Latest / Safe As Podcast / E18: When Emotion Leads Risk - risk as feelings and not just numbers

Transcript

E18: When Emotion Leads Risk - risk as feelings and not just numbers · Safe As Podcast

• 0:07
  Have you ever made a decision where the numbers screened one
• 0:10
  thing, but you got told you something else entirely?
• 0:14
  Maybe you're more terrified of flying than driving, even though
• 0:18
  statistics consistently show driving is far riskier.
• 0:22
  Perhaps you felt an inexplicable dread about something like that.
• 0:25
  On paper, poses little threat. We're constantly bombarded with
• 0:29
  definitions of risk, probabilities, data, objective
• 0:33
  assessments. We're told to be rational, to
• 0:35
  trust the numbers, to leave emotion out of it when it comes
• 0:39
  to understanding danger, what of that uneasy feeling in the
• 0:42
  stomach, That sudden surge of fear isn't just a rational
• 0:46
  noise? What if it's a primal, lightning
• 0:49
  fast intelligence honed by a millennia of human survival,
• 0:52
  telling you something critical about the world around you Good
• 0:56
  day everyone. I'm Ben Hutchinson and this is
• 0:58
  safe as a podcast dedicated to the thrifty analysis of safety,
• 1:02
  risk and performance research visit safetyinsights.org for
• 1:06
  more research. This paper today, risk as
• 1:10
  analysis and risk as feelings, some thoughts about effect,
• 1:14
  reason, risk and rationality by Paul slavik and his colleagues
• 1:18
  explores how we actually understand and react to risk
• 1:22
  published 2004 and risk analysis.
• 1:25
  It argues that emotion or effect that's a double FECT, not a
• 1:29
  double FECT effect, plays a crucial role in our decisions.
• 1:34
  The paper challenges the traditional idea that emotions
• 1:37
  make us irrational, instead suggesting that truly smart
• 1:41
  decisions in both our logical and emotional thinking working
• 1:44
  together. And this paper covers a lot of
• 1:47
  ground. But fundamentally, we confront
• 1:49
  risk in three ways. Risk as feelings.
• 1:53
  These are our fast, instinctive, intuitive reactions to danger.
• 1:57
  Risk is analysis which brings logic, reason and scientific
• 2:00
  deliberation to bear on hazard management.
• 2:03
  And when these ancient instincts conflict with our modern
• 2:06
  structured analytical frames, we become painfully aware of 1/3
• 2:10
  reality. Paul Slavic argues in another
• 2:12
  paper. Risk is politics.
• 2:15
  So some of the key findings in this paper, really there's two
• 2:19
  simple ways we can think about how people actually think and
• 2:22
  process the world. We have an analytical type of
• 2:26
  processing. This uses logic, numbers, and
• 2:29
  formal rules. It's slow, it takes effort, and
• 2:33
  we're consciously aware of it. It's driven by reason and needs
• 2:36
  evidence to justify conclusions. We also have an experiential
• 2:40
  processing mode. It's intuitive, fast, and mostly
• 2:44
  automatic. It relies on images and
• 2:47
  associations directly linked to emotions and feelings.
• 2:51
  This processing type represents risk as a feeling that tells us
• 2:55
  whether it's safe to walk down this dark street or drink this
• 2:58
  strange smelling water. It's holistic, it's guided by
• 3:02
  gut feelings from past experiences and enables
• 3:05
  immediate action. This processing type is vital
• 3:08
  for human survival and remains our most natural way to respond
• 3:11
  to risk. So these processing types work
• 3:14
  side by side and depend on each other.
• 3:17
  Analytical reasoning isn't more effective unless it's guided by
• 3:20
  emotion. It's just to update around.
• 3:23
  You've probably heard about System 1 and System 2 thinking
• 3:26
  popularised by Kahneman and others.
