00:06:59,410
Speaker 2: Ellsberg tried to make between ambiguity and risk. But I

117
00:06:59,460 --> 00:07:01,710
Speaker 2: think one thing that's really important to point out that

118
00:07:01,750 --> 00:07:05,090
Speaker 2: we didn't get into last time. is the distinction between

119
00:07:05,370 --> 00:07:10,100
Speaker 2: ambiguity aversion and risk aversion. I think risk aversion is

120
00:07:10,140 --> 00:07:13,280
Speaker 2: a concept more people are probably familiar with that comes

121
00:07:13,320 --> 00:07:16,800
Speaker 2: up in conversation sometimes. These are not the same thing.

122
00:07:16,820 --> 00:07:22,270
Speaker 2: A person who is risk averse doesn't like to make bets.

123
00:07:22,590 --> 00:07:26,010
Speaker 2: They would prefer a sure thing, even if the sure

124
00:07:26,110 --> 00:07:31,440
Speaker 2: thing is statistically less valuable than the risk. So, for example,

125
00:07:31,970 --> 00:07:34,750
Speaker 2: Imagine a fair coin flip game. We stipulate that this

126
00:07:34,790 --> 00:07:38,090
Speaker 2: is a fair coin. There's no cheating involved. And on

127
00:07:38,210 --> 00:07:40,450
Speaker 2: each round, when I flip the coin, I can either

128
00:07:40,750 --> 00:07:43,670
Speaker 2: just give you $ 40 no matter how the coin lands,

129
00:07:44,370 --> 00:07:47,630
Speaker 2: or I could give you $ 100 every time it lands

130
00:07:47,690 --> 00:07:51,430
Speaker 2: heads and nothing every time it lands tails. So with

131
00:07:51,450 --> 00:07:56,970
Speaker 2: the gambling condition here, you will on average win $ 50

132
00:07:55,810 --> 00:07:58,460
Speaker 2: per round. It's a 50-50 chance of winning, and the

133
00:07:58,500 --> 00:08:02,530
Speaker 2: payout is $ 100. So the odds here are clear. If

134
00:08:02,590 --> 00:08:04,810
Speaker 2: you play this game a thousand times in a row,

135
00:08:05,210 --> 00:08:07,770
Speaker 2: you will make more money by gambling than by taking

136
00:08:07,810 --> 00:08:12,100
Speaker 2: the guaranteed $ 40 every time. It's statistically all but a

137
00:08:12,120 --> 00:08:16,160
Speaker 2: sure thing. But still, some people would rather just take

138
00:08:16,220 --> 00:08:20,920
Speaker 2: the actual sure thing. They would rather just take the $ 40.

139
00:08:20,240 --> 00:08:24,540
Speaker 2: And that is risk aversion. It is a psychological tendency

140
00:08:24,620 --> 00:08:29,300
Speaker 2: to dislike and avoid taking a stake in uncertain outcomes,

141
00:08:29,420 --> 00:08:31,720
Speaker 2: even if you know the exact odds of those outcomes,

142
00:08:31,780 --> 00:08:34,660
Speaker 2: and potentially even if those odds will statistically bring you

143
00:08:34,820 --> 00:08:37,929
Speaker 2: more benefit than the sure thing. And another way of

144
00:08:37,970 --> 00:08:41,250
Speaker 2: thinking about risk aversion is that sometimes, to some people,

145
00:08:41,270 --> 00:08:45,590
Speaker 2: the feeling of security that you get from taking the

146
00:08:45,650 --> 00:08:49,670
Speaker 2: sure thing is actually more valuable than the money. It's

147
00:08:49,750 --> 00:08:52,250
Speaker 2: more valuable than the extra money that you would take

148
00:08:52,309 --> 00:08:54,910
Speaker 2: home on average if by placing a bet with a

149
00:08:54,950 --> 00:08:58,870
Speaker 2: higher expected value. And this emphasizes something that is true

150
00:08:59,309 --> 00:09:03,490
Speaker 2: in decision theory and economics. Value isn't just about getting

151
00:09:03,650 --> 00:09:07,050
Speaker 2: or keeping money. We use money because it's an easy

152
00:09:07,150 --> 00:09:11,010
Speaker 2: way to quantify value in experiments. But sometimes we would

153
00:09:11,210 --> 00:09:15,630
Speaker 2: pay money just to feel certainty and comfort. That feeling

154
00:09:15,720 --> 00:09:17,980
Speaker 2: is of higher value to us than the money is.

155
00:09:19,020 --> 00:09:21,300
Speaker 2: But then, of course, again, other other people at other

156
00:09:21,320 --> 00:09:23,719
Speaker 2: times would not pay that. Some people love the feeling

157
00:09:23,760 --> 00:09:25,840
Speaker 2: of risk. I mean, people gamble for pleasure.

158
00:09:25,860 --> 00:09:29,550
Speaker 1: Yeah, absolutely. But yeah, but other times it's like I

159
00:09:29,590 --> 00:09:31,170
Speaker 1: just don't I just don't want to take a chance

160
00:09:31,210 --> 00:09:33,250
Speaker 1: on this. I'd rather just go for what's going to

161
00:09:33,290 --> 00:09:35,090
Speaker 1: be straightforward and guaranteed.

162
00:09:35,530 --> 00:09:38,950
Speaker 2: Right. So that's risk and risk aversion. Ambiguity aversion is

163
00:09:39,070 --> 00:09:42,079
Speaker 2: different from risk aversion because it is not about a

164
00:09:42,120 --> 00:09:46,660
Speaker 2: dislike of risk, but specifically a desire to avoid ambiguous

165
00:09:46,780 --> 00:09:50,339
Speaker 2: risks with unclear odds. You might love gambling on a

166
00:09:50,360 --> 00:09:53,640
Speaker 2: 50-50 coin toss, but you might not want to play

167
00:09:53,679 --> 00:09:55,670
Speaker 2: the game where you have to guess how many black

168
00:09:55,710 --> 00:09:57,630
Speaker 2: balls or yellow balls there are and there's no way

169
00:09:57,670 --> 00:10:01,510
Speaker 2: to know. So a person could... technically be not very

170
00:10:01,620 --> 00:10:07,060
Speaker 2: risk-averse at all, but highly averse to ambiguity. Though the interaction,

171
00:10:07,440 --> 00:10:08,880
Speaker 2: the more I thought about it, the more I thought

172
00:10:08,920 --> 00:10:12,840
Speaker 2: that the interaction between these two tendencies in real life

173
00:10:13,140 --> 00:10:16,939
Speaker 2: is interesting because I think to many of us, it's

174
00:10:17,080 --> 00:10:21,050
Speaker 2: not always clear if a fear or a hesitancy we're

175
00:10:21,090 --> 00:10:26,330
Speaker 2: experiencing about making a decision is rooted in risk aversion

176
00:10:26,510 --> 00:10:29,969
Speaker 2: or ambiguity aversion. For instance, this came up because I

177
00:10:30,010 --> 00:10:33,170
Speaker 2: was trying to think of examples of risk aversion. And

178
00:10:33,390 --> 00:10:37,100
Speaker 2: I thought of the idea of what to do with savings.

179
00:10:37,160 --> 00:10:39,300
Speaker 2: Imagine you save up a little bit of extra money

180
00:10:39,320 --> 00:10:41,340
Speaker 2: at work from your paycheck, and you want to put

181
00:10:41,380 --> 00:10:44,600
Speaker 2: that toward retirement. And you have multiple options of what

182
00:10:44,640 --> 00:10:47,179
Speaker 2: to do with that money. Imagine a person's in that

183
00:10:47,240 --> 00:10:50,850
Speaker 2: scenario and they discover that they would really feel better

184
00:10:51,330 --> 00:10:54,089
Speaker 2: just keeping the money in a bank account, earning a

185
00:10:54,170 --> 00:10:58,270
Speaker 2: relatively low rate of interest rather than investing it in

186
00:10:58,350 --> 00:11:01,890
Speaker 2: like a stock market index fund where with the index fund,

187
00:11:02,080 --> 00:11:05,620
Speaker 2: it is commonly assumed it is standard financial wisdom that

188
00:11:05,640 --> 00:11:08,059
Speaker 2: the balance will grow at a faster rate than it

189
00:11:08,080 --> 00:11:10,340
Speaker 2: will just sitting in a savings account. And you will

190
00:11:10,420 --> 00:11:12,660
Speaker 2: end up with more money years down the road if

191
00:11:12,700 --> 00:11:15,459
Speaker 2: you put it in the index fund. This pattern of

192
00:11:15,500 --> 00:11:19,060
Speaker 2: behavior could be taken as an example of risk aversion,

193
00:11:19,160 --> 00:11:22,780
Speaker 2: because it's like, I would choose to probably end up

194
00:11:22,830 --> 00:11:25,750
Speaker 2: with less money just so I can avoid taking risks

195
00:11:25,870 --> 00:11:29,550
Speaker 2: with what I have. But then I thought, well, no,

196
00:11:29,650 --> 00:11:33,750
Speaker 2: not necessarily. What if that same impulse is actually a

197
00:11:33,929 --> 00:11:39,150
Speaker 2: better example of ambiguity aversion than risk aversion? Because how

198
00:11:39,200 --> 00:11:42,400
Speaker 2: do we actually know that a stock market index fund

199
00:11:42,880 --> 00:11:45,400
Speaker 2: will grow more over time than the balance of a

200
00:11:45,440 --> 00:11:50,420
Speaker 2: savings account? The answer is historical patterns. It's like statistically

201
00:11:50,480 --> 00:11:53,020
Speaker 2: true over the last hundred years or so, you know,

202
00:11:53,179 --> 00:11:56,750
Speaker 2: within certain economies, in living memory. But it's not actually

203
00:11:56,809 --> 00:12:00,869
Speaker 2: like a fixed odds casino game or a lab experiment

204
00:12:00,910 --> 00:12:04,410
Speaker 2: where the odds of winning are fair axiomatically. In the

205
00:12:04,450 --> 00:12:08,260
Speaker 2: real world, as the fine print always says, investing involves risk,

206
00:12:08,360 --> 00:12:12,500
Speaker 2: including loss of principal. and past performance does not guarantee

207
00:12:12,580 --> 00:12:17,350
Speaker 2: future returns. So maybe actually preferring the savings account is

208
00:12:17,410 --> 00:12:21,730
Speaker 2: a form of ambiguity aversion rather than risk aversion, responding

209
00:12:21,770 --> 00:12:25,170
Speaker 2: to the idea that, well, what are the odds that

210
00:12:25,350 --> 00:12:29,120
Speaker 2: suddenly somehow investing in the stock market becomes a bad

211
00:12:29,240 --> 00:12:32,280
Speaker 2: idea in a way that it has not within living memory?

212
00:12:32,860 --> 00:12:35,660
Speaker 2: It's hard to understand what the odds of something like

213
00:12:35,700 --> 00:12:40,570
Speaker 2: that are. It seems pretty unlikely, but it's fundamentally ambiguous.

214
00:12:40,650 --> 00:12:42,949
Speaker 2: You just can't say what the objective odds are. There's

215
00:12:42,970 --> 00:12:46,170
Speaker 2: really no way to calculate them. So sometimes there's actually

216
00:12:46,530 --> 00:12:51,469
Speaker 2: ambiguity about whether there is ambiguity, you know, because like

217
00:12:51,490 --> 00:12:54,230
Speaker 2: I think an invest an investment advisor might tell you

218
00:12:54,290 --> 00:12:57,489
Speaker 2: on one hand, like, well, you know, historical performance of

219
00:12:57,530 --> 00:12:59,530
Speaker 2: the stock market, you can pretty much depend on that.

220
00:12:59,570 --> 00:13:01,989
Speaker 2: That's a known risk when, you know, you have you

221
00:13:02,010 --> 00:13:05,840
Speaker 2: have statistics that tell you that you'll most likely get

222
00:13:05,880 --> 00:13:08,300
Speaker 2: this rate of return on average. And you just don't

223
00:13:08,360 --> 00:13:11,580
Speaker 2: know how seriously to take the possibility that the future

224
00:13:11,620 --> 00:13:12,940
Speaker 2: will not be like the past.

225
00:13:13,820 --> 00:13:16,810
Speaker 1: Yeah, absolutely. And another angle that I was looking at

226
00:13:16,850 --> 00:13:18,809
Speaker 1: here on all of this is that it's not just

227
00:13:18,850 --> 00:13:23,190
Speaker 1: the ambiguity and risk, but it's perceptions of ambiguity and risk. So,

228
00:13:23,710 --> 00:13:26,670
Speaker 1: you know, it ultimately isn't necessarily coming down as much

229
00:13:26,690 --> 00:13:30,619
Speaker 1: to like, what are actually the hard probabilities here? It

230
00:13:30,660 --> 00:13:34,240
Speaker 1: can skew greatly depending on what a particular individual's worldview is.

231
00:13:34,300 --> 00:13:36,880
Speaker 1: So you could have somebody like, You could have like

232
00:13:36,920 --> 00:13:42,890
Speaker 1: a Washington Bartholomew hog wallop figure. And he's decided not

233
00:13:42,929 --> 00:13:45,189
Speaker 1: only is he not going to invest in the stock market,

234
00:13:45,429 --> 00:13:46,870
Speaker 1: he's not even going to put his money in the bank.

