Lecture 9 - Decision Making & Problem Solving

what is the background of JDM?

• stands for Judgement and Decision Making (JDM)

  • Judgement: The mental process through which people draw conclusions from the evidence they counter

  • Behaviorism: Researchers were only interested in external behaviors, stimulus-response associations

  • Modern Cognitive Psychology: interested in mental concepts like desires and goals

what are three types of theories talked about in lecture?

• normative

  • the supposedly optimal decision based on a set of principles

  • ex. expected utility

• descriptive

  • how people actually make decisions

  • ex. the use of heuristics (mental shortcuts)

• prescriptive

  • how to help people make better decisions

  • ex. after understanding common biases, providing tools such as decision trees to inform decisions

what are the two types of systems in the dual process model? what are the limitation?

• System 1

  • Intuitive, Automatic, Immediate

  • Rely on heuristics

• System 2

  • Analytical, Controlled

  • Cognitively demanding

  • more likely to be correct

• Limitations

  • Distinction is not always clear-cut

  • Lack of biological basis

  • More descriptive in nature, can analyze why someone may have made a decision but have limited predictive power

• what is a heuristic?

  • what do they rely on?

• considered mental shortcuts

  • constructed based on prior experience

  • save time and energy

  • produce good enough/correct answers most of the time, but not guaranteed to be correct

• rely on attribute substitution

  • using one attribute (e.g., fluency or similarity) to make a judgment about another attribute (e.g., frequency or probability)

what are some common heuristics?

• availability: how easily something comes to mind

  • fluency effects

• representativeness: what you believe is typical/representative of a situation

  • base-rate neglect

  • conjunction fallacy

  • gambler’s fallacy

• anchoring and adjustment: base judgement on the first number presented and adjusting from there

• what is the availability heuristic?

  • how have we studied its fluency effects?

• Definition

  • making a judgement based on how easily something can be recalled, the easier it is for something to comes to mind, the more frequent / likely / significant it is assumed to be

• Course Evaluation Study

  • Two Groups:

    • relatively easy: list 2 ways in which this course could be improved

    • relatively difficult: list 10 ways in which this course could be improved

  • Groups then asked to list two favorite aspects of this course and rate it on a scale 1-7

  • Results: Paradoxically, course evaluations can be improved by soliciting more critical

    • 2 critical comments: 4.92

    • 10 critical comments: 5.52

  • Conclusion: A more positive judgement of the course is assumed because of the difficulty of coming up with 10 different things that could be improved (availability heuristic)

• Ratings of Assertiveness as a Function of Valence and Number of Recalled Behaviors

  • Two Groups:

    • List 6 (easy) or 12 (difficult) examples in which you . . .

      • “behaved very assertively and felt at ease”

      • “behaved unassertively and felt insecure”

  • Conclusion: When it’s easy to think of examples of being assertive, or when it’s hard to think of examples of being unassertive, people perceive themselves as more assertive

  • ease of recall affects judgment of self

what are two more studies discussed in lecture that investigated availability heuristic and fluency effects?

• Stock Study

  • Stocks w/ more pronounceable names (e.g., KAR) outperformed unpronounceable names (e.g., RDO)

  • significant difference on the first day, then the difference diminished potentially due to investors having additional diagnostic info on performance

• Food Additive Hazard Study

  • Groups asked to rate hazard posed by the following food additives

  • Results: The harder-to-pronounce ingredients received higher hazardousness ratings than the easier-to-pronounce ones

• Theme Park Study

  • Harder-to-pronounce names were perceived as riskier, regardless of whether it was desirable or undesirable risk

• what is the representativeness heuristic?

  • what is the assumption of homogeneity?

• Representativeness heuristic: the assumption that resemblance to the prototype reflect probability

  • assumption of homogeneity: an expectation that each individual is representative of the category overall

  • likelihood of category membership judged by resemblance

• what is base rate and base rate neglect?

  • what example was shown in class?

• base rate: actual rate of how common something is

• base rate neglect: the tendency to ignore the “prior probability” of an event

• BRN Example

  • Base rate only: There is a group of 70 lawyers and 30 engineers, what is their profession likely to be?

  • Base rate + diagnostic info: There are 70 lawyers and 30 engineers. This particular person likes carpentry, sailing, math puzzles; dislikes politics” What is their profession likely to be?

    • example of base rate neglect due to judgement being based on diagnostic info (ignore 70% chance)

      • this is known as Attribute Substitution: typicality substituted for probability

what is the conjunction fallacy?

• conjunction fallacy: an inference that the set of two or more conclusions is more likely than any single member of that set

• what is the gambler’s fallacy?

  • what are some examples of this?

• Gambler’s fallacy: failure to consider the independence of probabilistic events

  • chance viewed as “self-correcting” process

• Examples

  • After a long run on the roulette wheel, people think that black is “due to happen”

  • After having five girls, assuming the next child to be a boy

what is anchoring and adjustment?

• Anchoring and adjustment: the tendency to anchor estimate on first salient number then adjust up or down from there

In decision making, what are the two kinds of information one must consider to get the best outcome?

• Two types of Information

  • Utility: the importance/value of each outcome

  • Probability: how likely each outcome

• what is utility theory?

  • what are some problems with utility theory? what are some examples of these potential problems?

• utility theory:

  • assumes that humans are rational actions who choose the option that provides the most utility

  • utility can be thought of as the “expected value” of a choice

  • expected value = probability of an outcome x value of the outcome

• Problems

  • people have clear preferences even when the options have the same expected value

  • ex. Would you prefer to definitely gain $50 or have 50/50 chance to win $100

    • EV is +$50 for both, but people prefer definitely gaining $50

    • in this example, they are risk-averse

  • ex. Would you prefer to definitely lose $50 or have a 50/50 chance to lose $100 or nothing

    • EV is -$50 for both, but people prefer the 50/50 chance to lose $100

    • in this example, they are risk-seeking

• what is prospect theory?

  • how is it different from utility theory?

• Prospect theory accounts for the difference in psychological value between gains and losses

  • people hate losing a lot more than they like winning

• Prospect theory: The marginal impact of a change in value diminished with the distance from a relevant reference point

• Utility depends on subjective utility and subjective probability

• what are framing effects?

  • how does this apply to gains vs losses?

  • describe the disease problem. how does it apply?

  • describe the personal and impersonal version of the trolley problem. how does it apply?

• definition: a type of cognitive bias that describes how people react differently to the same information depending on how it’s presented, whether positively or negatively

• gains vs losses

  • risk-averse for gains

  • risk-seeking for losses

• Disease Problem

  • Background: Presents a gains vs losses problem but framed within a scenario of saving lives and sacrificing lives.

  • Results:

    • Prefer sure gain when positively-framed (guarantee saving some)

    • Prefer risky loss when negatively-framed (avoid loss of life)

• Trolley Problem

  • Impersonal: Pull switch a save more lives.

  • Personal: Push someone off and save more lives

  • Results: People are more willing to pull the switch than to push the person, even though the outcome would be the same

• what is the difference between relative risk reduction & absolute risk reduction?

  • what are the results when measuring people’s responses to either one?

• relative risk reduction (probability)

  • ex. If you have this test every 2 years, it will reduce your chance of dying from this cancer by around one third over the next 10 years

• absolute risk reduction (frequency)

  • ex. If you have this test every 2 years, it will reduce your chance of dying from this cancer from around 3 in 1000 to 2 in 1000 over the next 10 years.

• Results

  • People are more likely to accept the test in the probability condition

• how does having more options affect our ability to make decisions?

  • what study supports the answer to the question above?

• Having more options may lead to NOT making a decision

• Physician Decision Study

  • two groups

    • Group A (two decisions)

      • physicians less likely to refer patient to surgery

    • Group B (three decisions)

      • physicians more likely to refer patient to surgery

      • hypothesized that this result is because physicians could not decide between drug a or drug b which led them to default to surgery

• what is a problem space?

  • what are the two ways we can navigate a problem space?

• problem space: total set of possible moves within the constraints of the problem

• Two Ways to Navigate

  • Algorithm

    • A procedure that inspects every possible move in the space by applying operations over and over again until goal state is reached

    • slow but guaranteed to produce a correct solution

  • Heuristic

    • informal, intuitive, sometimes effective

what are the two types of problem-solving heuristics discussed in lecture?

• hill-climbing strategy

  • at each step in solving a problem, choose the option that moves you in the direction of your goal

  • note: does not work for problems that require you to move away from the goal for some steps

• means-end analysis

  • identify the difference between the current state and the goal state

  • choose an action that reduces the difference

  • usually divide the problem into smaller problems

• does experience always help when it comes to problem solving?

  • how does the concept of function fixedness help us understand the answer to the above question?

• Not always. Examples such as the “candle problems” and the “two-string problem” prove that this is not the case.

• Functional fixedness:

  • when the usual function of an object is emphasized, it will be far more difficult for a person to use that object in a novel manner

• Candle Problem

  • Given a matchbox, and a tac box with tacs in it one is asked to make a holder for a lit candle. Due to functional fixedness on the tacs and the matches one might not think to use the box of the tacs. They might fixate on the fact that its just a container rather than a possible solution to the problem.

• Two String Problem.

  • dont care

radiation problem

fortress problem

novice expert question