• 3:28
  Actually, more recent thinking based on a larger body of
• 3:31
  evidence suggests there probably isn't really a System 1 and
• 3:34
  System 2. Rather, it's probably more
• 3:37
  accurate to call them types. They're not distinct systems,
• 3:40
  but they're types of operating in the brain.
• 3:42
  So I'm going to stick with types, Sorry.
• 3:45
  The role of effect and the effect heuristic.
• 3:47
  So effect is a faint whisper of emotion.
• 3:50
  It's an automatic feeling of goodness or vadnais about
• 3:52
  something, how we emotionally respond and feel about it.
• 3:56
  These are effective reactions happening quickly and
• 3:58
  automatically, helping us to navigate a complex world.
• 4:02
  So the effect heuristic describes how we rely on these
• 4:05
  feelings as a mental shortcut. Using an overall emotional
• 4:09
  impression is far easier than and faster than weighing pros
• 4:12
  and cons analytically, especially when decisions are
• 4:15
  complex or we're short on mental resources or time.
• 4:19
  How does it affect these feelings?
• 4:21
  Influence our judgments? Some things that really heavily
• 4:25
  influence our judgments include dread and outrage.
• 4:29
  People's feelings of dread or dread risk about certain risks,
• 4:32
  like nuclear power, heavily influence their perception and
• 4:36
  acceptance, often differing significantly from expert
• 4:40
  assessments. This is linked to factors like
• 4:42
  voluntariness or controllability.
• 4:45
  The less controlled people believe they have over something
• 4:48
  or been exposed to something, generally the poorer they feel
• 4:51
  about it. And while risk and benefit often
• 4:54
  go together, in reality people tend to judge them as inversely
• 4:58
  related. If we feel good about an
• 5:01
  activity, we judge's risk as low and its benefits as high.
• 5:04
  For instance, we typically feel good about the freedom that
• 5:08
  driving provides us, and we feel in control of those decisions to
• 5:11
  drive, so we rate the risk as a lower.
• 5:14
  If we feel bad about something, we judge high risk and low
• 5:17
  benefit. This happens even faster under
• 5:20
  time pressure. For instance, people judge sort
• 5:22
  of good stocks as high return, low risk, and bad stocks as low
• 5:26
  return high risk simply based on their overall feeling.
• 5:31
  And the feelings towards risk can be disconnected from the
• 5:34
  statistical risk. For example, we fear shark
• 5:38
  attacks but not the drive to the beach.
• 5:41
  And importantly, emotion precedes analytical judgements.
• 5:46
  How we feel about something shapes what we think about
• 5:49
  something. Looking at probability and
• 5:52
  frequencies, there's this idea of called imagining the
• 5:55
  numerator. When given a choice, people
• 5:59
  sometimes choose a bowl with a high number of winning items.
• 6:02
  For instance, like 7 in 100 over one with better odds, one in 10
• 6:07
  because seeing more winning items felt like a better choice.
• 6:12
  And also how we frame the risk matters.
• 6:14
  So if we frame it as 10 out of every hundred patients can make
• 6:18
  it sound more dangerous than a 10% chance because the frequency
• 6:22
  creates more vivid, frightening images.
• 6:25
  So importantly, vivid, emotionally charged warnings are
• 6:29
  more effective than just statistics.
• 6:31
  Another example is, let's say, a doctor framing the risk of
• 6:34
  surgery. One death in 100 surgery sounds
• 6:38
  a bit scarier than 99 out of 100 successful surgeries.
