AI summary by TwinMind

Paul Slovic | Confronting the Deadly Arithmetic of Compassion | Talks at Google

Talks at Google

50m 22s

Speaker Introduction & Background

  • Paul Slovic: Stanford bachelor's, University of Michigan PhD in psychology (1976)
  • Founded Decision Research institute with Sarah Lichtenstein and Baruch Fishoff (1976)
  • Professor of psychology at University of Oregon since 1986
  • 60 years researching risk perception, starting with simple gambling experiments
  • Research evolved from casino gambling prototypes to real-life risks (genocides, nuclear war, climate change)
  • Extensive mentoring worldwide, organizes virtual get-togethers for community building

Evolution from Gambling to Risk Research

  • Started 1959 at University of Michigan studying personality theory
  • Got involved with professor testing economic theories about utility maximization in gambling decisions
  • Professor skeptical that people maximize utility through calculations
  • 60 years later still studying "gambles" - now real-life risks we face daily
  • Journey from abstract gambling experiments to humanitarian problems

Individual Risk Perception - System 1 vs System 2 Thinking

  • Initial assumption: people do mental calculations (probability × payoff = expectation)
  • Reality: most people rely on intuitive gut feelings, not calculations
  • Risk is mostly represented as a feeling, not result of calculation
  • System 1 (Stanovich & West, popularized by Kahneman 2011): fast, intuitive, gut feeling
    • Easy, feels natural, seems right, usually works
  • System 2: analytic, uses reasons, calculations, logic (taught in school)
  • Most time spent in System 1 mode because it's easy and efficient
  • Focus on when System 1 fails and how to defend against feeling-based mistakes

Community Risk Perception & Bystander Effect

  • Individual life valued very highly - people risk own lives to save others
  • As numbers increase, response doesn't increase proportionally
  • First life most important, second doesn't feel twice as important
  • By 6-7 lives, additional lives don't feel different
  • Feelings can't do math - arithmetic of compassion is flawed
  • Example: 200,000 vs 400,000 deaths in genocide feels the same (psychic numbing)
  • Large numbers bounce off without creating feelings
  • Failure to feel reality of large numbers leads to tolerating massive harms

The Deadly Arithmetic of Compassion - Three Components

1. Psychic Numbing

  • Concept originated with psychiatrist Robert Lifton studying atomic bomb rescue workers
  • Workers had to turn off feelings to function (good side of psychic numbing)
  • Dark side: insensitivity to large numbers of victims
  • Donation experiments: response declines as number of children needing help increases
  • Rwanda refugee camp experiment:
    • Scenario: official deciding whether to send pumping equipment for clean water
    • Could save 4,500 people indefinitely
    • Group 1: 4,500 out of 11,000 total in camp
    • Group 2: 4,500 out of 250,000 total in camp
    • Much higher willingness to help when saving 4,500 out of 11,000 vs 250,000
    • People respond to proportion saved, not absolute number

2. Pseudo Inefficacy (False Inefficacy)

  • Linked to sense of efficacy in making difference
  • Going from 250,000 to 245,000 doesn't feel effective
  • Even small numbers of people you can't help dampens response
  • Experiment: helping one child when six others can't be helped reduces donations
  • Even helping one of two children (instead of just one) reduces response
  • Warm glow effect: satisfaction from helping someone diminished when others can't be helped
  • False sense of inefficacy prevents helping even when you really can make difference

3. Prominence Effect (Security Prominence)

  • Occurs when values conflict (e.g., stopping genocide vs national security)
  • Moral obligation to stop genocide conflicts with:
    • National security concerns
    • Military costs
    • Political relationships with other nations
  • When facing conflict, people focus on most defensible dimensions
  • Security, homeland protection, cost-cutting more defensible than protecting "statistical lives"
  • Security dominates choices regardless of lives lost
  • Makes people choose in ways that violate stated values

Climate Change Application

  • Scientists warning for half century, only paying attention when consequences hit personally
  • All three elements apply to climate change:
    1. Psychic numbing: problem seems distant, happening to others, lacks emotional connection
    2. Pseudo inefficacy: individual actions feel like "drop in bucket"
    3. Prominence effect: near-term comforts/convenience outweigh long-term planetary benefits
  • Solutions require government/industry action:
    • Can't rely only on warnings (like UN reports)
    • Need carrots and sticks - rewards for right behavior, costs for wrong behavior
    • Provide options to satisfy needs with less climate harm
    • Higher costs for flood insurance, fossil fuel use
    • Renewable energy technologies and alternatives

Communication Strategies for Climate Change

  • Abstract numbers (1.5-2.5°C temperature rise) don't create emotional connection
  • Need translation to experiential terms:
    • Percentage of days over 40°C in specific regions
    • Visual representations of sea level rise effects on familiar cities/beaches
  • Make information come alive with feeling rather than abstract statistics

COVID-19 as Behavioral Problem

  • Exponential growth misunderstanding: people extrapolate linearly, vastly underestimate exponential curves
  • Early action most effective for exponential spread, but people didn't grasp urgency
  • Reward/punishment backwards:
    • Doing right things (masks, isolation): immediate punishment (restrictions), no immediate reward
    • Doing wrong things (parties, no masks): immediate reward (freedom), no immediate punishment
  • COVID as much behavioral problem as medical problem
  • Need to combine behavioral knowledge with medical knowledge

Resources & Websites

  • decisionresearch.org: Decision Research institute publications (free access)
  • arithmeticofcompassion.org: specialized site with:
    • Information on psychic numbing, pseudo inefficacy
    • Blogs connecting psychological factors to current world events
    • Take action section with practical steps
    • Welcome contributions for blog section

Personal Well-being & Research Approach

  • Finds purpose and excitement in learning things that may be helpful
  • Works collaboratively with community of researchers (not alone)
  • Admits to own psychic numbing - discusses problems from statistical distance
  • Wife works directly with refugees locally, experiences emotional reality
  • Contrasts abstract statistical work with personal, emotional refugee assistance work

TwinMind

Get summaries like this for your own meetings

TwinMind takes notes in the background during meetings, lectures, and calls, then writes the summary for you.

Free plan to get started

No bot joins your calls

Phone, desktop & Chrome apps

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome