Decisions

Decision making fascinates me.  In fact one of the reasons I entered into teaching and math specifically was to help the next generation with the skills to make better decisions.   I viewed math as a tool that would develop their cognitive wherewithal.  However decision making is more complicated than that.

Source of trouble Brief Summary, aspects, characteristics Evidence, Citations, Examples How to mitigate, minimize, address
Memory We base a lot of our decisions on Memory.  What we remember we did before, how we experienced something, or how we felt about what we experienced.  The problem with this are many:

  • Memory is reconstructed when we think about a thing. This means that we don’t remember the actual event but our perception of it.
  • Our memory is selective.  We do not remember all the details, but select only certain details that some executive function determines as relevant to he experience.
  • Our experience is subject to suggestion and therefore we are actually mis-remembering. Rather, we may be remember an implanted memory or someone else’s interpretation, or our changed interpretation of an event.
  • Hindsight is 20-20 we knew it all along…
  • We tend to keep a mental account of events to guide our decisions…and this may be flawed.
  Reconstructed Memory

  How to address this in the classroom:

  1. Have them construct notes first without copying to force the interpretation of what happened on the board.  For example: After students gain some experience with completing the square, have them outline the process in their own language.
  2. Write notes clearly on the board. 
  3. Pay students for taking notes.
  4. Always leave on good terms.  Handshakes, well-wishes, or puzzling thought.  Some thing that is their last “memory” of the class for the day.
Two systems We have two systems that process information.

  • System 1 is primal immediate, intuitive, emotive, reactive.  “It operates automatically and quickly, with little or no effort and no sense of voluntary control”
  • System 2 is thoughtful, effortful.  It allocates attention to the effortful mental activities that demand it, including complex computations.  The operation of system 2 are often associated with the subjective experience of agency, choice, and concentration.”
  • Kahneman, Daniel Thinking Fast and Slow. (p.20)
  • System 1 is subject to cognitive illusions (ibid p.28)
  • Priming affects our association (If we are hungry we see SO_P soup rather than soap).
  • We have a bias towards cognitive ease. The less we work in reading short sentences, easy to remember acronyms, fonts, and layouts, the more likely we are are to believe it or remember it. (ibid p.59-70)
  • Framing effects can trick system 1: Different ways of presenting the same information evoke different emotional responses (p. 88)
  • Overconfidence in our beliefs and suppresses doubt and ambiguity (p.88)
  • System 1 Links cognitive ease, pleasurable effects, and reduced vigilance (Ibid, p.104)
  • “Choices are not reality-bound because system 1 is not reality bound” (ibid p.364)

  1. Would you accept a gamble that offers 10% chance to win $95 and a 90% chance to lose $5?
  2. Would you pay $5 to participate in a lottery that offers a 10% chance to win $100 and a 90% chance to win nothing?

 (Although 1 and 2 have a same outcome, yet people favor 1) (Ibid p.364) the problem is framed differently.

 Next How to
Halo Effect When we are taken by someone or something, our perception of evidence surrounding that person is clouded.

  • Consider: “Hitler likes small puppies.” How could someone so terrible be associated with something so cute.
  • Or after a lifetime of positive impact and contribution to the community, person x is found to have committed an act of atrocity.  All their good works are erased in a moment.

We are quick to overlook fact or weight them inconsistently depending on whether we like or don’t like the person.

 

 Next How to
Emotions The emotional impact we feel affects our judgement.

  • When angry, overjoyed, stressed, depressed, we make decisions we might not ordinary make.
  • Also we may not interpret our emotions properly.  Males confused anxiety with sexual arousal.
  • Emotions can frequently override sensory input adversely affecting decision making capabilities.
  • Intensity and inescapability trigger psychological immune systems.
  • Regret is a strong emotion tied to among other things our recollections (which may be faulty), and our predictions (which also may be faulty).
Heuristics Heuristics allow us to quickly gauge or size up a situation without exact measurement.  “If I had to ball park it….”