235
00:13:47,010 --> 00:13:49,630
Speaker 1: It's better off in the tin can buried in his backyard.

236
00:13:49,770 --> 00:13:50,610
Speaker 2: Yeah, there you go.

237
00:13:50,910 --> 00:13:54,490
Speaker 1: Even though there may be very good statistics that show

238
00:13:54,850 --> 00:13:57,959
Speaker 1: that is less safe and that is a greater risk,

239
00:13:58,630 --> 00:14:02,100
Speaker 1: your worldview and perception could be, oh, no, absolutely. You

240
00:14:02,140 --> 00:14:05,750
Speaker 1: can't trust the banks at all. I can only trust myself,

241
00:14:05,890 --> 00:14:08,520
Speaker 1: and therefore it's safer in my yard with me standing

242
00:14:08,600 --> 00:14:09,420
Speaker 1: over it with a shotgun.

243
00:14:19,700 --> 00:14:23,250
Speaker 2: But anyway, so this all brings us back to this

244
00:14:23,310 --> 00:14:27,890
Speaker 2: distinction between risk aversion and ambiguity aversion, and how difficult

245
00:14:27,950 --> 00:14:29,970
Speaker 2: it can be to tell the difference between the two.

246
00:14:30,110 --> 00:14:33,190
Speaker 2: Like when we feel aversion to making a bet of

247
00:14:33,250 --> 00:14:35,810
Speaker 2: some kind, which by way of analogy, that doesn't have

248
00:14:35,850 --> 00:14:37,790
Speaker 2: to mean literally a bet with money, but just any

249
00:14:37,850 --> 00:14:42,850
Speaker 2: important decision in our lives. When we feel that fear, hesitancy,

250
00:14:43,010 --> 00:14:46,930
Speaker 2: or resistance, we often think of this as risk aversion,

251
00:14:47,500 --> 00:14:51,380
Speaker 2: but could it actually be more often ambiguity aversion, that

252
00:14:51,400 --> 00:14:55,240
Speaker 2: we're responding subconsciously to the fact that we don't know

253
00:14:55,280 --> 00:14:58,120
Speaker 2: how to calculate the odds for the decision we're about

254
00:14:58,160 --> 00:15:01,590
Speaker 2: to make, and that makes us uncomfortable. Now, one of

255
00:15:01,610 --> 00:15:04,350
Speaker 2: the big things we left open after we introduced the

256
00:15:04,390 --> 00:15:08,270
Speaker 2: concept of ambiguity aversion in the last episode is, how

257
00:15:08,530 --> 00:15:12,330
Speaker 2: is decision theory supposed to make sense of this finding,

258
00:15:12,510 --> 00:15:15,470
Speaker 2: of like the Ellsberg Paradox? And there has been a

259
00:15:15,650 --> 00:15:18,320
Speaker 2: great amount of ink spilled in answering this question in

260
00:15:18,360 --> 00:15:21,570
Speaker 2: the 50-something years since Ellsberg. I don't think I can

261
00:15:21,610 --> 00:15:24,330
Speaker 2: even attempt to cover the whole landscape of responses here.

262
00:15:24,430 --> 00:15:26,110
Speaker 2: So I want to be clear, this will not be

263
00:15:26,150 --> 00:15:30,710
Speaker 2: a comprehensive technical survey of the ambiguity aversion literature. I'm

264
00:15:30,730 --> 00:15:32,790
Speaker 2: just going to mention a couple of broad trends and

265
00:15:32,810 --> 00:15:36,990
Speaker 2: then a few ideas within each. So within the responses,

266
00:15:37,110 --> 00:15:41,710
Speaker 2: there are two main categories of thought. One assumes that

267
00:15:41,910 --> 00:15:47,369
Speaker 2: ambiguity aversion is actually a component of rational decision-making and

268
00:15:47,800 --> 00:15:52,080
Speaker 2: that it is, in a subjective sense, wise to avoid ambiguity,

269
00:15:52,480 --> 00:15:55,160
Speaker 2: even when you can't really prove that the ambiguous bets

270
00:15:55,240 --> 00:15:58,920
Speaker 2: will generally turn out bad. The other main way of

271
00:15:58,990 --> 00:16:03,590
Speaker 2: thinking is that ambiguity aversion is some type of error

272
00:16:03,850 --> 00:16:06,850
Speaker 2: or cognitive bias. And here I'm going to mention a

273
00:16:06,870 --> 00:16:08,990
Speaker 2: couple of sources. The first one I cited in our

274
00:16:09,050 --> 00:16:13,230
Speaker 2: last episode, that is a book chapter by Mark J.

275
00:16:13,270 --> 00:16:18,800
Speaker 2: Machina and Marciano Siniscalci, called Ambiguity and Ambiguity Aversion in

276
00:16:18,870 --> 00:16:24,270
Speaker 2: the Handbook of Economics of Risk and Uncertainty, published in 2014.

277
00:16:24,270 --> 00:16:25,770
Speaker 2: And then the second source I want to mention here

278
00:16:25,810 --> 00:16:30,650
Speaker 2: is by Nabeel I. Al-Najjar and Jonathan Weinstein, published in

279
00:16:30,710 --> 00:16:35,420
Speaker 2: the Journal of Economics and Philosophy, called The Ambiguity Aversion Literature,

280
00:16:35,480 --> 00:16:39,359
Speaker 2: a Critical Assessment. This one is more in that second

281
00:16:39,380 --> 00:16:43,320
Speaker 2: category I mentioned. It's an approach arguing that ambiguity aversion

282
00:16:43,380 --> 00:16:46,890
Speaker 2: is actually just a type of error, basically. So first,

283
00:16:46,950 --> 00:16:48,370
Speaker 2: I want to look at what are some of the

284
00:16:48,410 --> 00:16:52,320
Speaker 2: ways that ambiguity aversion and like the Ellsberg paradox might

285
00:16:52,380 --> 00:16:57,020
Speaker 2: actually be a form of rational, self-consistent decision making. One

286
00:16:57,060 --> 00:17:01,680
Speaker 2: theory here can be called the multiple priors model, also

287
00:17:01,720 --> 00:17:08,390
Speaker 2: known as Max-Min expected utility. Max-Min is M-A-X-M-I-N. Max-Min expected

288
00:17:08,490 --> 00:17:12,300
Speaker 2: utility or M-E-U. let's think about it like this. Imagine

289
00:17:12,359 --> 00:17:15,859
Speaker 2: you are asked to make a bet involving ambiguous odds.

290
00:17:16,020 --> 00:17:18,030
Speaker 2: Like we'll go back to the balls in the urn

291
00:17:18,070 --> 00:17:21,149
Speaker 2: example from the first episode. So am I more likely

292
00:17:21,210 --> 00:17:23,750
Speaker 2: to pull a red ball or a black ball out

293
00:17:23,790 --> 00:17:27,850
Speaker 2: of this urn? And so to remember the details there,

294
00:17:27,869 --> 00:17:30,830
Speaker 2: the three color urn experiment was that inside this urn,

295
00:17:30,869 --> 00:17:33,230
Speaker 2: you can't see inside. You don't know what, what color

296
00:17:33,250 --> 00:17:35,750
Speaker 2: you're reaching into grab, but there are exactly 30 red

297
00:17:35,770 --> 00:17:39,410
Speaker 2: balls inside. and then 60 balls that could be any

298
00:17:39,530 --> 00:17:42,920
Speaker 2: mix of yellow and black. So they could be 0

299
00:17:42,920 --> 00:17:46,500
Speaker 2: yellow and 60 black, 60 yellow and 0 black, 30

300
00:17:46,500 --> 00:17:48,899
Speaker 2: and 30, or anything in between. You have no way

301
00:17:48,940 --> 00:17:52,859
Speaker 2: of knowing the mix in advance. In these situations, Maxman

302
00:17:52,920 --> 00:17:56,940
Speaker 2: expected utility theory would say that instead of forming one

303
00:17:57,160 --> 00:18:01,149
Speaker 2: consistent belief about the number of black balls inside and

304
00:18:01,170 --> 00:18:05,190
Speaker 2: then placing bets according to that prediction, we allow ourselves

305
00:18:05,270 --> 00:18:10,420
Speaker 2: to consider multiple different probabilities, and we apply those probabilities

306
00:18:10,520 --> 00:18:16,190
Speaker 2: to different bets pessimistically. So essentially, for each new bet

307
00:18:16,230 --> 00:18:18,869
Speaker 2: that requires us to guess the number of black balls,

308
00:18:18,950 --> 00:18:22,190
Speaker 2: we ask ourselves, what would be the worst case scenario

309
00:18:22,230 --> 00:18:25,090
Speaker 2: for me here? And then we assume those worst case

310
00:18:25,250 --> 00:18:28,480
Speaker 2: odds when considering the bet. So if you're asked to

311
00:18:28,520 --> 00:18:32,379
Speaker 2: choose between black and red, There could be anywhere between

312
00:18:32,520 --> 00:18:35,500
Speaker 2: zero and 60 black balls. You just assume there are

313
00:18:35,520 --> 00:18:38,640
Speaker 2: zero and you bet on red. I tried to come

314
00:18:38,680 --> 00:18:41,600
Speaker 2: up with a real life decision sequence that might help

315
00:18:41,680 --> 00:18:45,820
Speaker 2: the contradictory behavior under MEU make sense. So here's what

316
00:18:45,840 --> 00:18:49,360
Speaker 2: I've got. Imagine this. You're making plans for the afternoon

317
00:18:49,660 --> 00:18:51,960
Speaker 2: and you don't know what the weather is going to be.

318
00:18:52,320 --> 00:18:54,419
Speaker 2: It might rain. It might not. Don't know how to

319
00:18:54,460 --> 00:18:57,760
Speaker 2: calculate that. You were thinking about maybe going out for

320
00:18:57,800 --> 00:19:01,180
Speaker 2: a walk. But then again, considering that it might rain,

321
00:19:01,340 --> 00:19:03,119
Speaker 2: you don't want to be caught out in a storm,

322
00:19:03,280 --> 00:19:05,880
Speaker 2: so you decide instead to stay home and do some

323
00:19:06,010 --> 00:19:10,050
Speaker 2: indoor activities. At the same time, you are also deciding

324
00:19:10,270 --> 00:19:12,850
Speaker 2: what to do about the flowers on your front porch.

325
00:19:12,930 --> 00:19:16,280
Speaker 2: They're drooping and they need some water. Now, it might rain,

326
00:19:16,560 --> 00:19:18,820
Speaker 2: and then they would get all the water they need

327
00:19:18,880 --> 00:19:21,960
Speaker 2: from that. But then again, it might not rain, so

328
00:19:22,000 --> 00:19:24,260
Speaker 2: you decide to make time to fill up a watering

329
00:19:24,300 --> 00:19:28,790
Speaker 2: can and water them manually. This could appear inconsistent because

330
00:19:28,810 --> 00:19:32,410
Speaker 2: you've made two different decisions, both on the bet that

331
00:19:32,450 --> 00:19:34,600
Speaker 2: it will rain in one case and on the bet

332
00:19:34,640 --> 00:19:37,940
Speaker 2: that it won't rain in the other case. It's possible

333
00:19:37,980 --> 00:19:42,919
Speaker 2: that that's just self-contradictory, or it could be perfectly self-consistent

334
00:19:43,060 --> 00:19:46,570
Speaker 2: and logical if you are considering a range of different

335
00:19:46,630 --> 00:19:49,909
Speaker 2: probabilities of rain for the afternoon and then applying the

336
00:19:50,220 --> 00:19:54,520
Speaker 2: most pessimistic one to each decision, rather than just assuming

337
00:19:54,680 --> 00:19:58,119
Speaker 2: one probability that applies to all decisions. Does that make sense?

338
00:19:58,720 --> 00:19:59,920
Speaker 1: Yeah, yeah, I think so.

339
00:20:00,640 --> 00:20:02,859
Speaker 2: And to be clear, there's no guarantee here that the

340
00:20:02,940 --> 00:20:07,520
Speaker 2: pessimistic assumptions will be correct, because again, the information environment

341
00:20:07,619 --> 00:20:11,480
Speaker 2: is truly ambiguous. You just can't know. That's what ambiguity is.

342
00:20:12,050 --> 00:20:15,970
Speaker 2: But considering multiple probabilities and acting according to the most

343
00:20:16,109 --> 00:20:20,449
Speaker 2: pessimistic one for each decision is theoretically a logical and

344
00:20:20,550 --> 00:20:24,910
Speaker 2: consistent strategy. So the authors that support this model argue

345
00:20:24,950 --> 00:20:27,930
Speaker 2: that it is actually a rational way ambiguity aversion might

346
00:20:28,030 --> 00:20:32,470
Speaker 2: produce the apparently self-contradictory results we observe. Like the apparent

347
00:20:32,510 --> 00:20:35,170
Speaker 2: contradiction comes from the fact that we're sort of coming

348
00:20:35,210 --> 00:20:38,450
Speaker 2: up with different predictions for each bet and instead of

349
00:20:38,650 --> 00:20:41,669
Speaker 2: one prediction for the actual full state of affairs.

