The Drunkard's Walk: How Randomness Rules our Lives
Author: Leonard Mlodinow
ISBN: 978-0-307-27517-2
APA Style Citation
Mlodinow, L (2008). The Drunkard's Walk: How Randomness Rules our Lives. New York, NY: Random House.
Buy this book
https://www.amazon.com/dp/B001NXK1XO/ref=dp-kindle-redirect?_encoding=UTF8&btkr=1
Author: Leonard Mlodinow
ISBN: 978-0-307-27517-2
APA Style Citation
Mlodinow, L (2008). The Drunkard's Walk: How Randomness Rules our Lives. New York, NY: Random House.
Buy this book
https://www.amazon.com/dp/B001NXK1XO/ref=dp-kindle-redirect?_encoding=UTF8&btkr=1
| schema_burglar_homebuyer_.docx |
Book Description
In the 2018-2019 NBA season, LeBron James had a free throw percentage of 51%, just over his all-time career free throw percentage of 50.4%. Let's propose a scenario in which LeBron James had made the last ten free throws he shot. Many basketball fans would consider this a "streak," attributing this recent run of success to the playoffs, playing against a long-time rival, increased fan support, or LeBron's new off-season training regimen. Those who believe in this “streak” for whatever reason will be more likely to assume that James will make his next free throw, disregarding the evidence that roughly 50% of the time he will make the shot and 50% he will miss. The Drunkard's Walkis not about an inebriated stroll through the city, rather it is about how we try to make sense of random events in our lives by adding some meaning and perceived control to these happenings.
The author, Mlodinow recalls a story from famed psychologist Daniel Kahneman. Kahneman explains that he was working with air force flight instructors trying to demonstrate that reinforcement was a far more effective method of instruction than punishment in terms of the speed of learning. The instructors disagreed, explaining that when a student completed a great maneuver, the instructor would issue praise, but inevitably at the next lesson, the same maneuver would be less impressive. Likewise, they argued that if the pilot in training made a horrible move, they would receive some intense negative feedback, and the next time out, their performance was generally better. After much contemplation, Kahneman realized that this did not challenge the conventional wisdom, instead, the instructors were witnessing regression to the mean. Most of the students were quite good as pilots, but were still training and would occasionally make a silly mistake and sometimes make a brilliant move; most of the time, however, their flying was slightly above average. It was not the instructors' positive feedback that was decreasing the quality of the flying, but the brilliant move was likely a fluke as was the mistake, so regardless of the feedback they received, the students flying should return to their typical level of performance on the next flight.
Mlodinow also explains that no executive is worth 25 million dollars and that the profits of a company are likely the result of many random market fluctuations and most of a company’s success (or failure) is far outside of the control of the CEO. Mlodinow uses the example of studios that produce movies. Heads of the movie houses are just as likely to predict a blockbuster as they are to predict a failure, but we want to believe that they have some internal vision that allows them to “just know” a blockbuster when they see one. It is far more likely that most film`s success or failure cannot be predicted with any certainty until it is released.
Our need to feel control over our lives and the world around us leads us to the frequent misconception that we have control over circumstances that are independent random events. If we flip a coin 20 times and it comes up as heads the first 19 times, most people would bet that on the next toss the coin will land as tails. They forget that the 20thtoss is an independent event that is not related to the results of the earlier trials. Just as each of LeBron James's free throws is an independent entity. People who win the lottery believe that their selection of specific numbers such as their family`s birthdays were the reason they won, not just a stroke of luck. Speaking of a stroke of luck, in large state lotteries the person is more likely to die in a car accident driving to buy the lottery ticket than they are to win the jackpot, but still, they persevere in the belief that they somehow influenced the outcome of this random event.
Some events however, we can predict. For example, we can predict the location of a prize behind one of three curtains (the two other curtains have goats behind them) with a higher than chance likelihood (66%) once one of the curtains has been eliminated. This is the famous Monty Hall Problem. We can predict the likelihood that a mother carrying fraternal twins will have a 75% chance that one of them is a girl (girl-boy), (boy-boy) (girl-girl), (boy-girl). Insurance companies can predict the risk of a driver based on their demographics, and the National Highway System can predict the number of accidents that will occur on a given year within a few hundred accidents. Demographers can produce relatively accurate predictions regarding the number of people who will move into or out of a state in a given year or two. To make these accurate predictions, we must overcome our expectations of past events and focus only on the situation in the present. Schemas are cognitive models we hold based on previous experiences, which then influence our perception in other situations. If I believe that men are safer drivers than women because the men in my household received fewer tickets than women or that, even if you present evidence to the contrary, I will have a difficult time accepting this information. Schemas can create selective attention allowing us to only focus on certain aspects of a situation while missing out on other important information.