  • When faced with a hard question, we frequently answer an easier and similar question.
  • Our heuristic may not be calibrated properly.
  • Kahneman, Daniel Thinking Fast and Slow. (p.97-105)
  • Think of the politician when asked question x answers question y (perhaps because of a prepared response, but also perhaps they did not have an answer and could not come up with one on the spot).
 Next How to
Small Numbers Large Samples are more precise than small samples.  Small samples yield extreme results more often than large samples do.  This misleads our read on patterns and may spark us to “misread” a situation. Next How to
The Anchor effect Anchors prime our perspective and change our intuitive and perception of the circumstance.”Starting points affect versions of the ending point.  If a person is fed certain input, the outcome is shaped by this.”We tend towards adjusting towards that anchor.
  • Kahneman, Daniel Thinking Fast and Slow. (p.118-128)
  • How many countries in the UN are African experiment.  Participants guess higher/lower depending on phone number they were asked to think about.
  • In negotiations, a “low” or “high” offer will pull negotiations in either direction…but not far from the initial offer.
  • In initial guess to unknown information, the guess is close to a number or information the participant was prompted to think about.
  • Charities anchor thought about giving with higher numbers on the slip.
  • Judges are more likely to sentence differently when anchored to different levels.
  • Marketing uses this technique to affect our perception of purchase price
  • Gilbert, Dan Stumbling on Happiness (part p.148)
 Next How to
Experience is subjective “The attentive person’s honest, real-time report is an imperfect approximation of her subjective experience, but it is the only game in town.”We frequently gauge our decisions by our happiness, perceived pleasure, or how we rank that experience of others….but it is different for everyone.The frequency and recency of our experience colors our interpretation and judgement.

When making a decision about something we’ve never done before we sometimes rely on the experience of others.  However their internal measures for the “goodness” of an experience may be radically different than our own.

Perception of the future and past is colored by our experience of today. (This affects our ability to make predictions)

Our experience also informs our ability to “Prefeel” events.   Which helps us decide on one course of action or another.

 Next How to
Context Context affects our perception of an event.People may be making a judgement or decision without proper context.People tend to self sort.

This means they tend to live in communities, read material, and surround themselves with people who don’t challenge their values and don’t supply a diversity of thought.The values ingrained in us shape our context.  Pacifists, religious zealots, or people from a culture a violence see their options differently and may come to different conclusions about what must be done.

Media and coverage affect are view of frequent or important events. They specifically affect the availability of information which affects are associative process (the frequency of event A, with the risks of event A, with the potential reward of event A). In a word they affect our base rate knowledge.
  • Gilbert, Dan Stumbling on Happiness (part p.83-104) Consider Adolf Fischer the “Haymarket Martyr” v. George Eastman.  Adolf … wrongly accussed was hung, yet stated it was the happiest day of his life while the wealthy and powerful George Eastman killed himself.
  • Bishop, Bill the Big Sort.
  • People overestimate Tornado and lightning as causes of death and underestimate asthma and pools as causes of death.  Why we see them more and our context for frequency calculations are off. Why we make bad decisions.
  • Paul Slovic, Dan Gilbert, Dan Kahneman speak to this in their work on availability of information and risk assessment. (Kahneman, p.147)
 Next How to
Calculation Errors People when making an assessment about the likelihood of an event or finding a percent of certainty.  They need to know two things: the Part and the Whole.  This ratio makes a fraction which can become a percent or probability.  However many factors interfere with this:

  • Finding the base rate can be tricky. Perhaps our ability to count or to acquire a sense of is skewed by any number of the factors listed here.
  • Counting is hard.  In many cases there are issues of samples, representative samples, or combinatorics that are simply beyond what we can quickly and without computing aids.
  • People misapply the rules of probability.  Conditional probability and others require different calculations than just counting.
  • People don’t understand the calculated result.
  • Consider the gambler’s dilemma. In effect the gambler fundamentally misunderstands the law of large numbers or is willing to forgo what the rules for other reasons.
  • The Linda Problem. Given a scenario people believe A is more likely than A and B. Where P(A) > P(A and B) as 1/x  > (1/x * 1/y)
 Next How to
Logical Fallacies Regression to the MeanCorrelation does not imply causation

Not all patterns continue or they could have multiple next steps

We tend to ignore absences when finding patterns.

 Next How to
Overconfidence We want to be right.  No one wants to go through life being wrong all the time.  We want to be right more often than not.  Needless to say this is magnified by the stakes involved.Illusions of skill: (Stock Traders)Illusions of pundits: Experts performed worse than if they would have if they had simply assigned equal probabilities to each of the potential outcomes…Those who know more forecast very slightly better than those who know less. But those with the most knowledge develops an enhanced illusion of their skill and becomes unrealistically overconfident.

Optimistic Bias causes to give overly robust estimates to their plan.  This also contributes to the Planning Fallacy.
Loss Aversion We would rather win less than loose more. Organisms that treat threats as more urgent than opportunities have a better chance to survive.  This causes people to make errors in value. Next How to
The Anchor effect Next summary Next Evidence Next How to
The Anchor effect Next summary Next Evidence Next How to

 

 

How do we combat poor decision making?

Strategy Brief Summary, aspects, characteristics Evidence, Citations, Examples
Mathematical Modeling Using a mathematical model or other algorithm frequently outperforms subjective heuristics and experts. Paul Meehls work Clinical versus Statistical PredictionKahneman, Daniel, Thinking Fast and Slow (p. 229-231)

 

Leave a Reply

Your email address will not be published. Required fields are marked *