350
00:20:42,490 --> 00:20:44,570
Speaker 1: Yeah, though, of course, making all your decisions based on

351
00:20:44,609 --> 00:20:47,430
Speaker 1: worst-case scenarios, not a great way to live your life.

352
00:20:47,830 --> 00:20:49,350
Speaker 2: Right, and I'm going to come back to that in

353
00:20:49,369 --> 00:20:51,390
Speaker 2: a minute, because that is part of the critical assessment.

354
00:20:52,070 --> 00:20:55,670
Speaker 2: Another theory that would make the Ellsberg paradox behavior rational

355
00:20:56,190 --> 00:21:00,689
Speaker 2: is what's known as Choquet expected utility, or CEU. And

356
00:21:00,710 --> 00:21:05,470
Speaker 2: the Choquet there, that's a French name. It's C-H-O-Q-U-E-T, Choquet.

357
00:21:06,170 --> 00:21:07,710
Speaker 1: Sounds delicious. Oh, yeah.

358
00:21:08,580 --> 00:21:12,120
Speaker 2: So Choquet expected utility is the idea that when we

359
00:21:12,160 --> 00:21:16,159
Speaker 2: evaluate a bet and the odds are ambiguous, we basically

360
00:21:16,500 --> 00:21:20,260
Speaker 2: we do come up with a single internal belief about probability,

361
00:21:20,320 --> 00:21:23,899
Speaker 2: but it takes a weird form. Instead of a probability,

362
00:21:24,359 --> 00:21:29,060
Speaker 2: we use what the literature calls a capacity. So most

363
00:21:29,080 --> 00:21:31,609
Speaker 2: of the time when we're trying to judge how likely

364
00:21:31,660 --> 00:21:35,699
Speaker 2: something is to happen. Together, the likelihood of that event

365
00:21:36,030 --> 00:21:39,810
Speaker 2: and the likelihood of its negation of not that event

366
00:21:40,230 --> 00:21:41,910
Speaker 2: have to sum up to 100 percent.

367
00:21:42,050 --> 00:21:42,350
Speaker 1: Right.

368
00:21:42,410 --> 00:21:45,230
Speaker 2: So like under normal probability, if there is a one

369
00:21:45,310 --> 00:21:48,109
Speaker 2: third chance a ball will be read, there is a

370
00:21:48,170 --> 00:21:51,149
Speaker 2: two thirds chance the ball will not be read. These

371
00:21:51,230 --> 00:21:55,730
Speaker 2: complementary probabilities should always sum to one or if expressed

372
00:21:55,790 --> 00:22:01,050
Speaker 2: as percentages up to 100 percent. CEU theory says that

373
00:22:01,369 --> 00:22:07,609
Speaker 2: our internal representation of likelihoods for ambiguous events is not

374
00:22:07,810 --> 00:22:11,830
Speaker 2: like this. Instead, when the odds are ambiguous, we operate

375
00:22:11,869 --> 00:22:16,070
Speaker 2: on the basis of a capacity, which mathematically incorporates our

376
00:22:16,450 --> 00:22:21,109
Speaker 2: lack of confidence in the situation by representing the chance

377
00:22:21,150 --> 00:22:25,280
Speaker 2: of an outcome and its negation together in a way

378
00:22:25,340 --> 00:22:28,859
Speaker 2: that does not have to add up to 100%. And

379
00:22:28,880 --> 00:22:32,080
Speaker 2: I apologize to the more informed people. I'm skipping over

380
00:22:32,119 --> 00:22:35,220
Speaker 2: some mathematical complexity here. But this is the simplified version

381
00:22:35,340 --> 00:22:38,120
Speaker 2: is if you take the example of the three color urn,

382
00:22:38,570 --> 00:22:43,470
Speaker 2: a person might begin with the objective information that, you know, 30,

383
00:22:43,470 --> 00:22:45,710
Speaker 2: one third of the balls are red. So, you know,

384
00:22:45,770 --> 00:22:48,030
Speaker 2: you actually have a one third chance that the ball

385
00:22:48,050 --> 00:22:50,070
Speaker 2: you draw is red. And then for the yellow and

386
00:22:50,109 --> 00:22:53,470
Speaker 2: black balls, instead of coming up with with guesses that

387
00:22:53,609 --> 00:22:56,150
Speaker 2: sum up to two thirds, which would be the actual

388
00:22:56,210 --> 00:22:59,530
Speaker 2: sort of objective probability, whatever the mix is, you pick

389
00:22:59,550 --> 00:23:03,300
Speaker 2: some smaller probabilities like you add in a maybe a

390
00:23:03,430 --> 00:23:05,920
Speaker 2: one fifth chance that the ball is yellow and a

391
00:23:06,080 --> 00:23:09,200
Speaker 2: one fifth chance that the ball is black. Obviously, these

392
00:23:09,340 --> 00:23:12,340
Speaker 2: these cannot be the real objective odds, again, with the

393
00:23:12,380 --> 00:23:15,970
Speaker 2: caveat on what are actually objective odds. But They can't

394
00:23:16,010 --> 00:23:18,690
Speaker 2: be the real they can't represent reality because they don't

395
00:23:18,810 --> 00:23:20,770
Speaker 2: add up to one and there are no other colors

396
00:23:20,810 --> 00:23:24,030
Speaker 2: in there. But again, the values we come up with

397
00:23:24,090 --> 00:23:26,629
Speaker 2: in these kinds of decision making processes don't have to

398
00:23:26,670 --> 00:23:30,429
Speaker 2: be perfect at predicting real outcomes. And in fact, they

399
00:23:30,490 --> 00:23:33,889
Speaker 2: can't be because we are missing information. That is the ambiguity.

400
00:23:34,550 --> 00:23:37,969
Speaker 2: The point is that this is another self-consistent way our

401
00:23:38,010 --> 00:23:41,879
Speaker 2: minds could work while producing the behaviors observed in these experiments,

402
00:23:41,970 --> 00:23:45,320
Speaker 2: like the missing percentage of of likely outcomes in the

403
00:23:45,340 --> 00:23:49,500
Speaker 2: calculation is missing to reflect the level of ambiguity we

404
00:23:49,520 --> 00:23:52,619
Speaker 2: feel about the bet. So that's two main models. And

405
00:23:52,640 --> 00:23:54,760
Speaker 2: there are other models for making sense of the Ellsberg

406
00:23:54,800 --> 00:23:58,640
Speaker 2: paradox behavior in these sorts of ways. But I think

407
00:23:58,770 --> 00:24:01,750
Speaker 2: these give you a general idea of how this strain

408
00:24:01,790 --> 00:24:05,010
Speaker 2: of thinking works. The other main way of thinking about

409
00:24:05,070 --> 00:24:08,650
Speaker 2: ambiguity aversion is that it is not actually rational. Like

410
00:24:08,670 --> 00:24:11,110
Speaker 2: you don't need to find a way of making rational

411
00:24:11,190 --> 00:24:13,960
Speaker 2: sense of it. Instead, it is a form of error

412
00:24:14,119 --> 00:24:17,159
Speaker 2: or cognitive bias. And this is the thrust of the

413
00:24:17,200 --> 00:24:20,980
Speaker 2: paper that I mentioned earlier from 2009 by Al-Najjar and Weinstein.

414
00:24:21,840 --> 00:24:25,560
Speaker 2: The authors here argue, in short, that ambiguity aversion is

415
00:24:25,840 --> 00:24:31,100
Speaker 2: actually a misapplied heuristic, meaning it's a mental shortcut that

416
00:24:31,580 --> 00:24:35,619
Speaker 2: might be truly useful in some scenarios. For example, in

417
00:24:35,680 --> 00:24:39,220
Speaker 2: situations where we need to be cautious to avoid being

418
00:24:39,359 --> 00:24:44,600
Speaker 2: scammed or manipulated. But this shortcut gets mistakenly applied to

419
00:24:44,660 --> 00:24:49,580
Speaker 2: scenarios where it is not helpful. And misapplying the ambiguity

420
00:24:49,640 --> 00:24:53,120
Speaker 2: version heuristic like this, they say, could cause you to, like,

421
00:24:53,290 --> 00:24:56,949
Speaker 2: make different bets about the same thing when conditions are

422
00:24:57,010 --> 00:24:59,790
Speaker 2: the same and no new information has been learned. So,

423
00:25:00,250 --> 00:25:04,410
Speaker 2: you know, like that's producing this contradictory or apparently irrational behavior.

424
00:25:05,010 --> 00:25:09,540
Speaker 2: Or it could cause you to treat non-informative information as informative.

425
00:25:11,580 --> 00:25:15,020
Speaker 2: They also argue that relying on heuristics like pessimism, coming

426
00:25:15,040 --> 00:25:18,040
Speaker 2: back to your comment earlier, Rob, again, that's always assuming

427
00:25:18,080 --> 00:25:22,340
Speaker 2: the worst case scenario for ambiguous bets. If you actually

428
00:25:22,400 --> 00:25:26,330
Speaker 2: do this, if you rely on a pessimistic heuristic in

429
00:25:26,369 --> 00:25:31,510
Speaker 2: real life, this will systematically produce incorrect predictions and bad outcomes.

430
00:25:31,570 --> 00:25:35,290
Speaker 2: It's actually not a good strategy to follow. Failure to

431
00:25:35,369 --> 00:25:38,790
Speaker 2: make bets or decisions in the presence of ambiguity also

432
00:25:38,810 --> 00:25:43,310
Speaker 2: means causes us to never gain useful information because a

433
00:25:43,369 --> 00:25:46,570
Speaker 2: lot of times we can only learn more about what

434
00:25:46,609 --> 00:25:51,109
Speaker 2: the real odds are by making decisions in ambiguous conditions.

435
00:25:51,170 --> 00:25:54,910
Speaker 2: You know, like past bets on ambiguity lead to more

436
00:25:55,010 --> 00:25:58,429
Speaker 2: informed bets with higher certainty in the future because the

437
00:25:58,470 --> 00:26:01,640
Speaker 2: previous bets and their outcomes are information-gathering tools.

438
00:26:02,520 --> 00:26:04,179
Speaker 1: Yeah, and of course we see this played out in

439
00:26:04,220 --> 00:26:06,980
Speaker 1: the simplified world of games. If you want to get

440
00:26:07,020 --> 00:26:10,240
Speaker 1: good at a particular card game or chess or anything

441
00:26:10,280 --> 00:26:14,530
Speaker 1: like that, you need to be prepared to be beat

442
00:26:14,790 --> 00:26:18,130
Speaker 1: a lot. You know, I mean, that's part of learning

443
00:26:18,230 --> 00:26:19,790
Speaker 1: what your odds are with a native in play.

444
00:26:20,109 --> 00:26:23,210
Speaker 2: That's right. So this strain of thinking says, no, ambiguity

445
00:26:23,230 --> 00:26:27,560
Speaker 2: aversion is not a rational strategy for making decisions. Instead,

446
00:26:27,730 --> 00:26:30,940
Speaker 2: it's more like an emotional bias that causes us to

447
00:26:31,220 --> 00:26:35,639
Speaker 2: act on, yes, self-contradictory beliefs and to make bad decisions

448
00:26:35,740 --> 00:26:40,200
Speaker 2: because we do not like the feeling of ambiguity. And honestly,

449
00:26:40,560 --> 00:26:43,919
Speaker 2: I'm not sure which strain of thinking is more on track.

450
00:26:43,940 --> 00:26:46,199
Speaker 2: I've been reading about these and I'm not I don't

451
00:26:46,240 --> 00:26:49,010
Speaker 2: know who's more convincing that like that it's part of

452
00:26:49,070 --> 00:26:54,710
Speaker 2: rationality overall or that it's not. The whole thing about ambiguity,

453
00:26:54,730 --> 00:26:57,530
Speaker 2: of course, is that it's ambiguous whether or not you

454
00:26:57,550 --> 00:27:00,550
Speaker 2: should bet on it. So there really is no objective answer.

455
00:27:01,109 --> 00:27:04,050
Speaker 2: But I think the I mean, what we're instead looking

456
00:27:04,130 --> 00:27:10,700
Speaker 2: for is. What is the most consistent way to explain

457
00:27:10,740 --> 00:27:15,419
Speaker 2: the subjective behavior rather than who's actually right about the

458
00:27:15,480 --> 00:27:17,790
Speaker 2: ambiguous outcomes? Because there's no way to know them in

459
00:27:17,830 --> 00:27:21,630
Speaker 2: advance by definition. I do think the critical assessment makes

460
00:27:21,690 --> 00:27:25,790
Speaker 2: a really good point about learning. You really do often

461
00:27:25,830 --> 00:27:28,950
Speaker 2: have to be bold and make decisions in the face

462
00:27:28,990 --> 00:27:32,429
Speaker 2: of ambiguity. so that there will be more information and

463
00:27:32,490 --> 00:27:35,520
Speaker 2: less ambiguity in future decisions. That seems like a highly

464
00:27:35,580 --> 00:27:36,660
Speaker 2: salient point to me.