Once we can identify the difference between random events and those that we can predict with some accuracy, we can stop developing algorithms for a drunkard's walk in which the drunk will eventually go somewhere, but there is no discernable pattern to his random ambling. Despite our inclinations to make meaning out of clouds and to think that a Volkswagen is smiling at us because of the orientation of the grill and headlights, we should let those things beyond our control be just that and seek to control what we can, namely things like studying to improve one's psychology grade in which great strides can be made with the proper efforts.
Other Related Resources
Scientific America: How Randomness Rules our World, and How Why We Cannot See It
https://www.scientificamerican.com/article/how-randomness-rules-our-world/
New York Post: "The Drunkard's Walk"
https://nypost.com/2008/05/18/the-drunkards-walk/
The Monty Hall Problem
https://www.montyhallproblem.com
Mathhacks: Simple explanations of some of the statistical examples from Drunkard's Walk
https://medium.com/i-math/the-drunkards-walk-explained-48a0205d304
Examples of people trying to plot the Drunkard's Walk using an algorithm
https://www.google.com/search?sa=X&q=drunkard+walk+algorithm&tbm=isch&source=univ&safe=active&ved=2ahUKEwiBlcORpv_hAhUFdJoKHY6rB9sQsAR6BAgJEAE&biw=1252&bih=814#imgrc=BOlvmEepa5hGQM:
Microsoft Research Talk with Author
https://www.youtube.com/watch?v=KRLwyq5dK48
Psychological Figures and Concepts
Aristotle
Charles Darwin
Rene Descartes
Francis Galton
William James
Daniel Kahneman
David Rosenhan
B.F. Skinner
Amos Tversky
amygdala
availability heuristic
behavioral modification
cerebral cortex
confirmation bias
conjunction fallacy
correlation coefficient
fMRI
gambler's fallacy
game theory
hindsight bias
inter-rater reliability
intuition
Monty Hall problem
normal distribution
outliers
paranoid delusion
perceptual sets
random sampling
regression towards the mean
representativeness heuristic
Rorschach inkblot
schema
schizophrenia
split-brain
standard deviation
statistical significance
stereotype
variance
In the 2018-2019 NBA season, LeBron James had a free throw percentage of 51%, just over his all-time career free throw percentage of 50.4%. Let's propose a scenario in which LeBron James had made the last ten free throws he shot. Many basketball fans would consider this a "streak," attributing this recent run of success to the playoffs, playing against a long-time rival, increased fan support, or LeBron's new off-season training regimen. Those who believe in this “streak” for whatever reason will be more likely to assume that James will make his next free throw, disregarding the evidence that roughly 50% of the time he will make the shot and 50% he will miss. The Drunkard's Walkis not about an inebriated stroll through the city, rather it is about how we try to make sense of random events in our lives by adding some meaning and perceived control to these happenings.
The author, Mlodinow recalls a story from famed psychologist Daniel Kahneman. Kahneman explains that he was working with air force flight instructors trying to demonstrate that reinforcement was a far more effective method of instruction than punishment in terms of the speed of learning. The instructors disagreed, explaining that when a student completed a great maneuver, the instructor would issue praise, but inevitably at the next lesson, the same maneuver would be less impressive. Likewise, they argued that if the pilot in training made a horrible move, they would receive some intense negative feedback, and the next time out, their performance was generally better. After much contemplation, Kahneman realized that this did not challenge the conventional wisdom, instead, the instructors were witnessing regression to the mean. Most of the students were quite good as pilots, but were still training and would occasionally make a silly mistake and sometimes make a brilliant move; most of the time, however, their flying was slightly above average. It was not the instructors' positive feedback that was decreasing the quality of the flying, but the brilliant move was likely a fluke as was the mistake, so regardless of the feedback they received, the students flying should return to their typical level of performance on the next flight.
