Columbia Atheists Message Board › New Meetup: Saturday Morning Science
Announcing a new Meetup for Columbia Atheists!
What: Saturday Morning Science
When: Saturday, February 6, 2010 10:00 PM
Monsanto auditorium of the Life Sciences Center
MU Campus Rollins & College
Columbia, MO 65201
How game theory explains stupid behavior
When markets collapse and people who should cooperate turn on each other instead, rational people are acting stupidly. This talk will explain how game theorists study some of our most irrational and stupid behavior.
Saturday mornings from 10:30-11:30 at the Monsanto auditorium of the Life Sciences Center. Come earlier for coffee.
Learn more here:
That would have been kind of a late morning...
Yeah, only noticed that after posting.
AM not PM.
We suck at following game theory advice
Game theory = mathematical theory describing how individuals choose actions when the outcome depends on how others choose
- Stock market
Lots of second guessing, iterative thinking
Stock market is like a beauty contest where everyone tries to pick who everyone else thinks is prettiest
Player's actions in a situation of risk is her strategy
- Suppose you choose a strategy
- Imagine everyone else picks that strategy too
- Would you now want to switch strategies?
-- If so, strategy is unstable
-- If stick with same strategy, it is stable = Nash equilibrium
- Goal = play stable strategies
Rock paper scissors
- There's no stable strategy other than choosing randomly
- Over time, you'll win 1/3 of the time
Nash's golden rule
- Play the strategy you'd want to play even if everyone else played it too
Hedge funds violate Nash's golden rule
- Black-Scholes Theorem gives value of options
-- Only works if you're the only one using the formula
-- Fund using the formula did great until formula went public, then it tanked.
- They know that we know that they know that we know that they know...
- Can only comfortably reason about 2-3 iterations of this kind of problem
- Make decisions according to Nash's golden rule
- But be realistic about who you're playing against
-- Not everyone is an idiot
-- Not everyone is brilliant
Animals play game theory too
- Generally seem to naturally choose stable strategies
- Spiders looking for a nest will usually walk away if it's occupied rather than trying to evict the current spider
-- Don't start a war, just walk away and find another spot
- Young horses use chewing behavior to indicate "I'm just a baby". Older horse will back off picking on them.
People have evolved ways of dealing with others
- Speak the same language
- Have common ways of communicating
Why aren't we better at dealing with others?
- Limits to brain function?
- Something more subtle?
Most of our behavior is instinctive. Even what we learn, we're prone to learning.
- Innately good at learning the 'typical human stuff'
- If we were raised by honey bees, we'd never learn all the normal honey bee stuff.
Many cases where there are multiple Nash equilibria
- Usually pick one strategy then it becomes entrenched & becomes the single best strategy
- No inherent benefit to driving on left or right side of the road
- Once driving on the right becomes most common/standard, driving on the right is now the best strategy
Cultures become entrenched from the bottom up. Governments react to that culture. Thus, if you get rid of the government, the culture is still entrenched so you haven't changed too much.
Leaders & soldiers & citizens etc aren't playing the same 'game'. Each group has their own point of view and cost/benefit analyses that often take the other 'games' into account to some greater or lesser extent
Game theory is ideal rationality but in real life, there are gaps between "what we should do" and "what we actually do"
- We do learn and get closer to Nash equilibrium
- However, by changing your behavior to adapt to other people, you are still out of date because the other people have probably changed their behavior and the 'standard' has now changed. You're always playing catch up.
Economy in recent past incentivizes destabilizing the market. Need to find a way to make incentives for a stable economy.
Game theory says spiders should start attacking other spiders rather than backing off if hiding places are rare.
- Risk of injury from fight vs. risk of not being able to find another home.
Wow Carla, You are awesome at taking notes.