465
00:27:37,280 --> 00:27:40,840
Speaker 1: Yeah, yeah. And would you say, based on what you

466
00:27:40,880 --> 00:27:45,639
Speaker 1: were looking at, that it sounds like ambiguity aversion is

467
00:27:45,700 --> 00:27:49,909
Speaker 1: itself adaptive in the grand scheme of things? Maybe it's

468
00:27:49,970 --> 00:27:53,730
Speaker 1: one of these situations like type 1 errors in cognition,

469
00:27:53,770 --> 00:27:58,130
Speaker 1: false positives versus type 2 errors in cognition, false negatives,

470
00:27:58,190 --> 00:28:02,070
Speaker 1: where in sort of the hypothetical tigers hiding in the

471
00:28:02,109 --> 00:28:06,129
Speaker 1: bushes scenario, this is absolutely adaptive. But then when taken

472
00:28:06,350 --> 00:28:09,310
Speaker 1: into other aspects of life, especially modern life and all

473
00:28:09,350 --> 00:28:11,810
Speaker 1: its complexity, it's not necessarily useful.

474
00:28:12,310 --> 00:28:14,050
Speaker 2: Well, I mean, I think you could think of it

475
00:28:14,090 --> 00:28:16,430
Speaker 2: that way. And that might be one way of approaching

476
00:28:16,510 --> 00:28:21,520
Speaker 2: the error theory, right? That it's a mental shortcut that

477
00:28:21,580 --> 00:28:24,200
Speaker 2: is useful in some scenarios, again, especially when you're like,

478
00:28:24,650 --> 00:28:27,970
Speaker 2: trying to avoid being tricked or trying to avoid being

479
00:28:28,010 --> 00:28:31,320
Speaker 2: taken advantage of by somebody who has superior information to you.

480
00:28:32,000 --> 00:28:35,460
Speaker 2: But then you're misapplying it to scenarios where that's not

481
00:28:35,500 --> 00:28:38,760
Speaker 2: really the case. You're, you're just like, you're using a

482
00:28:38,820 --> 00:28:41,680
Speaker 2: defensive mental shortcut in a case where you don't really

483
00:28:41,760 --> 00:28:45,490
Speaker 2: need to. And it's causing you to act with unnecessary

484
00:28:45,550 --> 00:28:48,670
Speaker 2: levels of defense, which produce irrational behavior. And we've talked

485
00:28:48,690 --> 00:28:50,410
Speaker 2: about other ways that that can be true. So I

486
00:28:50,830 --> 00:28:53,270
Speaker 2: think that fits more with the error way of thinking.

487
00:28:54,010 --> 00:28:56,550
Speaker 2: I think the other way of thinking would just say that, yeah,

488
00:28:56,590 --> 00:28:59,670
Speaker 2: I mean, this would probably assume, it doesn't really comment

489
00:28:59,730 --> 00:29:02,370
Speaker 2: on this, but it would just assume probably that it's

490
00:29:02,430 --> 00:29:08,120
Speaker 2: adaptive because it is just part of our consistent decision-making toolkit.

491
00:29:08,280 --> 00:29:12,140
Speaker 2: And it's just the same way that our brains decide

492
00:29:12,180 --> 00:29:15,219
Speaker 2: that it's better to take a bet with a 90%

493
00:29:15,220 --> 00:29:18,120
Speaker 2: chance of winning than a 10% chance of winning. It

494
00:29:18,200 --> 00:29:21,740
Speaker 2: also has this other function, which is this ambiguity aversion function.

495
00:29:21,760 --> 00:29:25,590
Speaker 1: Yeah, and like any toolkit, You want to use multiple

496
00:29:25,630 --> 00:29:29,030
Speaker 1: tools depending on the actual situation. If you have an

497
00:29:29,710 --> 00:29:33,400
Speaker 1: expansive toolkit and it's the screwdriver every time, even when

498
00:29:33,420 --> 00:29:36,219
Speaker 1: you need to hammer a nail, yeah, that's probably not

499
00:29:36,240 --> 00:29:38,660
Speaker 1: going to work out too well. But when you got

500
00:29:38,680 --> 00:29:42,440
Speaker 1: to screw a screw in. you know, you're golden. Assuming

501
00:29:42,480 --> 00:29:44,440
Speaker 1: it's the right doodad at the end and not one,

502
00:29:44,580 --> 00:29:45,340
Speaker 1: it's not the star.

503
00:29:46,100 --> 00:29:48,700
Speaker 2: Do you go through the same one-two process every time

504
00:29:48,740 --> 00:29:50,820
Speaker 2: you find a sharp screw out on the road or

505
00:29:50,860 --> 00:29:53,480
Speaker 2: on the sidewalk where like, I hit the same two

506
00:29:53,520 --> 00:29:56,320
Speaker 2: thoughts every time and be like, first of all, it's like, ooh,

507
00:29:56,340 --> 00:29:59,160
Speaker 2: I pick it up glad somebody a child didn't step

508
00:29:59,220 --> 00:30:00,960
Speaker 2: on this or a car hit this with their tire

509
00:30:01,000 --> 00:30:02,820
Speaker 2: and then the second thing is what did this come

510
00:30:02,920 --> 00:30:06,180
Speaker 2: out of was it holding something important together.

511
00:30:06,160 --> 00:30:08,220
Speaker 1: Oh i never do the second step i'm i always

512
00:30:08,260 --> 00:30:09,959
Speaker 1: just stick at the first step where i'm like oh

513
00:30:10,060 --> 00:30:11,940
Speaker 1: glad nobody stepped on that or got that in their

514
00:30:11,960 --> 00:30:15,210
Speaker 1: tire i have saved the day and then i carry on.

515
00:30:15,510 --> 00:30:17,530
Speaker 2: I sometimes if it's like near my house i like

516
00:30:17,570 --> 00:30:21,190
Speaker 2: look at my house and i'm like it's just holding

517
00:30:21,290 --> 00:30:23,340
Speaker 2: something on i don't know If.

518
00:30:23,320 --> 00:30:25,700
Speaker 1: It's in the house, definitely. And I guess if it's

519
00:30:25,740 --> 00:30:27,580
Speaker 1: in close proximity to the house. But if it's in

520
00:30:27,600 --> 00:30:31,040
Speaker 1: the road, I'm just assuming, you know, I don't know.

521
00:30:31,080 --> 00:30:32,520
Speaker 1: I just never think about it falling out of a

522
00:30:32,540 --> 00:30:37,070
Speaker 1: vehicle in a dangerous capacity. But maybe I should. Maybe

523
00:30:37,110 --> 00:30:39,620
Speaker 1: that's an additional level of worry that I should begin

524
00:30:39,680 --> 00:30:40,610
Speaker 1: to employ.

525
00:30:41,330 --> 00:30:43,490
Speaker 2: I didn't mean a vehicle specifically. I mean, it might

526
00:30:43,530 --> 00:30:45,530
Speaker 2: be or a house or anything. There are all kinds

527
00:30:45,570 --> 00:30:47,730
Speaker 2: of things that really need to remain stuck together.

528
00:30:48,130 --> 00:30:50,760
Speaker 1: Yeah, I guess I just assume it's like it's work vehicles.

529
00:30:50,800 --> 00:30:53,860
Speaker 1: They have surplus screws rolling around and they're just rolling

530
00:30:53,940 --> 00:30:54,600
Speaker 1: out onto the road.

531
00:30:55,480 --> 00:30:58,020
Speaker 2: Let's hope that's the case most of the time. But

532
00:30:58,040 --> 00:30:59,320
Speaker 2: I'm glad I could share that anxiety.

533
00:31:09,600 --> 00:31:11,900
Speaker 1: All right. A few other angles we want to touch

534
00:31:11,980 --> 00:31:13,740
Speaker 1: on here. This first one, I'm not going to go

535
00:31:13,810 --> 00:31:15,330
Speaker 1: in super deep on this one because this one is

536
00:31:15,370 --> 00:31:21,090
Speaker 1: another highly economic topic. But there's what's known as the

537
00:31:21,130 --> 00:31:24,230
Speaker 1: competence effect. So we've already touched on the basics of

538
00:31:24,250 --> 00:31:25,730
Speaker 1: this a little bit, but this is the term for

539
00:31:25,750 --> 00:31:29,930
Speaker 1: the phenomenon by which we tend to be more ambiguity-averse

540
00:31:30,070 --> 00:31:34,220
Speaker 1: in fields where we feel we lack experience or expertise.

541
00:31:34,880 --> 00:31:38,100
Speaker 1: In economics and finance, this manifests in investors who are

542
00:31:38,140 --> 00:31:41,300
Speaker 1: more willing to invest in familiar domestic stocks rather than

543
00:31:41,360 --> 00:31:45,050
Speaker 1: complex foreign markets. And if you're like me, that sentence

544
00:31:45,110 --> 00:31:47,330
Speaker 1: is like, yeah, well, I'm uncomfortable with either. I don't

545
00:31:47,370 --> 00:31:48,810
Speaker 1: know my way. I don't know one from the other.

546
00:31:48,850 --> 00:31:53,920
Speaker 1: So just mark me down as ambiguity averse to either.

547
00:31:53,940 --> 00:31:58,020
Speaker 1: Seems to be making the point, yeah. But this is

548
00:31:58,720 --> 00:32:01,480
Speaker 1: apparently a thing. I believe this was first looked at

549
00:32:01,520 --> 00:32:04,780
Speaker 1: in a paper by Chip Heath and Amos Tversky titled

550
00:32:05,100 --> 00:32:08,290
Speaker 1: Preference and Competence in Choice Under Uncertainty, published in the

551
00:32:08,330 --> 00:32:11,870
Speaker 1: Journal of Risk and Uncertainty in 1991. And this found

552
00:32:11,890 --> 00:32:14,330
Speaker 1: that people tended to stick to gut instincts in a

553
00:32:14,370 --> 00:32:18,480
Speaker 1: familiar area versus even against a rational, diversified approach that

554
00:32:18,520 --> 00:32:20,980
Speaker 1: gets into areas they're not that familiar with.

555
00:32:21,720 --> 00:32:24,320
Speaker 2: This rings true for me. I know I am more

556
00:32:24,520 --> 00:32:28,520
Speaker 2: ambiguity averse in domains of life that I don't understand

557
00:32:28,620 --> 00:32:32,400
Speaker 2: very well, which is funny because we have to make

558
00:32:32,500 --> 00:32:37,290
Speaker 2: decisions with ambiguous odds of success in both unfamiliar and

559
00:32:37,480 --> 00:32:42,520
Speaker 2: in familiar domains. I mean, ambiguity does not leave reality

560
00:32:42,580 --> 00:32:44,600
Speaker 2: just because you know what you're doing in a certain

561
00:32:44,700 --> 00:32:48,000
Speaker 2: knowledge domain or hobby area or something like that. Having experience,

562
00:32:48,680 --> 00:32:50,240
Speaker 2: you can have all the experience in the world and

563
00:32:50,310 --> 00:32:51,950
Speaker 2: still there are things where it's just like you don't

564
00:32:52,010 --> 00:32:55,950
Speaker 2: know what your odds are. So I think maybe the

565
00:32:56,010 --> 00:33:00,510
Speaker 2: explanation for this difference is that in the unfamiliar domains,

566
00:33:01,850 --> 00:33:07,660
Speaker 2: I'm more likely to worry that ambiguity is artificial and

567
00:33:07,720 --> 00:33:11,880
Speaker 2: that I am being pricked or manipulated by someone who

568
00:33:11,960 --> 00:33:17,040
Speaker 2: has superior knowledge of that domain. Whereas in familiar situations,

569
00:33:17,860 --> 00:33:21,790
Speaker 2: I understand what amount of information it is normal to have,

570
00:33:22,240 --> 00:33:25,730
Speaker 2: and thus how much ambiguity is just natural and unavoidable,

571
00:33:26,150 --> 00:33:29,410
Speaker 2: and thus the ambiguity that is there becomes easier to embrace.

572
00:33:30,550 --> 00:33:31,230
Speaker 2: Does that make sense?

573
00:33:31,570 --> 00:33:34,390
Speaker 1: Yeah, yeah. And of course, all of this ends up

574
00:33:34,430 --> 00:33:37,540
Speaker 1: coming back to the topic of Not only like what

575
00:33:37,560 --> 00:33:39,160
Speaker 1: do you know? But what do you not know and

576
00:33:39,220 --> 00:33:41,340
Speaker 1: what do you think that you know that you actually

577
00:33:41,460 --> 00:33:44,560
Speaker 1: don't know? Yeah And we've talked about this before on

578
00:33:44,600 --> 00:33:46,640
Speaker 1: the show and it always reminds me of a line

579
00:33:46,660 --> 00:33:48,920
Speaker 1: I had to look this up. I'd forgotten the source,

580
00:33:48,940 --> 00:33:51,100
Speaker 1: but it was from the movie body heat You have

581
00:33:51,140 --> 00:33:53,720
Speaker 1: a character played by Mickey Rourke. He says it and

582
00:33:53,760 --> 00:33:56,320
Speaker 1: this is a I'm gonna paraphrase the quote a little

583
00:33:56,340 --> 00:33:58,260
Speaker 1: bit He says anytime you try to commit a crime

584
00:33:58,280 --> 00:34:00,610
Speaker 1: there are about 50 ways that you could mess up

585
00:34:00,660 --> 00:34:02,750
Speaker 1: and if you can think of 25 of them you're

586
00:34:02,770 --> 00:34:06,080
Speaker 1: a genius and Now, I don't know if those numbers

587
00:34:06,120 --> 00:34:08,080
Speaker 1: are exact, but I like the spirit of the thing.