Mlodinow also explains that no executive is worth 25 million dollars and that the profits of a company are likely the result of many random market fluctuations and most of a company’s success (or failure) is far outside of the control of the CEO. Mlodinow uses the example of studios that produce movies. Heads of the movie houses are just as likely to predict a blockbuster as they are to predict a failure, but we want to believe that they have some internal vision that allows them to “just know” a blockbuster when they see one. It is far more likely that most film`s success or failure cannot be predicted with any certainty until it is released.
Our need to feel control over our lives and the world around us leads us to the frequent misconception that we have control over circumstances that are independent random events. If we flip a coin 20 times and it comes up as heads the first 19 times, most people would bet that on the next toss the coin will land as tails. They forget that the 20thtoss is an independent event that is not related to the results of the earlier trials. Just as each of LeBron James's free throws is an independent entity. People who win the lottery believe that their selection of specific numbers such as their family`s birthdays were the reason they won, not just a stroke of luck. Speaking of a stroke of luck, in large state lotteries the person is more likely to die in a car accident driving to buy the lottery ticket than they are to win the jackpot, but still, they persevere in the belief that they somehow influenced the outcome of this random event.
Some events however, we can predict. For example, we can predict the location of a prize behind one of three curtains (the two other curtains have goats behind them) with a higher than chance likelihood (66%) once one of the curtains has been eliminated. This is the famous Monty Hall Problem. We can predict the likelihood that a mother carrying fraternal twins will have a 75% chance that one of them is a girl (girl-boy), (boy-boy) (girl-girl), (boy-girl). Insurance companies can predict the risk of a driver based on their demographics, and the National Highway System can predict the number of accidents that will occur on a given year within a few hundred accidents. Demographers can produce relatively accurate predictions regarding the number of people who will move into or out of a state in a given year or two. To make these accurate predictions, we must overcome our expectations of past events and focus only on the situation in the present. Schemas are cognitive models we hold based on previous experiences, which then influence our perception in other situations. If I believe that men are safer drivers than women because the men in my household received fewer tickets than women or that, even if you present evidence to the contrary, I will have a difficult time accepting this information. Schemas can create selective attention allowing us to only focus on certain aspects of a situation while missing out on other important information.
Once we can identify the difference between random events and those that we can predict with some accuracy, we can stop developing algorithms for a drunkard's walk in which the drunk will eventually go somewhere, but there is no discernable pattern to his random ambling. Despite our inclinations to make meaning out of clouds and to think that a Volkswagen is smiling at us because of the orientation of the grill and headlights, we should let those things beyond our control be just that and seek to control what we can, namely things like studying to improve one's psychology grade in which great strides can be made with the proper efforts.
Other Related Resources
Scientific America: How Randomness Rules our World, and How Why We Cannot See It
https://www.scientificamerican.com/article/how-randomness-rules-our-world/
New York Post: "The Drunkard's Walk"
https://nypost.com/2008/05/18/the-drunkards-walk/
The Monty Hall Problem
https://www.montyhallproblem.com
Mathhacks: Simple explanations of some of the statistical examples from Drunkard's Walk
https://medium.com/i-math/the-drunkards-walk-explained-48a0205d304
Examples of people trying to plot the Drunkard's Walk using an algorithm
https://www.google.com/search?sa=X&q=drunkard+walk+algorithm&tbm=isch&source=univ&safe=active&ved=2ahUKEwiBlcORpv_hAhUFdJoKHY6rB9sQsAR6BAgJEAE&biw=1252&bih=814#imgrc=BOlvmEepa5hGQM:
Microsoft Research Talk with Author
https://www.youtube.com/watch?v=KRLwyq5dK48
Psychological Figures and Concepts
Aristotle
Charles Darwin
Rene Descartes
Francis Galton
William James
Daniel Kahneman
David Rosenhan
B.F. Skinner
Amos Tversky
amygdala
availability heuristic
behavioral modification
cerebral cortex
confirmation bias
conjunction fallacy
correlation coefficient
fMRI
gambler's fallacy
game theory
hindsight bias
inter-rater reliability
intuition
Monty Hall problem
normal distribution
outliers
paranoid delusion
perceptual sets
random sampling
regression towards the mean
representativeness heuristic
Rorschach inkblot
schema
schizophrenia
split-brain
standard deviation
statistical significance
stereotype
variance
RSS Feed