Thinking Fast and Slow by Daniel Kahneman

This outline of the book "Thinking Fast and Slow" is a work in progress (July 2013)

not sure which chapter
- system 1 engages system 2 when it thinks it does not have an answer
- some questions seem plausible so system 1 answers them, and does so incorrectly
  - bat and ball
  - lily pads on lake doubling
  - how many animals of each kind did moses bring on the ark?
- experiment with priming of two posters, flowers vs eye, near “honesty jar” for use of office break room items, eye condition increased donations
  - Bateson, M., Nettle, D., & Roberts, G. (2006). Cues of being watched enhance cooperation in a real-world setting. Biology letters, 2(3), 412-414.
- word completion S__P could be SOAP vs SOUP, depending on priming of food or cleanliness
- people who lied over phone preferred mouthwash, people who lied over email preferred soap
- people primed with ideas of old age walked slower down a hall [apparently not true]
chapter 1: the characters of the story
- example of automatic: interpreting anger on woman’s face
- (p20) example of deliberate: 17 x 24
- system 1: automatic, fast, with little or no effot, no sense of voluntary control
- system 2
  - allocates attention to the effortful mental activities that demand it, including complex computations
  - associated with subjective experience of agency, choice, concentration
chapter 2: attention and effort
- blindness as in Invisible Gorrila
- dilation of pupils clearly shows mental effort
- skill in task decreases demand for energy
chapter 3: the lazy controller
- state of flow “a state of effortless concentration so deep they lose their sense of of time, themselves, of their problems”
- people who are cognitively busy (e.g., remembering digits) are more likely to choose chocolate cake over fruit salad, make selfish choices, use bad language, make superficial judgments
- voluntary effort (cognitive, emotional, physical) draws on same pool of mental energy
- Israel judges approved more paroles after lunch
chapter 4: the associative machine
- system 1 makes associations automatically
chapter 5: cognitive easy
- ways to increase cognitive ease of sentence
  - clear font
  - (p63) high quality paper for high contrast
  - (p63) blue or red (vs green,yellow, pale blue)
  - (p63) simple language
  - (p63) make memorable with rhyme
  - repetition
  - priming
  - good mood
  - putting pencil sideways in mouth to fake a smile
- effects of ease
  - feels familiar
  - feels true
  - feels good
  - feels effortless
  - more intuitive
  - more creative
- effects of lack of ease
  - vigilance
  - suspicion
  - invest more effort
  - feel less comfortable
  - make fewer errors
- (p60) “Becoming Famous Overnight” study
- “a chicken has three legs” is quicker to falsify than “a chicken has four legs” because some animals do have four legs
- repetition can make people believe falsehoods
- (p66) ease and familiarity can be measured as subtle smiles
- (p66) mere exposure effect: people prefer things they have seen
  - ad shown in student newspaper with strange words with varying frequency
  - later people rated it as a good thing correlated with how often it was shown
- (p68) remote association test (rat): triads of words like sore shoulder sweat (which has solution cold)
  - people can feel association is true before knowing the solution
  - mood affects performance on RAT
(03:20) people see causality directly like they see color
Chapter 6: Norms, Surprises, and Causes
- When it saves time, is usually right, and cost of mistake is wrong, a heuristic is good
- System 1 makes a bet
Chapter 7: A machine for jumping to conclusions
- jumping to conclusions is efficient if
  - conclusions are likely true
  - cost of occasional mistake is low
  - jump saves time and effort
- jumping to conclusions is risky if
  - when the situation is unfamiliar
  - the stakes are high
  - there is no time to collect more information
- ambiguous information is interpreted in its context
- system 1 picks the most likely interpretation and unconsciously hides the others because system 1 is trusting
- halo effect
  - tendency to like or dislike everything about a person or thing
  - system 1’s representation is a model that is simpler than the real thing
  - can be self-reinforcing
  - contradictory attributes and new information are interpreted by earlier impressions
  - first impressions make a disproportionate effect
- (p84) James Surowiecki Wisdom of the Crowds: multiple estimates are more accurate when they are independent
  - decorrelate error
  - witnesses in police/justice should not talk to each other
  - before a meeting, each person should write his position
- (p85) WYSIATI-What you see is all there is
  - Information not retrieved by System 1 may as well not exist
  - (p86) system 1 is insensitive to the quantity and quality of information that gives rise to impressions and intuitions
  - (p86-7) people who read background on legal dispute and then heard lawyer’s review of one side were more confident than those who heard both lawyers, so consistency was more important than completeness
  - system 1 likes coherence
  - causes overconfidence, even if important information is missing
  - framing effects: 90% fat free vs 10% fat
  - base rate neglect
chapter 8: how judgments happen
- system 1
  - continuously makes basic assessment
    - is everything normal?
    - is there a threat or opportunity?
    - how are things going?
    - approach or avoid?
    - friend or foe?
    - is a person trustworthy? dominant?
  - substitutes one judgment for another
  - translates values across dimensions
- system 2
  - on demand answers certain questions
  - can handle many attributes
- sets and prototypes
  - system 1 can easily judge the average of a length of scattered lines but not the sum because system 1 represents by prototypes and exemplars
  - groups of participants stated willingness to pay (WTP) to save 2K, 20K, or 200K birds after Exxon Valdez. No difference in WTP by quantity.
- (p93) intensity matching
  - e.g. “how tall is a man who is as tall as Julie was precocious?” (she could read at four years old)
- mental shotgun
  - excess computations
  - in experiment it took longer to distinguish heard rhymes vote vs goat (different spelling) than vote vs note (same spelling) because of excess computations
  - similar experiment with sentences literally true or false, subjects (excessively) whether metaphorically true
- chapter 9: answering an easier question
  - e.g. like or dislike people before you know them
  - target question vs heuristic question
  - e.g., “how much would you like to contribute to save an endangered species” vs “how much emotion do i feel when i think of dying dolphins”
  - after substitution, the substituted answer must be fitted back to original question, so for example expressing feeling of dolphins in dollars
  - often system 2 does not notice
- 3d heuristic
- mood heuristic for happiness
  - experiment: asked same two questions in random order
  - q1: “how happy are you these days?”
  - q2: “how many dates did you have last month?”
  - if q1 asked first, no correlation
  - if q2 asked first, strong correlation because students had in mind an answer to related question
- affect heuristic
  - e.g., liking or disliking nuclear power affects believes about benefits and risks
  - (p103) in attitudes, system 2 is more of an apologist for the emotions of system 1 than a critic of those emotions