2018-01-29 RRG Notes
- Scott realizes that his experience with psychotherapy isn't anything like those of his colleagues
- Scott's patients gave calm and considered analyses of their problems
- His colleague's patients all had dramatic emotional breakdowns
- Scott's supervisor noted that he seemed to be uncomfortable with dramatic expressions of emotion, even though he was actively trying to hide that fact
- Scott was able to turn this around into a reputation for being able to deal successfully with really difficult patients who have a lot of emotional breakdowns
- This ability of Scott's has been described as a "niceness field"
- This means that Scott's lack of success with psychodynamic therapies might be due more to Scott's own personality than to the limitations of psychodynamic therapy itself
- Paranoia and Williams Syndrome
- Paranoia is a common symptom of a lot of psychiatric disorders, most notably schizophrenia
- The troubling thing about paranoia is how gradual it is
- Instead of thinking the CIA is after you with mind-control rays, you'll just interpret ambiguous social signals a bit more negatively
- This can lead to a self-reinforcing feedback loop, as the person becomes more and more standoffish in response to perceived slights from others
- Williams Syndrome is the opposite of paranoia
- People with Williams Syndrome are are "pathologically trusting"
- Literally incapable of distrust
- Williams Syndrome is usually, but not always coupled with mental retardation
- However, IQ doesn't seem to have much of an impact on William's Syndrome - it seems like threat detection is an automated process, not well controlled by conscious analysis
- Psychiatric disorders are often just the extremes of normal human variation
- For every person who is diagnosed autistic, there are a dozen people who are awkward and weird
- For every intellectually disabled person, there are a dozen that are just kind of slow
- Maybe for every person diagnosed with Williams Syndrome, there are a dozen that are just more "trusting" than others
- Our sense data is underdetermined
- Each data point that our senses receive can be interpreted in multiple ways
- This is especially true of social cues
- Most people are able to navigate this ambiguity with context, i.e. priors
- However, these priors can vary from person to person
- Human society permits quite a lot of variation before we decide that someone is so off-base that they need to be excluded
- Just as there's a spectrum from smart to dumb or a spectrum from introverted to extraverted, there may be a spectrum from completely trusting to completely paranoid
- Bubbles
- 46% of Americans are young-earth creationists
- However, even though Scott isn't selecting friends on the basis of politics, religion or class, he has approximately zero friends who are young-earth creationists
- Maybe instead of Scott excluding young-earth creationists, the young-earth creationists are excluding Scott - if it's their policy to not make friends with people who don't believe in young-earth creationism, then it's possible for them to end up in a bubble, even though people outside the bubble aren't necessarily excluding them
- Some other bubbles that Scott lives in:
- Transgender - people in Scott's circle of friends are 20x as likely to be transgender as the general population
- 2x as many Asians as the general population
- Half has many African Americans as the general population
- Depression, OCD and autism are high
- Drug addiction and alcoholism are low
- Programmers overrepresented at 10x the Bay Area average
- None of these bubbles were intentionally created
- Moreover, some of these bubbles have persisted in the face of conscious efforts to pop them
- This goes double for relationships - even though Scott doesn't think of himself as having a "type" all of the people he's dated have been similar in ways that he didn't expect when he first met them
- This bubble theory is something that Scott thinks about when he meets serial abuse victims
- Serial abuse victims are people who have been abused by multiple people in a row
- Often abused by the people they to go to seek relief from the abuse
- Offensive explanation: seek out abusers because for some reason they've internalized a model that defines abusive relationships as "correct"
- While this may be true of some victims, it doesn't seem to be true of many
- Go to great lengths to avoid abusers, but it doesn't seem to matter
- In the same way that Scott finds himself in a bubble of transgender programmers, these people might find themselves in a bubble of abusers
- Discrimination
- Some women in the tech. industry experience a constant litany of harassment and discrimination, whereas other women go their entire careers without experiencing a single harassment event
- There doesn't seem to be any correlation between industries, companies or physical attractiveness
- Given the baseline rates of discrimination reported by others, it can't be just luck for some of these people to go their entire lives without experiencing any discrimination at all
- These two forces, self-selected bubbles and the ambiguity of social cues combine to create different worlds for different people
- People unconsciously self-select into bubbles
- People vary in how they perceive social interactions
- Discrimination is rarely as blatant as people being called out for their race or gender directly
- There is usually some room for interpretation ("Was I being discriminated against there?") which means that different people will perceive discriminatory experiences differently
- Are people basically good or basically evil?
- Some people say that the world is full of hypocritical backstabbers
- Others say the world is full of basically decent people who are hampered by communications difficulties and differences in values
- Both groups are basically correct, because they just see different slices of the world
- This applies on all sorts of axes, not just good/evil
- Are people basically rational or basically emotional?
- Are people welcoming of outsiders or shunning of outsiders?
- Etc
- The concept of "privilege" gets part of the way to capturing these differences of experience, but privilege has the limitation of insisting that these differences have to line up along predefined categories like race or class
- Knowing that someone is living in a different world from you can go a long way towards making their behaviors more comprehensible
- The Narrative Fallacy
- A typical biography starts by describing the subject's younger life and tries to show how the young person was an early version of the person that they would become
- Steve Jobs
- Biographers play up the fact that he was adopted and imply that this led to a need for him to prove himself
- Nassim Taleb
- Describes how a professor who had read his earlier book ascribed his ability to separate cause and effect to his growing up in an Eastern Orthodox society
- The problem is that both of these narratives are contradicted by their subjects
- Steve Jobs actively denied that his being adopted had anything to do with his later success
- Nassim Taleb looked at others in the financial industry who came from the same background as him, and found that none of them become skeptical empricists
- The narrative fallacy exists because of a biological problem
- Too much sensory information to process events independently
- We have to put things in order so that we can process the world around us
- The world does not make sense without cause and effect
- While our tendency to order the world into narratives works well in general, sometimes it causes us to make errors
- The problem with narrative is that it lures us into believing we can explain the past through cause-and-effect when we hear a story that supports our prior beliefs
- Example: sports
- Every profile of an athlete has roughly the same form
- Natural gift for the sport
- Parents or coaches that pushed them to strive for excellence
- Hard work ethic
- Some kind of adversity of impactful life-event
- But we don't stop to ask ourselves why this person succeeded when the thousands of other people who have the same backgrounds failed
- Narratives cause us to miss the influences of luck and timing
- Narratives also cause us to ignore the mathematical rules of probability
- Nassim Taleb talks about the example of a good detective novel making it seem like every suspect must be the criminal, right up until the final reveal
- Kahneman talks about how people are willing to ignore the fact two things together must be less likely than one thing alone if the two things together fit a pre-packaged narrative
- Narratives also ignore regression to the mean
- All stories of success have a fair amount of luck in them
- Eventually this luck runs out and the person or business reverts to the mean
- This doesn't mean that they're any worse, only that they're not as lucky as they used to be
- The problem with narratives is that we make them predictive and by doing so we make them more real than they actually are
- A close cousin of the narrative fallacy is the reason-respecting tendency
- People are willing to comply with those who give reasons for those orders, even when those reasons are meaningless or irrational
- Example: people are able to jump to the head of the line at a copy machine by simply stating that they have to make copies, even though everyone else in the line has exactly the same reason
- This is because reasons allow us to build narratives
- This is why teaching that gives reasons for facts is so much more effective than teaching that asks us to memorize the bare facts themselves
- This means that the best teaching, learning and storytelling methods (those that involve reasons and narrative) can also cause us to make our worst mistakes
- So how do we help ourselves out of this quagmire?
- Become aware of the problem
- The key question to ask is, "Out of the population of X subject to the same initial conditions, how many turned out similarly to Y?"