588
00:34:08,760 --> 00:34:12,160
Speaker 1: Like he's saying, you're going into a situation and you

589
00:34:12,180 --> 00:34:15,980
Speaker 1: have this great degree of overconfidence. You don't realize all

590
00:34:16,000 --> 00:34:19,640
Speaker 1: of the ways you could potentially fail. And nobody can

591
00:34:19,680 --> 00:34:23,259
Speaker 1: have perfect knowledge. This is a quote that does recognize

592
00:34:23,300 --> 00:34:25,450
Speaker 1: that there will always be a certain amount of ambiguity.

593
00:34:26,590 --> 00:34:30,930
Speaker 1: But if you have expertise, that sort of cloud of

594
00:34:31,010 --> 00:34:32,910
Speaker 1: ambiguity is going to be smaller. It's not going to

595
00:34:32,950 --> 00:34:35,580
Speaker 1: go away. completely, but it will be smaller.

596
00:34:35,900 --> 00:34:37,900
Speaker 2: Well, I think, yeah. And I think this, yeah, the

597
00:34:37,960 --> 00:34:41,100
Speaker 2: spirit of the quote is that like people often go

598
00:34:41,160 --> 00:34:43,640
Speaker 2: into a crime thinking they're betting on red, that they're

599
00:34:43,660 --> 00:34:46,620
Speaker 2: taking a known risk when in fact they're actually, they're

600
00:34:46,640 --> 00:34:49,739
Speaker 2: going into an ambiguous risk where they don't understand how

601
00:34:49,800 --> 00:34:50,839
Speaker 2: much risk there is.

602
00:34:51,900 --> 00:34:56,080
Speaker 1: So anyway, minor, minor cinematic diversion there. Now, another higher

603
00:34:56,120 --> 00:35:01,530
Speaker 1: stakes area for the individual concerning ambiguity and ambiguity aversion

604
00:35:01,550 --> 00:35:05,310
Speaker 1: is is how ambiguity often factors into the healthcare experience.

605
00:35:05,770 --> 00:35:09,410
Speaker 1: I was reading about this. So there are two broad

606
00:35:09,469 --> 00:35:13,870
Speaker 1: categories within healthcare. There's diagnostic ambiguity. This is where the

607
00:35:13,950 --> 00:35:17,380
Speaker 1: underlying disease, there's ambiguity surrounding the underlying disease and how

608
00:35:17,440 --> 00:35:21,060
Speaker 1: it works. And then there's therapeutic ambiguity, like how do

609
00:35:21,080 --> 00:35:23,740
Speaker 1: we treat it or even cure it? And I imagine

610
00:35:23,780 --> 00:35:26,839
Speaker 1: a lot of you out there have encountered examples of this.

611
00:35:27,700 --> 00:35:31,110
Speaker 1: And these different ambiguities can lead to rather opposite choices.

612
00:35:31,850 --> 00:35:34,450
Speaker 1: But due to ambiguity aversion, a patient might choose a

613
00:35:34,489 --> 00:35:37,830
Speaker 1: treatment with known but terrible side effects over one with

614
00:35:37,890 --> 00:35:41,580
Speaker 1: a chance of, say, a better outcome, but also unknown risks.

615
00:35:42,500 --> 00:35:45,140
Speaker 2: Or because of ambiguity aversion, I think very often in

616
00:35:45,180 --> 00:35:50,160
Speaker 2: health care, we'll gravitate towards someone who offers them false

617
00:35:50,260 --> 00:35:56,109
Speaker 2: certainty versus someone who is honestly communicating like unknowns about

618
00:35:56,150 --> 00:35:56,970
Speaker 2: the level of risk.

619
00:35:57,590 --> 00:36:01,210
Speaker 1: Yeah, communication is key, and this becomes important in a

620
00:36:01,230 --> 00:36:04,370
Speaker 1: number of different ways. It has an impact, of course,

621
00:36:04,469 --> 00:36:08,870
Speaker 1: on public health and highlights the importance of science communication

622
00:36:08,930 --> 00:36:11,430
Speaker 1: in health care, as well as the dangers of medical

623
00:36:11,530 --> 00:36:16,620
Speaker 1: misinformation and conspiracy thinking. For instance, in the context of vaccination,

624
00:36:17,200 --> 00:36:21,239
Speaker 1: ambiguity aversion can definitely take the form of vaccine hesitancy.

625
00:36:22,560 --> 00:36:25,239
Speaker 1: And I want to stress again here that the ambiguity

626
00:36:25,300 --> 00:36:29,240
Speaker 1: need only be perceived ambiguity. So if one has already

627
00:36:29,340 --> 00:36:31,839
Speaker 1: bought into such statement says, well, no one really knows

628
00:36:32,540 --> 00:36:34,779
Speaker 1: what these vaccines are or no one really knows how

629
00:36:34,820 --> 00:36:40,010
Speaker 1: they work. Just assuming that those statements are true, there's

630
00:36:40,030 --> 00:36:42,810
Speaker 1: some large degree of truth to them, then there may

631
00:36:42,910 --> 00:36:47,049
Speaker 1: be well enough ambiguity, subjective ambiguity in place in the

632
00:36:47,070 --> 00:36:51,250
Speaker 1: individual's mind to make them rather adverse to this perceived

633
00:36:51,330 --> 00:36:54,339
Speaker 1: ambiguity and more likely to choose options that bring with

634
00:36:54,380 --> 00:36:59,450
Speaker 1: them greater risk. but more of a communicated idea of certainty,

635
00:36:59,780 --> 00:37:04,440
Speaker 1: like dubious alternative preventative and treatment methods, or just the

636
00:37:04,540 --> 00:37:05,799
Speaker 1: idea of doing nothing at all.

637
00:37:05,960 --> 00:37:08,040
Speaker 2: The body knows how to take care of itself, that

638
00:37:08,080 --> 00:37:10,920
Speaker 2: kind of thinking. Yeah. Which sounds very certain when delivered

639
00:37:10,960 --> 00:37:13,399
Speaker 2: by a confident and charismatic speaker. Yeah.

640
00:37:13,719 --> 00:37:16,020
Speaker 1: I was looking at a paper in behavioral medicine from

641
00:37:16,100 --> 00:37:21,279
Speaker 1: earlier this year, testing psychological correlates of ambiguity aversion in

642
00:37:21,500 --> 00:37:26,850
Speaker 1: the contest of COVID-19 vaccination. And this was by Simonovic

643
00:37:27,130 --> 00:37:31,390
Speaker 1: et al. And they found four different major takeaways here

644
00:37:31,410 --> 00:37:34,250
Speaker 1: that I thought were interesting. They found that Americans who

645
00:37:34,310 --> 00:37:39,480
Speaker 1: perceived higher ambiguity about COVID-19 vaccines reported lower worry and

646
00:37:39,540 --> 00:37:43,359
Speaker 1: lower perceived severity of COVID-19, which were each associated with

647
00:37:43,420 --> 00:37:49,500
Speaker 1: lower vaccination intentions and lower information seeking about vaccines. COVID-19 vaccines.

648
00:37:49,800 --> 00:37:52,580
Speaker 2: OK, so if I'm understanding that right, Americans who had

649
00:37:53,020 --> 00:37:57,600
Speaker 2: lower confidence in the efficacy and safety of COVID vaccines

650
00:37:58,040 --> 00:38:02,180
Speaker 2: were also, for whatever reason, less worried about COVID infection.

651
00:38:03,060 --> 00:38:05,890
Speaker 1: Right. But at the same time, they were also just

652
00:38:06,180 --> 00:38:07,950
Speaker 1: they were and this is kind of a no brainer.

653
00:38:08,050 --> 00:38:08,270
Speaker 2: Right.

654
00:38:08,290 --> 00:38:10,630
Speaker 1: It would seem to follow they would be less inclined

655
00:38:10,650 --> 00:38:14,100
Speaker 1: to have any intention of being vaccinated. And then would

656
00:38:14,120 --> 00:38:16,960
Speaker 1: be less inclined to seek out any information about said

657
00:38:17,000 --> 00:38:19,319
Speaker 1: vaccinations because, you know, their mind at this point is

658
00:38:19,340 --> 00:38:20,130
Speaker 1: already kind of made up.

659
00:38:20,420 --> 00:38:22,660
Speaker 2: Well, I think that's right. But it's also interesting to

660
00:38:22,980 --> 00:38:28,280
Speaker 2: note like the counterintuitive ways, forms that having your mind

661
00:38:28,390 --> 00:38:30,630
Speaker 2: made up in this way can take, because sometimes the

662
00:38:30,670 --> 00:38:33,390
Speaker 2: way that I think that is expressed is like sometimes

663
00:38:33,450 --> 00:38:35,570
Speaker 2: it's just a statement of like, oh, yeah, nobody knows.

664
00:38:35,630 --> 00:38:38,029
Speaker 2: Nobody can know about the vaccines. And so it just

665
00:38:38,070 --> 00:38:43,340
Speaker 2: becomes like this. infinitely and unsolvably dangerous thing out there

666
00:38:43,420 --> 00:38:46,620
Speaker 2: that you can never really have confidence in. Your mind

667
00:38:46,640 --> 00:38:48,910
Speaker 2: is made up about it, but it's made up in

668
00:38:48,950 --> 00:38:51,390
Speaker 2: a way that always just sort of like holds it

669
00:38:51,469 --> 00:38:54,450
Speaker 2: as an unresolved danger. Yeah, yeah, yeah.

670
00:38:55,890 --> 00:38:59,670
Speaker 1: And I'm sure that individuals on that side of the

671
00:38:59,730 --> 00:39:02,890
Speaker 1: argument might well make the counter argument that like, okay, well,

672
00:39:02,910 --> 00:39:05,370
Speaker 1: people who trust in the vaccine, they trust blindly. They've

673
00:39:05,390 --> 00:39:06,630
Speaker 1: just made up their mind and they're not going to

674
00:39:06,690 --> 00:39:10,270
Speaker 1: listen to any of the The criticisms. And I mean,

675
00:39:10,290 --> 00:39:13,049
Speaker 1: I disagree with that. But on the other hand, you know,

676
00:39:13,070 --> 00:39:16,790
Speaker 1: I feel like speaking personally, like I can't be like

677
00:39:17,170 --> 00:39:19,110
Speaker 1: I'm not going to pretend to be like Dr. Robert

678
00:39:19,230 --> 00:39:22,190
Speaker 1: all the time where I'm making all of all of these.

679
00:39:22,250 --> 00:39:25,140
Speaker 1: I'm doing all my own research on on, you know,

680
00:39:25,320 --> 00:39:27,759
Speaker 1: like health and dental concerns. Like ultimately, I have to

681
00:39:27,800 --> 00:39:31,549
Speaker 1: trust in a professional who. knows what they're doing and

682
00:39:31,650 --> 00:39:35,370
Speaker 1: is certified in what they're doing and leave it to

683
00:39:35,410 --> 00:39:37,930
Speaker 1: them to tell me what I should do because I

684
00:39:38,070 --> 00:39:40,489
Speaker 1: don't want to have all these decisions on me. I

685
00:39:40,530 --> 00:39:43,360
Speaker 1: don't want to do my own research on everything. I've

686
00:39:43,380 --> 00:39:45,379
Speaker 1: got enough research to do in and out during the week.

687
00:39:45,940 --> 00:39:49,020
Speaker 2: Well, I mean, about medical matters, When people say do

688
00:39:49,060 --> 00:39:52,739
Speaker 2: your own research, 99 point whatever percent of the time,

689
00:39:52,780 --> 00:39:55,740
Speaker 2: somebody who says that either is not doing any research

690
00:39:55,760 --> 00:39:58,520
Speaker 2: at all. And that just means go with your gut feeling.

691
00:39:58,800 --> 00:40:01,380
Speaker 2: Or if they are doing research, it just means doing

692
00:40:01,420 --> 00:40:05,410
Speaker 2: very bad research. It means relying on sources that you

693
00:40:05,450 --> 00:40:08,050
Speaker 2: have no objective reason for thinking or giving you good

694
00:40:08,090 --> 00:40:11,390
Speaker 2: information and probably are just confirming your priors.

695
00:40:11,550 --> 00:40:15,430
Speaker 1: Yeah, like I choose to research my health topic by

696
00:40:15,530 --> 00:40:17,130
Speaker 1: picking up Chariots of the Gods.

697
00:40:17,610 --> 00:40:20,740
Speaker 2: Yeah, I mean, I would argue, of course, like doctors

698
00:40:20,780 --> 00:40:23,480
Speaker 2: and health care professionals can be wrong. That's possible and

699
00:40:23,600 --> 00:40:26,000
Speaker 2: happens all the time. But I think on average, you're

700
00:40:26,020 --> 00:40:28,060
Speaker 2: going to do a lot better just like listening to

701
00:40:28,100 --> 00:40:31,000
Speaker 2: doctors than than like reading a Facebook post and then

702
00:40:31,040 --> 00:40:33,739
Speaker 2: making up your own mind about medicine. Yeah.

703
00:40:34,380 --> 00:40:36,340
Speaker 1: So this particular paper, I'm going to breeze through the

704
00:40:36,380 --> 00:40:39,040
Speaker 1: other main bullet point. Basically, they came down to people

705
00:40:39,080 --> 00:40:43,279
Speaker 1: who perceived higher ambiguity about COVID-19 vaccines. They reported higher

706
00:40:43,420 --> 00:40:47,939
Speaker 1: anger about the COVID-19 vaccines. which was associated with lower

707
00:40:47,980 --> 00:40:52,600
Speaker 1: perceived severity of COVID-19. They reported lower happiness about the vaccines,

708
00:40:52,640 --> 00:40:55,410
Speaker 1: which was associated with both lower worry and lower perceived

709
00:40:55,570 --> 00:40:59,590
Speaker 1: severity of COVID-19. And they also found that both Americans

710
00:40:59,630 --> 00:41:01,790
Speaker 1: and Israelis who were looked at in the study who

711
00:41:01,810 --> 00:41:05,170
Speaker 1: received higher ambiguity about COVID-19 vaccines reported lower feelings of

712
00:41:05,230 --> 00:41:09,010
Speaker 1: relaxation from the COVID-19 vaccine, which was associated with lower

713
00:41:09,150 --> 00:41:11,570
Speaker 1: perceived severity of the illness itself.

714
00:41:12,050 --> 00:41:15,390
Speaker 2: So you have these constellations of different tendencies acting together.

715
00:41:15,410 --> 00:41:20,069
Speaker 2: You've got like a lower confidence or lower trust in a, uh,

716
00:41:20,600 --> 00:41:23,799
Speaker 2: in a healthcare treatment that, uh, also seems to go

717
00:41:23,880 --> 00:41:28,320
Speaker 2: along with having, uh, uh, lower concern or, or, uh,

718
00:41:28,420 --> 00:41:31,780
Speaker 2: lower level of seriousness about the condition that that treatment

719
00:41:31,860 --> 00:41:36,009
Speaker 2: is supposed to treat less emotional positivity about the idea

720
00:41:36,050 --> 00:41:39,010
Speaker 2: of taking the treatment and, uh, and all of this together.

721
00:41:39,130 --> 00:41:42,150
Speaker 2: So it, I mean, well, one thing this sort of

722
00:41:42,190 --> 00:41:45,770
Speaker 2: suggests to me is that, uh, Of course, this is obvious,

723
00:41:45,850 --> 00:41:50,310
Speaker 2: but it just reminds us that when we are making decisions,

724
00:41:50,930 --> 00:41:54,200
Speaker 2: we do to some extent try to make rational decisions

725
00:41:54,260 --> 00:41:56,260
Speaker 2: based on our beliefs. But we're also just like highly

726
00:41:56,320 --> 00:42:00,739
Speaker 2: emotional creatures and our decision making is deeply entangled with

727
00:42:01,840 --> 00:42:05,430
Speaker 2: feelings we have about the things we're making decisions about.

728
00:42:05,510 --> 00:42:07,770
Speaker 2: You know, it's not really possible for us to approach

729
00:42:07,870 --> 00:42:12,290
Speaker 2: every decision in life. as just purely odds and benefit

730
00:42:12,350 --> 00:42:16,290
Speaker 2: maximizing machines. We have feelings about things, and those feelings,

731
00:42:16,350 --> 00:42:18,670
Speaker 2: whether we want them to or not, do influence the

732
00:42:18,690 --> 00:42:19,670
Speaker 2: way we make decisions.

733
00:42:19,850 --> 00:42:23,290
Speaker 1: Yeah, and they will override reason. And all of this

734
00:42:23,310 --> 00:42:27,319
Speaker 1: is susceptible to this sort of thing. But yeah, another

735
00:42:27,360 --> 00:42:29,839
Speaker 1: way that I think it's interesting and dreadful to think

736
00:42:29,880 --> 00:42:34,100
Speaker 1: about all of this is that, again, there's generally some

737
00:42:34,160 --> 00:42:38,760
Speaker 1: degree of ambiguity with both medical diagnosis and medical treatment.

738
00:42:39,670 --> 00:42:42,170
Speaker 1: And I can speak from experience, and again, most of

739
00:42:42,210 --> 00:42:45,170
Speaker 1: you can relate to this as well. It can be

740
00:42:45,230 --> 00:42:47,529
Speaker 1: frustrating when you go to a doctor and instead of

741
00:42:47,610 --> 00:42:50,660
Speaker 1: a clear course of action that you might want or

742
00:42:50,700 --> 00:42:53,940
Speaker 1: a definite solution that you're seeking, you instead get a

743
00:42:53,980 --> 00:42:56,640
Speaker 1: certain amount of ambiguity about what might be wrong and

744
00:42:56,680 --> 00:42:58,900
Speaker 1: how it might be addressed. Because at the end of

745
00:42:58,920 --> 00:43:01,100
Speaker 1: the day, I think we all want the Star Trek

746
00:43:01,160 --> 00:43:05,009
Speaker 1: medical tricorder experience where someone's able to just scan us,

747
00:43:05,570 --> 00:43:08,410
Speaker 1: perfect scan, see what ails us, and then tell us

748
00:43:08,469 --> 00:43:10,890
Speaker 1: exactly what they need to do to fix this, probably

749
00:43:10,930 --> 00:43:12,700
Speaker 1: with another device that they just kind of run up

750
00:43:12,739 --> 00:43:16,300
Speaker 1: and down our body. But even with today's technology, it's

751
00:43:16,360 --> 00:43:19,940
Speaker 1: more complex than that in most cases. Meanwhile, where do

752
00:43:19,980 --> 00:43:24,040
Speaker 1: we actually encounter this lack of ambiguity in the messaging? Again,

753
00:43:24,460 --> 00:43:27,500
Speaker 1: it comes generally in different forms of alternative medicine or

754
00:43:27,540 --> 00:43:31,520
Speaker 1: in snake oils or conspiracy thinking, because none of these

755
00:43:31,940 --> 00:43:35,220
Speaker 1: voices ever says, look, there's absolutely no evidence this works,

756
00:43:35,320 --> 00:43:37,500
Speaker 1: but give it a shot, or We can't know for

757
00:43:37,620 --> 00:43:40,940
Speaker 1: sure if lizard people run a secret government in Antarctica. But,

758
00:43:40,960 --> 00:43:45,700
Speaker 1: you know, there's always plenty of ambiguity in their evidence,

759
00:43:45,760 --> 00:43:48,569
Speaker 1: of course, if supplied at all. But you'll often find

760
00:43:48,610 --> 00:43:52,270
Speaker 1: them maneuvering around that in their messaging, which, again, is

761
00:43:52,410 --> 00:43:56,990
Speaker 1: low on ambiguity and high uncertainty. And like you mentioned earlier,

762
00:43:57,830 --> 00:44:00,690
Speaker 1: when it comes to legitimate medical messaging, it tends to

763
00:44:00,730 --> 00:44:04,509
Speaker 1: communicate both probability and risk. You'll get that list of

764
00:44:04,590 --> 00:44:07,910
Speaker 1: fast talk side effects at the end of of your

765
00:44:08,090 --> 00:44:12,150
Speaker 1: advertisement for an actual certified medication. And again, this is

766
00:44:12,190 --> 00:44:26,010
Speaker 1: just the inherent uncertainty in even very modern medicine. Another

767
00:44:26,030 --> 00:44:30,170
Speaker 1: paper I looked at was 2022's Vaccine Hesitancy and Cognitive

768
00:44:30,210 --> 00:44:33,660
Speaker 1: Biases by Casagliani et al. And this points out that

769
00:44:33,760 --> 00:44:39,220
Speaker 1: vaccination education strategies long followed a fact-based approach. But given

770
00:44:39,739 --> 00:44:43,360
Speaker 1: our aversion to ambiguity, along with bias and other heuristics,

771
00:44:44,100 --> 00:44:47,720
Speaker 1: these various cognitive biases in messaging have to be factored

772
00:44:47,840 --> 00:44:50,340
Speaker 1: into the outreach. And of course, this gets into the

773
00:44:50,390 --> 00:44:53,810
Speaker 1: complex way in which public health and vaccine messaging have

774
00:44:53,830 --> 00:44:57,950
Speaker 1: to attempt to counteract all the noise, sometimes by leaning

775
00:44:57,989 --> 00:45:02,150
Speaker 1: more on empathetic and audience-based approaches rather than fact-based debates.

776
00:45:02,650 --> 00:45:06,589
Speaker 1: Because again, in many cases, the actual medical facts have

777
00:45:06,650 --> 00:45:12,549
Speaker 1: baked in ambiguity while the non-scientific counterclaims do not. Again,

778
00:45:12,570 --> 00:45:16,910
Speaker 1: this is absolutely frustrating and increasingly dangerous, an increasingly dangerous

779
00:45:16,930 --> 00:45:20,629
Speaker 1: part of our reality. But yeah, the voice saying that

780
00:45:21,170 --> 00:45:23,890
Speaker 1: the vaccine has microchips in it that will turn you

781
00:45:23,930 --> 00:45:27,489
Speaker 1: into robots. There's no ambiguity to that statement. They're very clear,

782
00:45:27,510 --> 00:45:36,880
Speaker 1: they're very certain. Whereas, again, actual medical facts and actual

783
00:45:37,100 --> 00:45:41,259
Speaker 1: medicine-based arguments are going to have some level of ambiguity

784
00:45:41,760 --> 00:45:42,400
Speaker 1: in their messaging.

785
00:45:42,880 --> 00:45:45,830
Speaker 2: This is one of the things that makes communication about

786
00:45:45,890 --> 00:45:48,910
Speaker 2: science generally, but especially about something very high stakes like

787
00:45:50,150 --> 00:45:54,320
Speaker 2: medical science, medical treatments, makes it so difficult is because

788
00:45:54,340 --> 00:45:57,930
Speaker 2: you have to You have to find this very difficult

789
00:45:57,989 --> 00:46:02,210
Speaker 2: balance of like you are compelled ethically to be forthright

790
00:46:02,290 --> 00:46:05,049
Speaker 2: and honest about what you know. But you also have

791
00:46:05,070 --> 00:46:07,460
Speaker 2: to wait to find a way to present that truth

792
00:46:07,640 --> 00:46:12,219
Speaker 2: in a way that will be rhetorically effective and hit correctly.

793
00:46:13,100 --> 00:46:16,660
Speaker 2: And and that's so like you are balancing things that

794
00:46:16,700 --> 00:46:19,960
Speaker 2: somebody who is not bound by an obligation to be

795
00:46:20,180 --> 00:46:22,920
Speaker 2: ethical and honest is not bound like they can only

796
00:46:23,000 --> 00:46:25,799
Speaker 2: focus on the rhetorical part of the appeal. Like, all

797
00:46:25,840 --> 00:46:28,640
Speaker 2: they have to worry about is what I'm saying exciting

798
00:46:28,680 --> 00:46:32,700
Speaker 2: and charismatic and convincing. They don't have the other side

799
00:46:32,719 --> 00:46:33,720
Speaker 2: of the scale to balance.

800
00:46:34,239 --> 00:46:38,120
Speaker 1: That's right. That's right. And obviously this easily spills over

801
00:46:38,260 --> 00:46:46,230
Speaker 1: into other areas, you know, pseudo-archaeology, pseudo-paleontology, other pseudosciences, alternative history, ufology,

802
00:46:46,750 --> 00:46:51,250
Speaker 1: and more. Because as we, I think, really strive to

803
00:46:51,290 --> 00:46:55,300
Speaker 1: drive home in related episodes on this show, actual history

804
00:46:55,380 --> 00:47:01,040
Speaker 1: and archaeology inherently contain ambiguity. All histories are, to some

805
00:47:01,060 --> 00:47:04,400
Speaker 1: degree or another, imperfect. Some are just more imperfect than others.

806
00:47:06,080 --> 00:47:07,600
Speaker 1: As much as we know about certain periods of the

807
00:47:07,640 --> 00:47:10,230
Speaker 1: past and certain individuals and events, we still have gaps.

808
00:47:10,310 --> 00:47:15,270
Speaker 1: We have disagreements about interpretations. We have uncertainties. And then

809
00:47:15,370 --> 00:47:21,589
Speaker 1: I'm reminded, too, of particularly Alan Moore's use of the

810
00:47:21,820 --> 00:47:26,770
Speaker 1: Cox snowflake. That's K-O-C-H. in 1989's From Hell, his graphic

811
00:47:26,810 --> 00:47:31,630
Speaker 1: novel about Jack the Ripper. So the Koch snowflake is

812
00:47:31,670 --> 00:47:35,379
Speaker 1: a fractal curve that was first described by Niels Fabian

813
00:47:35,430 --> 00:47:39,440
Speaker 1: Helge von Koch in 1904. The idea here is that

814
00:47:39,480 --> 00:47:42,640
Speaker 1: it has a finite area, but an infinitely long perimeter.

815
00:47:43,260 --> 00:47:48,620
Speaker 1: So lots of crinulations. And Moore's treatment of this is

816
00:47:48,700 --> 00:47:54,040
Speaker 1: that history has a shape with infinite perimeter and finite area.

817
00:47:54,610 --> 00:47:58,359
Speaker 1: a labyrinth of connections and repeated patterns that we never

818
00:47:58,440 --> 00:47:59,360
Speaker 1: fully comprehend.

819
00:48:00,300 --> 00:48:02,460
Speaker 2: Ah, that's interesting. So if I understand that right, it

820
00:48:03,140 --> 00:48:07,040
Speaker 2: sort of explains why, as you read more about history

821
00:48:07,080 --> 00:48:10,970
Speaker 2: and learn more about history, you can, by incorporating more

822
00:48:11,010 --> 00:48:13,590
Speaker 2: and more knowledge and holding it parallel in your mind,

823
00:48:14,030 --> 00:48:17,550
Speaker 2: you can have in some ways a stronger and realer

824
00:48:17,670 --> 00:48:23,230
Speaker 2: picture of the past, but also the increasing informational complexity

825
00:48:23,250 --> 00:48:28,060
Speaker 2: it makes everything increasingly blurry as well like it you're

826
00:48:28,140 --> 00:48:32,479
Speaker 2: always finding it becomes harder to have uh it becomes

827
00:48:32,540 --> 00:48:36,080
Speaker 2: harder to have like causal through lines and clear narratives

828
00:48:36,160 --> 00:48:38,540
Speaker 2: about why history happens because the more you know the

829
00:48:38,580 --> 00:48:42,940
Speaker 2: more other causal factors you can consider when thinking about

830
00:48:42,980 --> 00:48:46,240
Speaker 2: like why something happened in history and ultimately it gets

831
00:48:46,280 --> 00:48:48,990
Speaker 2: to the point where it's so complex that it's like

832
00:48:49,090 --> 00:48:50,770
Speaker 2: who knows why anything happened.

833
00:48:50,790 --> 00:48:53,150
Speaker 1: Right right and that was his point about like who

834
00:48:53,190 --> 00:48:56,160
Speaker 1: was jack the ripper You know, we end up with

835
00:48:56,340 --> 00:48:59,140
Speaker 1: all of these, just so many books, so many theories,

836
00:48:59,320 --> 00:49:03,120
Speaker 1: so much literature and writing and conspiracy theories and so forth.

837
00:49:03,510 --> 00:49:06,009
Speaker 1: And for the most part, it's all within the confines

838
00:49:06,050 --> 00:49:09,970
Speaker 1: of that snowflake. The actual answer is outside the snowflake.

839
00:49:10,330 --> 00:49:14,070
Speaker 1: We can never actually determine it with any degree of accuracy.

840
00:49:14,510 --> 00:49:19,590
Speaker 1: It would just be conjecture. So, you know, at the

841
00:49:19,610 --> 00:49:23,540
Speaker 1: same time, time, conspiracy thinking itself engages in something akin

842
00:49:23,640 --> 00:49:26,620
Speaker 1: to the Cox snowflake, you know, because it's often an

843
00:49:26,719 --> 00:49:29,500
Speaker 1: infinite search for hidden details, whether those details are there

844
00:49:29,560 --> 00:49:32,240
Speaker 1: or not. But still, the basic premise holds true. Actual

845
00:49:32,280 --> 00:49:36,040
Speaker 1: history is complex, while the narrative provided by conspiracy thinking

846
00:49:36,100 --> 00:49:39,540
Speaker 1: is generally simple and single track. leading to a singular

847
00:49:39,620 --> 00:49:43,660
Speaker 1: force at the center of the entire conspiracy. It will

848
00:49:45,120 --> 00:49:46,940
Speaker 1: be more akin to like, why did this happen? Well,

849
00:49:46,980 --> 00:49:52,430
Speaker 1: it's this evil cabal or this evil corporation leader and

850
00:49:52,469 --> 00:49:55,510
Speaker 1: so forth, as opposed to, well, there are multiple societal

851
00:49:55,550 --> 00:50:00,050
Speaker 1: factors and historical anomalies that play into this particular reality.

852
00:50:00,850 --> 00:50:05,810
Speaker 1: And again, more ambiguity in real life, less ambiguity or

853
00:50:05,870 --> 00:50:08,470
Speaker 1: no ambiguity at all in the conspiracy argument.

854
00:50:08,880 --> 00:50:09,060
Speaker 2: Yeah.

855
00:50:10,040 --> 00:50:11,980
Speaker 1: It comes back to, it's like stocks, you know, I

856
00:50:12,000 --> 00:50:13,600
Speaker 1: don't know anything about stocks, but I know that when

857
00:50:13,640 --> 00:50:16,520
Speaker 1: I see that little advertisement at the bottom of a

858
00:50:16,560 --> 00:50:20,060
Speaker 1: blog where, you know, and there's a picture of like

859
00:50:20,100 --> 00:50:22,800
Speaker 1: an old person smiling and it's like, this is a

860
00:50:22,920 --> 00:50:25,989
Speaker 1: surefire bet stock that so-and-so has just identified. It's like,

861
00:50:26,070 --> 00:50:30,310
Speaker 1: I'm instantly suspicious because it seems a little bit too

862
00:50:30,350 --> 00:50:32,950
Speaker 1: good to be true. It seems like there's not much

863
00:50:33,050 --> 00:50:36,910
Speaker 1: ambiguity to this particular suggestion. Would be a risky click.

864
00:50:37,960 --> 00:50:40,020
Speaker 1: All right, one more area I want to touch here

865
00:50:40,060 --> 00:50:42,980
Speaker 1: because I was looking at a really recent paper on

866
00:50:43,100 --> 00:50:47,580
Speaker 1: ambiguity aversion. Okay, here's a hypothetical scenario. You're considering a

867
00:50:47,640 --> 00:50:50,770
Speaker 1: couple of vacation options, and we'll assume for the purposes

868
00:50:50,830 --> 00:50:53,710
Speaker 1: of this experiment that the cost and other matters are

869
00:50:53,810 --> 00:50:57,150
Speaker 1: equal here. One is a trip to a family vacation

870
00:50:57,170 --> 00:51:00,469
Speaker 1: destination that you've visited numerous times before, and the other

871
00:51:00,570 --> 00:51:03,390
Speaker 1: is an entirely new experience. Maybe it's visiting a country

872
00:51:03,410 --> 00:51:07,140
Speaker 1: you've never traveled to or something of that nature. And

873
00:51:07,160 --> 00:51:09,990
Speaker 1: then as you're making up your mind, what do you do? Oh,

874
00:51:10,070 --> 00:51:14,250
Speaker 1: you flip on the news. And surprise, surprise, it's all negative.

875
00:51:14,330 --> 00:51:18,690
Speaker 1: It's all troubling. Would this exposure to negative news influence

876
00:51:18,730 --> 00:51:22,110
Speaker 1: your decision making and push you toward the less ambiguous choice?

877
00:51:22,880 --> 00:51:27,319
Speaker 1: Or would you still basically make your decision as if

878
00:51:27,360 --> 00:51:29,600
Speaker 1: you would have if you didn't watch the news at all?

879
00:51:30,300 --> 00:51:32,620
Speaker 2: Well, I guess it could go multiple ways. But yeah,

880
00:51:32,700 --> 00:51:37,120
Speaker 2: I would tend to assume that increasing my anxiety would

881
00:51:37,160 --> 00:51:40,520
Speaker 2: make me more ambiguity averse and that I would be

882
00:51:40,760 --> 00:51:46,640
Speaker 2: seeking more known odds and familiar experiences if my anxiety

883
00:51:46,700 --> 00:51:47,460
Speaker 2: levels are higher.

884
00:51:47,480 --> 00:51:51,480
Speaker 1: Yeah, yeah. You know, prior to reading this paper, I

885
00:51:51,520 --> 00:51:54,980
Speaker 1: might have jumped at this thinking as well. I've very

886
00:51:55,060 --> 00:51:57,540
Speaker 1: rarely altered travel plans due to something I saw in

887
00:51:57,580 --> 00:51:59,400
Speaker 1: the news, but I'll be the first to admit that

888
00:51:59,480 --> 00:52:01,819
Speaker 1: news items can rattle my resolve a bit. You know,

889
00:52:01,890 --> 00:52:05,020
Speaker 1: they can make me more anxious. So I would have guessed, yeah,

890
00:52:05,140 --> 00:52:07,910
Speaker 1: negative news can surely push one away from risk, away

891
00:52:07,950 --> 00:52:10,810
Speaker 1: from ambiguity, and back into the familiar, and more to

892
00:52:10,850 --> 00:52:13,870
Speaker 1: the point, might be likely to do so on the whole.

893
00:52:14,610 --> 00:52:16,330
Speaker 2: Now, one of the only things I would think that

894
00:52:16,450 --> 00:52:18,360
Speaker 2: might push me back in the other direction is a

895
00:52:18,390 --> 00:52:22,320
Speaker 2: finding we talked about in the previous episode where essentially

896
00:52:22,600 --> 00:52:29,930
Speaker 2: if the known odds are bad, people's ambiguity aversion to

897
00:52:30,010 --> 00:52:33,810
Speaker 2: take the other bet lessens, right? So like in the

898
00:52:33,850 --> 00:52:36,930
Speaker 2: three color urn experiment, if instead of 30 red balls,

899
00:52:36,969 --> 00:52:40,339
Speaker 2: there were like five red balls, people at that point

900
00:52:40,380 --> 00:52:43,640
Speaker 2: become more likely to bet on black because it's like, well,

901
00:52:43,820 --> 00:52:46,600
Speaker 2: the odds of the thing you know about are terrible.

902
00:52:47,040 --> 00:52:50,759
Speaker 2: So why not take a gamble on something else? So

903
00:52:51,300 --> 00:52:54,160
Speaker 2: I wonder maybe if that could push you back the

904
00:52:54,219 --> 00:52:58,210
Speaker 2: other way. Like my intuition is that anxiety makes you

905
00:52:58,270 --> 00:53:01,250
Speaker 2: more ambiguity averse and just want to go with something

906
00:53:01,410 --> 00:53:04,450
Speaker 2: where the odds are clear. But maybe anxiety makes you

907
00:53:04,510 --> 00:53:08,870
Speaker 2: believe that, like, essentially the odds of stasis are really bad,

908
00:53:09,370 --> 00:53:11,489
Speaker 2: and thus you might as well take a gamble on

909
00:53:11,570 --> 00:53:13,850
Speaker 2: something different. Yeah, yeah.

910
00:53:14,550 --> 00:53:19,730
Speaker 1: You know, we also have to acknowledge that news reporting

911
00:53:19,770 --> 00:53:24,879
Speaker 1: can certainly change the perceived odds of something happening. Like,

912
00:53:25,140 --> 00:53:27,780
Speaker 1: if you were to watch a news network, a hypothetical

913
00:53:27,820 --> 00:53:31,920
Speaker 1: news network that was just 24-hour coverage of snakebites, that's

914
00:53:31,980 --> 00:53:35,230
Speaker 1: all they covered. Anytime, anywhere in the world someone is

915
00:53:35,250 --> 00:53:37,290
Speaker 1: bitten by a snake, they are there to cover it

916
00:53:37,410 --> 00:53:40,860
Speaker 1: in full. This would maybe give you certainly a heightened

917
00:53:40,920 --> 00:53:44,520
Speaker 1: awareness of maybe the legitimate risks of snakebite, but also

918
00:53:44,560 --> 00:53:48,520
Speaker 1: maybe an inflated feeling about your personal risks of snakebite.

919
00:53:49,000 --> 00:53:49,340
Speaker 2: Yeah.

920
00:53:49,540 --> 00:53:49,739
Speaker 1: Okay.

921
00:53:49,780 --> 00:53:51,920
Speaker 2: So that's different in that what I was just talking

922
00:53:51,940 --> 00:53:54,800
Speaker 2: about was like just how general emotional affect could affect

923
00:53:54,860 --> 00:53:57,700
Speaker 2: the way you reason, but it could also be content specific.

924
00:53:57,739 --> 00:53:58,360
Speaker 2: You're right.

925
00:53:58,380 --> 00:54:02,620
Speaker 1: Yeah. But anyway, mostly what we're dealing here with is

926
00:54:02,660 --> 00:54:07,280
Speaker 1: the effect here. So a recent study from Adelphi University

927
00:54:07,460 --> 00:54:09,960
Speaker 1: looked into this. It was published in Frontiers in Psychology

928
00:54:10,020 --> 00:54:13,760
Speaker 1: entitled Negative News Exposure Does Not Affect Risk or Ambiguity Aversion.

929
00:54:14,520 --> 00:54:16,620
Speaker 1: So their answer is pretty much in the headline there,

930
00:54:16,680 --> 00:54:19,299
Speaker 1: as is often the case. Now, this is another one

931
00:54:19,320 --> 00:54:21,740
Speaker 1: of those small studies. I believe they worked with groups

932
00:54:21,800 --> 00:54:22,720
Speaker 1: here of like 84 and 229. So, you know, as always,

933
00:54:22,739 --> 00:54:30,049
Speaker 1: more research is required. This is not something you can

934
00:54:30,090 --> 00:54:32,589
Speaker 1: just take to the bank, but it's an interesting possibility.

935
00:54:33,290 --> 00:54:35,989
Speaker 1: The authors point out that while studies have found that

936
00:54:36,050 --> 00:54:40,430
Speaker 1: negative effect can increase risk aversion and ambiguity aversion, we

937
00:54:40,450 --> 00:54:44,430
Speaker 1: didn't know if, quote, these effects generalize to more realistic

938
00:54:44,530 --> 00:54:47,640
Speaker 1: negative stimuli, such as watching the news.

939
00:54:48,880 --> 00:54:52,160
Speaker 2: Oh, OK. So maybe in previous studies, it's been found

940
00:54:52,219 --> 00:54:57,920
Speaker 2: that negative emotions can increase ambiguity aversion and risk aversion.

941
00:54:58,510 --> 00:55:01,310
Speaker 2: And so you would just assume that stimuli like watching

942
00:55:01,350 --> 00:55:05,430
Speaker 2: the news would also increase those aversions, but maybe it doesn't.

943
00:55:06,110 --> 00:55:07,880
Speaker 1: Yeah. And you can see why this would be something

944
00:55:07,940 --> 00:55:10,460
Speaker 1: worth knowing. I mean, not only from a general standpoint

945
00:55:10,500 --> 00:55:14,960
Speaker 1: of understanding, you know, how people make choices based on

946
00:55:15,120 --> 00:55:17,780
Speaker 1: their mood and what influences their mood. That's certainly valid.

947
00:55:17,800 --> 00:55:20,469
Speaker 1: And I imagine that's the main point of the research here.

948
00:55:20,510 --> 00:55:23,779
Speaker 1: But also you can imagine it being very A very

949
00:55:23,820 --> 00:55:28,460
Speaker 1: interesting topic for individuals buying advertisement time on, say, a

950
00:55:28,480 --> 00:55:33,860
Speaker 1: 24-hour news network, especially if they want to have an

951
00:55:33,920 --> 00:55:36,360
Speaker 1: advertisement for some sort of adventure travel or something of

952
00:55:36,410 --> 00:55:40,870
Speaker 1: that nature. Bungee cords, I don't know, whatever you happen

953
00:55:40,890 --> 00:55:44,129
Speaker 1: to be selling. So in this particular study, one group

954
00:55:44,170 --> 00:55:46,870
Speaker 1: was exposed to negative news. And in this case, it

955
00:55:46,890 --> 00:55:49,219
Speaker 1: was like news about a car crash. And the other

956
00:55:49,260 --> 00:55:51,959
Speaker 1: group was exposed to neutral news. So not great news,

957
00:55:52,020 --> 00:55:54,440
Speaker 1: not good news, not like a real fluff piece, but

958
00:55:54,500 --> 00:55:58,810
Speaker 1: just something about train schedules, something nice and boring. And

959
00:55:58,830 --> 00:56:01,890
Speaker 1: maybe it's a little interesting. And they found that although

960
00:56:01,930 --> 00:56:05,810
Speaker 1: participants who watched negative news reported a significant increase in

961
00:56:05,870 --> 00:56:08,950
Speaker 1: negative effect, they did not differ from the neutral news

962
00:56:08,989 --> 00:56:11,330
Speaker 1: group in their risk or ambiguity preferences.

963
00:56:11,969 --> 00:56:15,670
Speaker 2: Interesting. OK, so they said past studies had found negative

964
00:56:15,730 --> 00:56:21,310
Speaker 2: effect increases ambiguity aversion. They showed people bad news, and

965
00:56:21,370 --> 00:56:27,320
Speaker 2: it did increase negative affect, but did not increase ambiguity aversion. Right, right.

966
00:56:27,440 --> 00:56:30,400
Speaker 1: So I don't think it means go ahead and consume

967
00:56:30,420 --> 00:56:32,060
Speaker 1: all the negative news you want. It's not going to

968
00:56:32,100 --> 00:56:35,609
Speaker 1: impact you. Like, no, it can and will impact you.

969
00:56:35,910 --> 00:56:38,650
Speaker 1: But what the studies seem to be suggesting, rather, is

970
00:56:38,730 --> 00:56:45,290
Speaker 1: that sort of incidental or realistic consumption of negative news

971
00:56:46,380 --> 00:56:48,820
Speaker 1: that exposure is not going to be enough of a

972
00:56:48,900 --> 00:56:51,319
Speaker 1: force in and of itself to move the scale and

973
00:56:51,460 --> 00:56:56,520
Speaker 1: actually impact your decision-making process. Again, on the whole, this

974
00:56:56,550 --> 00:56:59,090
Speaker 1: would be a generalization, and again, it's based on a

975
00:56:59,160 --> 00:57:01,610
Speaker 1: smaller study, and there's so many factors on top of

976
00:57:01,630 --> 00:57:04,690
Speaker 1: this that you could throw into the scenario. But both

977
00:57:04,770 --> 00:57:08,330
Speaker 1: groups were equally likely to choose a certain gamble over

978
00:57:08,370 --> 00:57:11,489
Speaker 1: an ambiguous one following the consumption of their meat.

979
00:57:12,050 --> 00:57:13,610
Speaker 2: Oh, well, some just occurred to me. I have no

980
00:57:13,650 --> 00:57:15,770
Speaker 2: reason to think this is really true, because I'm sure

981
00:57:15,810 --> 00:57:18,850
Speaker 2: the authors would have done a pretty good job of

982
00:57:19,390 --> 00:57:21,800
Speaker 2: planning this out. But what if the control group, the

983
00:57:21,840 --> 00:57:26,700
Speaker 2: train schedule news, was actually pretty demoralizing? It was supposed

984
00:57:26,740 --> 00:57:29,940
Speaker 2: to be a neutral control, but actually that made people

985
00:57:30,160 --> 00:57:31,520
Speaker 2: equally ambiguity-averse.

986
00:57:32,320 --> 00:57:34,100
Speaker 1: That would have been an interesting outcome. No, no, no.

987
00:57:34,140 --> 00:57:39,950
Speaker 1: But my understanding is that basically the normal parameters... of

988
00:57:40,050 --> 00:57:43,880
Speaker 1: ambiguity aversion were in place. So people were still averse

989
00:57:43,920 --> 00:57:48,140
Speaker 1: to ambiguity, but in equal measure, whether they had just

990
00:57:48,220 --> 00:57:50,320
Speaker 1: learned about a car crash or about the exciting world

991
00:57:50,360 --> 00:57:51,280
Speaker 1: of train schedules.

992
00:57:51,640 --> 00:57:57,410
Speaker 2: Yeah, so they basically just conformed roughly to Ellsberg's predictions. Yes, exactly.

993
00:57:57,890 --> 00:58:01,190
Speaker 1: But again, so many studies come out about this. There

994
00:58:01,330 --> 00:58:03,130
Speaker 1: are new ones all the time. It's just such a

995
00:58:03,840 --> 00:58:08,510
Speaker 1: fascinating way to sort of get in and crunch how

996
00:58:08,670 --> 00:58:12,850
Speaker 1: we make decisions in life based on how we deal

997
00:58:12,870 --> 00:58:16,189
Speaker 1: with ambiguity and the inherent ambiguity of life.

998
00:58:16,710 --> 00:58:20,910
Speaker 2: Yeah. Again, I can't emphasize this enough. Because these studies

999
00:58:20,970 --> 00:58:24,590
Speaker 2: want to quantify the effect, they need to use these

1000
00:58:24,930 --> 00:58:31,820
Speaker 2: fixed games that include certain odds and risk conditions where

1001
00:58:31,860 --> 00:58:34,810
Speaker 2: the odds are very clear. But In real life, you're

1002
00:58:34,830 --> 00:58:38,780
Speaker 2: dealing with basically top-to-bottom ambiguity. It's real ambiguity all the time. Like,

1003
00:58:38,840 --> 00:58:43,340
Speaker 2: really rarely do we encounter decisions where our probability of

1004
00:58:43,400 --> 00:58:48,620
Speaker 2: success is absolutely clear. It's clearer than others. Some decisions

1005
00:58:48,660 --> 00:58:52,060
Speaker 2: have clearer odds than others, but you never really know

1006
00:58:52,100 --> 00:58:52,860
Speaker 2: what your odds are.

1007
00:58:53,680 --> 00:58:55,370
Speaker 1: Yeah, I mean, that's one of the reasons we've talked

1008
00:58:55,390 --> 00:58:56,530
Speaker 1: about this before on the show. It's one of the

1009
00:58:56,550 --> 00:59:00,370
Speaker 1: reasons we like things like zombie apocalypse scenarios. Not because

1010
00:59:00,410 --> 00:59:03,450
Speaker 1: we necessarily like the idea of the world ending. or

1011
00:59:03,520 --> 00:59:05,260
Speaker 1: of the dead coming back to life and trying to

1012
00:59:05,300 --> 00:59:09,229
Speaker 1: eat our brains. But baked into those scenarios, there's often

1013
00:59:09,290 --> 00:59:13,830
Speaker 1: like a simple of how the world works, a removal

1014
00:59:13,910 --> 00:59:16,870
Speaker 1: of some degree of the ambiguity. And there's often still

1015
00:59:16,890 --> 00:59:19,190
Speaker 1: in any good fiction, I think you're going to have

1016
00:59:19,240 --> 00:59:23,030
Speaker 1: a contemplation of ambiguity as well. But in broad strokes,

1017
00:59:23,070 --> 00:59:26,000
Speaker 1: you might have a scenario boiled down to like, OK, well,

1018
00:59:26,040 --> 00:59:28,340
Speaker 1: now it's the living versus the dead. Now it's the

1019
00:59:28,360 --> 00:59:32,220
Speaker 1: clear good guys versus the bad guys. And what is

1020
00:59:32,260 --> 00:59:34,720
Speaker 1: the answer? Well, it's always blasting it with a shotgun, right?

1021
00:59:35,000 --> 00:59:39,280
Speaker 2: Isn't it interesting how in most stories, any major change

1022
00:59:39,400 --> 00:59:43,280
Speaker 2: that happens to the characters is because of whatever the

1023
00:59:43,320 --> 00:59:47,060
Speaker 2: struggle in the story is about? It's not like totally random,

1024
00:59:47,120 --> 00:59:50,180
Speaker 2: out of nowhere, exogenous events come in and completely change

1025
00:59:50,220 --> 00:59:54,560
Speaker 2: the story. Occasionally that happens. And people, I think, often

1026
00:59:54,620 --> 00:59:57,400
Speaker 2: find that quite interesting in storytelling because it's pretty rare.

1027
00:59:58,540 --> 01:00:02,440
Speaker 1: Where the cause of the scenario is ambiguous, you mean?

1028
01:00:02,500 --> 01:00:04,640
Speaker 2: Where like a major change maybe comes in the middle

1029
01:00:04,680 --> 01:00:06,680
Speaker 2: of the story and it has nothing to do with

1030
01:00:06,700 --> 01:00:08,880
Speaker 2: what the main struggle or plot of the story is.

1031
01:00:08,940 --> 01:00:11,680
Speaker 2: It just comes out of nowhere. Yeah. I can't think

1032
01:00:11,720 --> 01:00:13,240
Speaker 2: of an example off the top of my head, but

1033
01:00:13,570 --> 01:00:15,470
Speaker 2: I know there are some like this. Write in with

1034
01:00:15,510 --> 01:00:16,110
Speaker 2: your example.

1035
01:00:16,690 --> 01:00:18,250
Speaker 1: All right. We're going to go and close up this episode,

1036
01:00:18,270 --> 01:00:19,770
Speaker 1: but we'd love to hear from everyone out there. We

1037
01:00:19,790 --> 01:00:23,010
Speaker 1: know that everyone has experience with ambiguity in life, judging it,

1038
01:00:23,290 --> 01:00:26,669
Speaker 1: being adverse to it, rolling the dice anyway, and so forth.

1039
01:00:27,140 --> 01:00:29,240
Speaker 1: So write in. We would love to hear from you.

1040
01:00:29,740 --> 01:00:31,340
Speaker 1: Just a reminder that Stuff to Build Your Mind is

1041
01:00:31,400 --> 01:00:33,760
Speaker 1: primarily a science and culture podcast with core episodes on

1042
01:00:33,800 --> 01:00:36,700
Speaker 1: Tuesdays and Thursdays, short-form episodes on Wednesdays, and on Fridays

1043
01:00:36,720 --> 01:00:39,160
Speaker 1: we set aside most serious concerns to just talk about

1044
01:00:39,180 --> 01:00:41,100
Speaker 1: a weird film on Weird House Cinema.

1045
01:00:41,420 --> 01:00:45,180
Speaker 2: Huge thanks, as always, to our excellent audio producer, J.J. Posway.

1046
01:00:45,340 --> 01:00:46,760
Speaker 2: If you would like to get in touch with us

1047
01:00:46,800 --> 01:00:49,240
Speaker 2: with feedback on this episode or any other, to suggest

1048
01:00:49,300 --> 01:00:51,340
Speaker 2: a topic for the future, or just to say hello...

1049
01:00:51,660 --> 01:01:03,090
Speaker 2: you can email us at contact at stufftoblowyourmind.com. Stuff to

1050
01:01:03,110 --> 01:01:05,240
Speaker 2: Blow Your Mind is a production of iHeartRadio.

1051
01:01:05,620 --> 01:01:08,940
Speaker 1: For more podcasts from iHeartRadio, visit the iHeartRadio app. Apple

1052
01:01:08,960 --> 01:01:11,340
Speaker 1: Podcasts are wherever you listen to your favorite shows.

1053
01:01:20,180 --> 01:01:20,860
Speaker 2: Thank you.