emotional adaptation and the end of economics n.
Download
Skip this Video
Loading SlideShow in 5 Seconds..
Emotional Adaptation and the End of Economics PowerPoint Presentation
Download Presentation
Emotional Adaptation and the End of Economics

Loading in 2 Seconds...

play fullscreen
1 / 79

Emotional Adaptation and the End of Economics - PowerPoint PPT Presentation


  • 86 Views
  • Uploaded on

Emotional Adaptation and the End of Economics. Peter A. Ubel, M.D. Center for Behavioral and Decision Sciences in Medicine Ann Arbor VAMC University of Michigan Health System . The answer depends on who you ask. TTO utility of ESRD (0-1 scale) Patients = .56 Community = .39

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about 'Emotional Adaptation and the End of Economics' - ismael


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
emotional adaptation and the end of economics
Emotional Adaptation and the End of Economics

Peter A. Ubel, M.D.

Center for Behavioral and Decision Sciences in Medicine

Ann Arbor VAMC

University of Michigan Health System

the answer depends on who you ask
The answer depends on who you ask
  • TTO utility of ESRD (0-1 scale)
    • Patients = .56
    • Community = .39
  • Moods (-2 to +2 scale)
    • Patients = .66
    • Community prediction of patients = -.17
economic importance of this discrepancy
Economic Importance of This Discrepancy
  • Whose utilities should we include in cost effectiveness analyses?
as if the who question wasn t enough
As if the “who” question wasn’t enough!
  • We need to figure out what question to ask
  • Specifically: should policy decisions be based on
    • decision utility or
    • experience utility
a quick and inaccurate history of economics
A Quick and Inaccurate History of Economics
  • Economics = Science of utility maximization
  • Original notion of utility
    • Jeremy Bentham
    • Balance of pleasure & pain = Experience utility
  • More recent view of utility
    • Revealed preferences
    • Rational people’s free choices lead to utility maximization

= Decision utility

what happened in this study
What happened in this study?
  • Experience utility –
    • The 60 second bucket was better than the 90
  • Decision utility –
    • The 90 second bucket was better
  • People misremembered their 2 experiences, causing them to make a bad decision
goals of talk
Goals of Talk
  • Potential flaw of decision utility
    • Based on mispredictions and misrememberings of experience utility
  • Potential flaw of experience utility
    • Goals of healthcare go beyond mood maximization
  • Point to future research directions
    • Empirical
    • Normative
global versus momentary reports of well being swb
Global versus Momentary Reports of Well-being (SWB)
  • People have difficulty describing average emotions over time
  • There may be a discrepancy between
    • moment to moment SWB
    • general evaluation of SWB
no pain no gain

High

Pain

Pain

60

60

90

Time

Time

No pain, no gain

High

Pain

imagine a dialysis patient s quality of life
Imagine a Dialysis Patient’s Quality of Life
  • How happy are you right now (0 - 10)?
    • 6
    • 5
    • 5
    • 6
    • 8
  • How happy do you feel generally?
    • 7
slide17

+2

Very pleasant

+1

Slightly pleasant

0

Neutral

-1

Slightly unpleasant

-2

Very unpleasant

100

Entry: Percentage of time, during a typical week, that you are in each of the following moods?

predicted moods

Predicted Mood

Patients

.78

Controls

.61

Predicted Moods
predicted vs actual mood

Actual Mood

Patients

.78

.66

Controls

.61

.80

Predicted vs. Actual Mood

Predicted Mood

specific moods 0 6 scale palm data

4 positive measures

5 negative measures

Patients

3.21

1.00

Controls

3.23

.99

Specific Moods (0-6 scale)- Palm Data -
a whole lot of mispredicting going on
A Whole Lot of Mispredicting Going On
  • Patients
    • Mispredict life without kidney disease
  • General public
    • Mispredict life with kidney disease
  • These mispredictions of experience utility could influence decision utility
hope i die before i get old
Hope I die before I get old!
  • Common belief (among rock stars, at least)
    • Getting old stinks
  • But happiness data suggest that
    • Happiness relatively stable across adult years
      • Not much change between 30 and 70
      • If anything, less negative affect with age
  • What do real people believe?
    • < 40 year olds
    • > 60 year olds
  • And how do beliefs compare to “reality”?
study design
Study Design
  • Internet study of people
    • 21 – 40 (n = 264; median age = 31)
    • 60 and up (n = 255; median age = 68)
  • Asked to:
    • Report current happiness (0 – 10 scale)
    • Estimate happiness of typical
      • 30 year old
      • 70 year old
results self reported happiness
Results: Self reported happiness