• 6:42
  Also, when we're evaluating interventions like in healthcare
• 6:45
  and medicine, people respond often times more strongly to the
• 6:49
  proportion of lives saved. For instance, this intervention
• 6:53
  will save 98% of the participants than the actual
• 6:57
  #150 lives were saved. Because proportions carry
• 7:01
  clearer emotional meaning, people can also be really
• 7:04
  insensitive to probabilities. So for outcomes with strong
• 7:08
  emotional meaning, like a lottery jackpot or cancer,
• 7:11
  changes in probability often have little impact on our
• 7:14
  feelings. So the way that we feel about
• 7:17
  lottery for instance. So whether the chance is 1 in 10
• 7:20
  million or one in 10,000, we often feel the same way about
• 7:24
  it. This helps explain why people
• 7:26
  gamble and buy insurance simultaneously, why fears about
• 7:29
  hazards like nuclear power persist despite extremely low
• 7:33
  probabilities of major failures. But it still has happened, of
• 7:37
  course. How can these feelings go wrong?
• 7:39
  So failures of the experiential system, well, usually helpful.
• 7:43
  Sometimes they can lead us a little bit astray.
• 7:46
  So, for instance, manipulation our emotional reaction can be
• 7:49
  exploited by advertising or propaganda.
• 7:52
  Cults, for instance, play on this very well.
• 7:54
  There's also inherent biases and noise.
• 7:57
  So like something called psychophysical numbing, we're
• 8:00
  insensitive to small changes near 0, so for instance 0 to 1
• 8:04
  deaths, but struggle to emotionally grasp larger changes
• 8:07
  further, from zero 500 to 600 deaths.
• 8:11
  As a 1947 Washington Post article about Stalin
• 8:15
  depressingly quipped, if only one man dies of hunger, that is
• 8:19
  a tragedy. If millions die, that's only
• 8:22
  statistics. Our risk judgments are also
• 8:25
  sensitive to visceral reactions. Strong feelings in the present,
• 8:30
  like hunger or cravings, are hard to accurately call or
• 8:33
  anticipate in the future. So as I said, I've covered a lot
• 8:36
  of ground there. So what can we make of these
• 8:38
  findings? And being very simplistic about
• 8:40
  a very large, broad body of research.
• 8:44
  So it's a bit more difficult to come up with some practical
• 8:46
  implications. But how about some of these?
• 8:48
  And some of these came from the paper.
• 8:50
  One suggestion is using reason to calm strong emotions.
• 8:54
  So when our risk is failing, processing types overweighs the
• 8:58
  frightening but remote consequences like terrorism.
• 9:01
  Our analytical mind can sometimes be really effective to
• 9:04
  help provide perspective on the actual likelihoods.
• 9:07
  Not always. Sometimes you need to fight
• 9:09
  emotion with emotion. They give an example in the
• 9:11
  paper. So fear might lead someone to
• 9:13
  buy a handgun, but analysis shows it's far more lucky to
• 9:16
  harm their own family than an intruder.
• 9:19
  Another suggestion is adding doses of feeling to logic.
• 9:22
  Even highly analytical tasks like proving a theorem or
• 9:26
  playing chess benefit from intuition, like a mathematician
• 9:30
  sensing if a proof sort of looks good, or a chess master feeling
• 9:34
  and move feels right. So risk analysis also needs to
• 9:37
  consider softer values like dread or fairness that drive
• 9:41
  public concerns and sort of counteract this psychophysical
• 9:45
  numbing and make statistics feel real.
• 9:48
  Effective communication can use vivid personal stories or
• 9:51
  physical representations. Another example from the paper
• 9:54
  was piling 38,000 pairs of shoes to represent handgun deaths.
• 9:59
  So in conclusion, meaning is deeply tied to our emotions,
• 10:03
  numbers and statistics about risk.
• 10:05
  Unless they're infused with affect, feelings may not be
• 10:08
  truly representative of risk. How we perceive risk or acted
• 10:12
  upon according to paper rationality is not only a
• 10:15
  product of the analytical minds, but of the experiential mind as
• 10:19
  well. That's it on Safe as I'm Ben
• 10:22
  Hutchinson, please help share, rate and review and
• 10:26
  checkoutsafetyinsights.org for more research.
• 10:29
  Finally, feel free to support Safe As by shouting a coffee
• 10:33
  link in the show notes.