- "What hard-to-measure causes may have played a role?"
- Modern scientific thought is built on top of efforts to solve this problem
- The entire notion of a hypothesis comes from the fact that people recognized that simple narrative explanations were not sufficient to explain the world
- Narratives have to be experimentally tested before they can be accepted as true cause-and-effect relationships
- Another question we can ask ourselves is, "Of the population not subject to initial conditions X, how many ended up with the results of Y?"
- Which basketball players had intact families, easy childhoods and ended up in the NBA anyway?
- Which corporations lacked the traits talked about in business books but ended up successful anyway
- We can also reduce our vulnerability to the narrative fallacy just by reducing the number of narratives we consume
- Stop watching TV news
- Be skeptical of biographies, memoirs and personal histories
- Be careful of writers who claim to be writing facts, but are talented at painting a narrative (Malcolm Gladwell, Thomas Friedman)
- We can reduce the power of narrative in our own lives by keeping journals
- Whenever you're about to make a risky or uncertain decision, write down exactly why you're making that decision
- When the decision eventually succeeds or fails, you can then look back at that document and evaluate your decision-making rather than coming up with a convenient narrative that explains why success or failure was inevitable
- When searching for truth, favor experimentation over storytelling
- Some people just don't have visual imaginations
- Assumed that when other people were talking about visualizing objects, they were speaking metaphorically
- Got so good at talking about mental experiences as if they were visual that people with visual imaginations thought that they were having visual experiences
- Only when Galton actually surveyed people did we find out that there is in fact a broad variation in people's ability to form mental imagery
- Some people don't have the ability to smell (anosmia)
- Can go for years without realizing that they don't have the ability
- Often realize that they're not able to smell only when they're asked specifically about smells in great detail
- So what other "fundamental" experiences are people missing out on
- Asexuality - for most people, sex isn't gross or weird
- Emotional blunting - Scott may not have had emotions for about 5 years when he was on SSRIs
- Thought that everyone else was just being dramatic and overexuberant
- Even when he noticed himself not having emotions, he dismissed the fact
- Only learned later that emotional blunting is a common side effect of SSRIs
- Passion for music
- Scott doesn't really enjoy jazz - at best gets some kind of half-hearted feeling that he he could snap his fingers to the beat if he really tried
- Meanwhile his brother feel in love with jazz and is now a professional jazz musician
- The narrative fallacy is our tendency to turn everything into a story
- Unfortunately the real world has very few examples of linear chains of cause and effect
- Most outcomes are probabilistic, direct causation is rare, and events are complex and interrelated
- Our brains are engines designed to analyze the environment, pick out important parts and use those to extrapolate
- In the ancestral environment, simple linear extrapolation was "good enough"
- Unfortunately, the world is much more complex today
- The ability to simplify, cluster and chain ideas is what allows us to get away with a relatively small working memory and a slow neuron firing speed
- This narrative fallacy shows up in a number of lower level biases
- Availability heuristic - we make predictions based upon what we find easiest to remember - often this is what has the most compelling narrative attached to it
- Hindsight bias - past events "obviously" and "inevitably" cause future ones
- Consistency bias - we reinterpret past events and actions to be consistent with new information
- Confirmation bias - we only look for data to support the narrative conclusions we've already arrived at
- That said, we need narrative in order to have a single coherent self
- When people have damage to the frontal lobe and lose the ability to process higher-order input, they lose the ability to organize their lives and actions
- In the extreme case, they do not speak unless spoken to and do not move unless very hungry
- People with damage to other regions of the brain lose specific abilities, but remain the same person otherwise
- People who lose the ability to construct narratives lose their selves
- In the other extreme, narcissists over narrativize their lives, and make everything about themselves
- So what should we do about this?
- Make conjectures and run experiments
- Force beliefs to be falsifiable
- Make beliefs pay rent in anticipated experiences