10

0

7.3

Happiness

6.6

<40 > 60

results predictions of 21 40 year olds
Results: Predictions of 21–40 year olds

10

0

7.3

7.0

Happiness

6.6

6.2

<40 > 60

results predictions of old people
Results: Predictions of old people

7.6

10

0

7.3

7.0

Happiness

6.6

6.3

6.2

<40 > 60

a whole lot of mispredicting and misremembering going on
A whole lot of mispredicting and misremembering going on
  • Young people mispredict happiness as they age
  • Old people misremember happiness of their youth
  • People are out of touch with their own experience
misestimating the benefits of kidney transplantation
Misestimating the benefits of kidney transplantation
  • Surveyed patients waiting for kidney transplant at University of Pennsylvania
    • Measured QoL
    • Asked them to predict QoL 1 year after successful transplant
  • Resurveyed them after transplant
    • Measured QoL
    • Asked them to remember pre-transplant QoL
mispredictions
Mispredictions

Prediction for Post-tx

Actual Post-tx

Domain

Pre-tx

QoL (0-100) 66 91

83

Travel (days/yr) 9 20

12

Work (hrs/wk) 12 32

15

Energy (1-5) 3.2 5.1

4.3

more mispredictions
More mispredictions

Prediction for Post-tx

Actual Post-tx

Domain

Pre-tx

Happy w/ life 2.8 3.9

3.5

…health 2.4 3.9

3.6

…std. of lvg. 2.6 3.7

3.1

…work 2.5 3.6

3.1

…love life 3.5 3.8

3.5

…family life 3.8 3.8

3.8

…social life 3.5 3.8

3.6

…spiritual life 3.4 3.7

3.6

…leisure life 3.0 3.8

3.3

misremembering esrd
Misremembering ESRD

Time

Pre-tx QoL

Post-tx QoL

Pre-tx 65 91

Post-tx:

Immediate 57 78

6 months 55 80

12 months 48 83

colostomy patient survey
Colostomy Patient Survey
  • Surveyed people who have received colostomies at UM within last 5 years
    • some had permanent and some had reversed colostomies
      • 94 permanent
      • 100 reversed
  • What do these two groups think of life with a colostomy?
adapting to life with a colostomy
Adapting to life with a Colostomy
  • Overall quality of life (0-100)
    • permanent = 67
    • reversed = 71
  • Overall positive affect positive mood (0-4)
    • permanent = 3.1
    • reversed = 3.1
  • Overall negative mood (0-4)
    • permanent = 1.8
    • reversed = 1.9

So ... little to no difference in mood or quality of life

how happy now 5 yrs ago
How Happy: Now & 5 yrs. ago

7

6.8

6.4

6.1

6

6.0

5

Now 5 yrs. ago

how bad do these groups think it is to have a colostomy
How bad do these groups think it is to have a colostomy?
  • Time tradeoff (TTO) utility question
    • Imagine you will live 10 years with a colostomy then die.
    • How many months (0-120) would you give up to get rid of the colostomy?
      • permanent = 18 months
  • reversed = 44 months
rational decision making and revealed preferences
Rational decision making and revealed preferences
  • Ua = P1U1 + P2U2 + …
  • Ub = P7U7 + P8U8 + …
  • If I chose A over B
    • Then Ua > Ub
  • Similarly, if I’m willing to pay
    • $10 for X
    • $15 for Y
      • Then Uy > Ux
flaws with revealed preference assumptions
Flaws with revealed preference assumptions
  • People mispredict utilities
    • As I’ve shown already
  • Even given utilities
    • People don’t always integrate p’s and u’s in rational manner
  • Here’s an example
imagine you have colon cancer
Imagine you have Colon Cancer

