[MUD-Dev] Charlie Munger on the Psychology of Human Misjudgment [summary of points]

[Sasha Hart]

The lecture identifies a number of possible sources of bias in
decisions.  It's dense, and a lot of ground is covered by
reference. I found the ordering and relation of the points very
difficult to follow, and the presentation of each point to be a
little murky. So here's my summary (with my skew and my mistakes as
well).

  1. Payoffs and their value affect decision-making.

  2. Contrasts in the magnitude of payoffs themselves affect
  decision-making.

    - Prospect: the contrast between what I expected and what I got.

    - Behavioral contrast: the contrast between what I've been
    getting and what I just got.

    - Envy: the contrast between what other people are getting and
    what I get

  3. Actions are affected by predictions of payoffs. Predictions of
  payoffs are themselves made on the basis of limited information,
  and are often heuristic.

    A. Pavlovian conditioning - past predictors, assessed almost
    correlationally, come to elicit preparation for what they
    predict.

    B. The opinions of many others, or of authorities, affect our
    judgements.
 
    C. The availability heuristic - If I hear a story about a plane
    crashing and the next day someone asks me about the likelihood
    of plane crashes, it's higher than if I hadn't heard the story.
    Even higher if the story was quite vivid.

    D. Persistence can be induced by irregularity of payoffs -
    payoffs serve as degraded symbols of their own availability,
    e.g., if the slot machine pays out variably, a single failure to
    pay out may mean less (an interpretation of the variable ratio).

  4. Beliefs (often also predictions, having effect on
  decision-making) are affected in ways which are apparently
  inappropriate, not just adaptations to limited information.

    A. People hang on to beliefs in the face of disconfirming
    evidence.

    B. People believe things about themselves, their family, or
    their stuff which are rosier than seems warranted.

    C. People believe things which make them happy because it would
    otherwise be too painful.

  5. Malfunctions (big category)

    A. Normal memory & cognitive limitations

    B. Stress induced changes

    C. Drug addictions, mental illnesses and declines

  6. Behavior also skews with typical social demands:

    A. Reciprocity - e.g., players more inclined to cooperate/defect
    on someone who cooperates/defects, respectively.

    B. When it comes to beliefs which are about incentives or
    instrumental to their achievement, we can misrepresent facts or
    beliefs to others (deliberately or without realizing, e.g. via
    self-deception).

    C. When there is an expectation to do or say something, one will
    often bend to that demand despite not really knowing what to do
    or say that is appropriate.  (??)

    D. A lot more that should have been mentioned, and should be
    here, but isn't.

    E. This has a rich relationship with the decisions & beliefs
    stuff above.

The substantiation, comprehensiveness and representativeness of the
claims is variable, but some of that comes with the territory,
AFAIC.

Sasha

_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev