🪴 Maixnor WU

Decision Making

Nov 04, 20254 min read

Rationality

  • assumptions:
    • unlimited cognitive resources
    • perfect information
    • utility maximization

Realistic Decision Making

  • don’t have limitless cognitive capacity or willpower
  • not entirely selfish
  • psychological realism → better predictability
  • more realism is not always better
    • “all models are wrong, but some are better for what you need”
  • problem of Unfalsifiability in psych models
  • risky decision: optimize for expected value:
  • Knightean Uncertainty vs radical uncertainty (not knowing anything)

Choice under Risk

  • you know the outcomes and the percentages (lottery setting)
  • not just expected value → St Petersburg Paradox
  • expected utility more important in decision making
    • utility is not a linear function, more like a log function
  • not maximizing expected value, but expected utility
    • large gains should not be weighted as heavy as small gains

4 Axioms of Rational Behavior

Decision Making

  • person vs situation
  • todo
  • 1 problem recognition
  • 2 information search
    • internal (memories) vs external (research, availability)
    • active progressing is done in Working Memory

Intuition

  • intuition depends a lot on the information you have
    • shit in - shit out
  • the better your inputs (memories, biases, etc) are the better your intuitions are the better decisions you will make
  • if you have a bad (inaccurate) gut feeling you can change this with better inputs
  • justification only after making a choice → only looking for supporting evidence, not for counteracting arguments

Indifference

  • if choices are equal or only marginally different
  • problem: infinite decision time, Buyers Remorse
  • solution: picking randomly

Others

  • asking other people
  • e.g. an export on the topic, a friend, a trusted individual
  • just speaking it out changes a lot
  • outsourcing: coin, person, algorithm, /dev/random

Dealing with Limitations

  • trade off between efficiency and accuracy (Simon, 1957)
  • moral wriggle room (putting off a decision)
    • e.g. not opening the door to a salesman to say no

Construction

  • perception is influenced by
    • context factors
    • previous experiences
    • expectations
    • motivations
  • framing
    • experienced entrepreneurs perceive less uncertainty and are less affected by framing issues

Prospect Theory

  • editing phase
    • simplify decisions (cutting off branches of decision tree)
    • reduce consideration set
    • cancelling options out or simplifying
  • evaluation phase
    • consider values relative to one another
    • consideration of probabilities
  • value of losses weighted more than value of losses
  • value is rather wealth changes, not absolute wealth level
  • risk-averse for gains, risk-seeking for losses
    • gambling addiction
  • probability weighting
    • overweighting low probabilities, underweighting high probabilities
    • overestimate luck, good and bad
    • certainty (say 1%, 99%) are undervalued

Heuristics

Mental shortcuts to make judgement quickly and efficiently.

& Aronson el al. 2019, p. 64

Availability

  • if you think of it, it must be important
  • not as many people die from shark attacks as from lightning strikes
  • but people are worried more of sharks, because they can think of it
  • if availability is good or not depends on what your sample is
  • emotional events will be more available
  • problematic implications:
    • anything that is not available is not regarded
    • distortion or abuse of public order
    • bad allocation of funds

Insensitivity to Sample Size Bias

  • failing to appreciate role of sample size

Conjunction Fallacy

  • an and gate can only ever reduce the probability
  • “certain” combinations are considered more probable than the individual factors, although it can only be more probable

Confirmation Bias

  • accepting supporting data as is
  • trying to reject disproving data
  • “You see what you expect to see”
  • Dissonance
  • remedy: Devils Advocate, outside counsel
  • dark side of dissonance:
    • escalation of commitment
    • sunk cost fallacy
    • irrational ignorance towards relevant information
    • ethical dissonance
    • overweighting of expensive advise
    • discounting of feedback
      • just because a bad person said a good thing does not make it bad
    • absenteeism
    • resistance to change

Examples Confirmation Bias

  • thinking a university course is bad, then you perceive it as bad even though you would like otherwise if not told that before

Anchoring

  • todo
  • price set in beginning is really important

Backlinks