Surgery A

    • 80% cure without complications
    • 16% die of disease
    • 1% colostomy
    • 1% intermittent bowel obstruction
    • 1% wound infection
    • 1% diarrhea
  • Surgery B
    • 80% cure without complications
    • 20% die
  • Which surgery would you choose?
give me colostomy or give me death
Give me colostomy or give me death!
  • When we ask people to choose between being dead or having one of these four complications
    • >90% prefer each of the four complications to death
  • To be consistent with these preferences
    • <10% should choose the uncomplicated surgery
our survey says
Our Survey says . . .
  • 50-60% of people choose surgery B, the uncomplicated surgery
kahneman s case against decision utility
Kahneman’s case against Decision Utility
  • Based on
    • mispredictions of utility
    • poor integration of problem and utility
  • If we want to maximize utility, we should measure experience utility and devise policies/practices that maximize it
narrowing the scope of debate
Narrowing the scope of debate
  • Lots of problems with utilitarianism
    • consequentialism: ends over means
    • maximizes average “utility” without regard to distribution
    • etc. etc.
  • But consequences domatter to decisions
    • choose dinner at restaurant
    • buy a movie ticket
    • invest in a health care system
  • Surely we want to understand distinction between decision and experience utility
advantages of experience utility as welfare criterion
Advantages of experience utility as welfare criterion
  • People generally want to be happy
    • But they are often unaware of what would make them happy
  • Recent advances allow for more accurate measures of happiness, mood and other experiences
    • Ecological Momentary Assessment
    • Experience Sampling
    • DRM
  • Policy should be informed by
    • Actual experience
    • Not mispredicted experience
current approach to experience utility
Current approach to experience utility
  • Focus = mood
  • Outcomes = maximization of mean mood
    • Integral of momentary affect
  • Thus, for example
    • -3, -3, 4, 4, 4 is better than
    • 1, 1, 1, 1, 1
a thought experiment
A thought experiment
  • Imagine that you are about to receive a below the knee amputation (BKA)
    • You will recover, physically, quickly
    • You will receive a top-of-the-line prosthesis
    • Physical function – almost normal
      • Able to play sports
      • Sprinting and jumping mildly reduced
a thought experiment continued
A thought experiment - continued
  • Imagine also that you completely adapt emotionally
    • Mood indistinguishable, on average, from prior to BKA
    • Some pangs of
      • Loss
      • Stigma
    • Balanced by positive emotions from lessons learned
a question about our thought experiment
A question about our thought experiment
  • How much would you pay to avoid BKA?
hedonic adaptation
Hedonic Adaptation
  • Our claim
    • Adaptation to positive and negative circumstances is not
      • Immediate
      • Universal
      • Complete
    • But is often
      • Rapid
      • Common
      • And substantial
relevance of adaptation to debate about experience utility
Relevance of adaptation to debate about experience utility
  • If moods largely return to normal after good and bad circumstances
  • Then policies based on experience utility
    • Won’t care too much about people’s circumstances
counterargument
Counterargument
  • Experience utility consists of things other than mood
  • Our reply
    • We agree, partially
      • Thus, the job of experience utility proponents should be to better capture the aspects of experience that people care about
      • But we also expect that people care about things that go beyond what would ever qualify as experience utility
what is missing from experience utility
What is missing from experience utility?
  • Mill’s higher and lower pleasures
    • Better to be an unhappy person than a happy pig
    • Consider: wine connoisseur
  • Meaning and purpose
    • Raising young children
  • Evaluation of experiences matters
    • Consider two movies
what is missing from experience utility1
What is missing from experience utility?
  • Capabilities
    • Walking in the woods with your children
  • Brief episodes
    • Death of a loved one
  • Self-identity
    • My mini-van
  • Moral considerations
    • I do it even though it won’t make me happy
empirical exploration of distinction between experience and decision utility
Empirical exploration of distinction between experience and decision utility
  • Developed intervention to help people take account of adaptation when making affective forecasts
    • Think of bad event from more than 6 months ago
      • more or less upsetting than predicted
      • emotions stronger or weaker over time?
    • List the 2 most upsetting things about becoming paraplegic
    • Do you think these 2 things would become more or less upsetting over time?
thinking about adaptation changed qol estimates
Thinking about adaptation changed QoL estimates

QoL Rating (0 - 100)

Disability N Before After P

Paraplegia 123

Paraplegia 56

.003

.001

47

-

52

62

thinking about adaptation changed policy recommendations
Thinking about adaptation changed policy recommendations
  • Given choice between saving the lives of
    • 100 people who can be returned to perfect health
    • X people who would experience onset of paraplegia
      • X = 1000
  • When given same choice after thinking about adaptation
    • X = 101
thinking about adaptation di not change decision utility
Thinking about adaptation di not change decision utility
  • Standard gamble to elicit utility of paraplegia
    • What chance of death would you take to be cured of paraplegia
  • Time Tradeoff elicitation
    • Imagine you will live 10 more years
    • How many months of that time would you give up to be cured of paraplegia
  • Adaptation exercise
    • Did not influence responses to either elicitation
these 3 studies suggest that
These 3 studies suggest that
  • Thinking about adaptation changes
    • QoL estimates
      • And potentially experience utility estimates
  • But does not change
    • People’s decision utility
what criteria should guide public policy1
What criteria should guide public policy?
  • Two candidates, with limitations
    • Decision utility/revealed preferences
    • Experience utility/ moment-to-moment mood
  • Given that neither option is ideal
    • What should we use?
  • We propose two answers
    • Because there are two types of policies that require a welfare criterion
two kinds of policy
Two kinds of policy
  • Policies that are designed to directly affect the behavior of individuals
    • Tax laws promoting IRAs
    • FDA rules requiring food labels
    • Seat belt laws
  • Policies that operate at a level higher than the individual
    • Local government deciding how much to spend on schools vs. cops
    • Federal government corporate taxation policy
  • Distinction between these two types of policy is not sharp
policies that guide individual behavior
Policies that guide individual behavior
  • Standard approach
    • Policies should enable people to make free, informed choices
    • e.g. FDA food labels
  • A step further
    • Libertarian paternalism, soft paternalism, pruned decision utility…
    • Set up decisions so that people are more likely to make optimal decision
      • e.g. Default options for insurance
      • e.g. Physician recommendations
but what criteria on should guide things like default options
But what criteria/on should guide things like default options?
  • Fact = People influenced by status quo bias/defaults options bias in choosing insurance
    • So how do you chose the proper defaults?
  • Our proposal
    • Base policy on experience utility
      • A maximally thick account of experience
    • Inform people about experience when feasible
    • And let them decide
why experience utility
Why experience utility?
  • Because experience utility does matter to people
    • And without “help”, they’ll mispredict it and make potentially bad decisions
  • And because they can still chose to base their decision on non-affective or non-experiential issues
policies that operate above the level of the individual
Policies that operate above the level of the individual
  • We don’t have a simple answer
    • No single, appropriate welfare criterion
  • Policy makers need to consider experience utility
    • But need to take account of other things that matter to people
    • Some hybrid of
      • Decision utility
      • Experience utility
possible approaches
Possible approaches
  • Deliberative democracy
  • Informed decision utility
  • But I’d like to see experimental studies evaluation those approaches
backing up our claim
Backing up our claim
  • Test of global vs. momentary affect
    • Dialysis study: EMA design
  • Test of conversational norms
    • Parkinson’s study
  • Test of scale recalibration
    • Diabetes and emphysema study
acknowledgements
Dylan Smith

George Loewenstein

Heather Pond Lacey

Laura Damschroder

Chris Jepson

Harv Feldman

Norbert Schwarz

Ryan Sherriff

Brian Zikmund-Fisher

Jason Riis

NIH

Dept. of Veterans Affairs

Acknowledgements