Psych 355

Front 1

8.1
Explain decision making under uncertainty. Give an example. What is the rational (normative) approach to making a decision?

Back 1

normative approach-find the alternative with the highest possible utility, the prob of the alternative occuring must be taken into account.
probability-measurement of hte degree of uncertainty about an outcome, when there is more than 1 possible outcome
outcome is usually expressed as a # between 0 (will not occur) and 1 (completely certain will occur).
Alternatives-items which to choose from
further evidence-before calculating if exact probabilities are not present, must be taken into account
evidence-making decision about the probable value would be if the #'s are not present
expected utility-(measure of value to decider) must be measured
decider trying to maximize expected utility of the alternative(s) in their decision
taking all these variables into account, rational approach to making the decision is to X the expected utility (value) of each outcome by the probability of that outcome and sum the totals across the outcomes.
choose the alternative that has the highest possible utility (now that prob has been taken into account)
ex-uncertain whether/not to play lottery look @ alternatives and prob:
Alt A buy ticket for $1: win-50% chance of $50. Lose-50% chance losing $1
win=25.50, love -.50=$24
alt 2: dont play and keep $1
based on what the decider thinks is valuable (keeping $1 or risking it and possibly winning $50) decision is made

Front 2

8.2 What is MAUT (be sure to explain utility, alternatives, dimensions, weighting, tradeoffs)? what is the method designed to accomplish?

Back 2

MAUT-normative theory to make a decision when there is no prob present and there are several alternatives, each of which rated differently on their attributes (dimensions or characteristics upon which the decider evaluates each alternative)
4 steps:
1. Weight the dimensions-decide how important each individual dimension is to you. ex. students asked to write several dimensions that contribute to their decision about what to major in. then weight (application to career was rated highest, rep on campus lowest)
2. rate the utility of each alternative on each dimension. utility is subjective measure of value to the decision-maker (try to maximize) alternatives are choices to choose between (psych, chem) there can be any # of dimensions, and utility of each alt. can be varied (psych high utility for applicability to career, but low utility for reputation on campus)
3. combine all the weights and ratings for each alternative (where space between rows is rep. of weight the dimensions have for each) X weights by the utility rating to get the overall rating for each alternatives
4. choose highest overall utility of the alternatives (psych scored higher in more heavily rated utility measures, helps it to receive highest score)
purpose is to use this strategy to find the alt that has the best value.
prerequisite: have to be willing to make compromises/tradeoffs-where 1 dimension allowed to compensate for another alternative (can't be 1 dimension that is end-all deciding factor, and other don't matter)
ex. more appropriate for career (psych major) must compensate for low rating on reputation on campus category

Front 3

8.3
Describe 2 "errors" (from normative theory) that ppl make inevaluating multiple dimensions to make choices among several alternatives (define weights and eliminating by aspects) Explain why each is an error from normative approach.

Back 3

fail to take combined weights and ratings of the alternatives into account or simply eliminate choices by aspects.
Gelloti-students complete 2 worksheets 1: write out simensions and weighted these and utility of each alternative (in college major)
2: showed same alternatives as WS 1, asked to make "global: rating of each
when compared, WS showed Ss rating on top dimensions corresponded perfectly w/ the highest weighted alternative. ERROR: don't take into account all weights, b/c limited capacity of working memory prevent from assessing each alt weight and comparing
error: eliminate alternatives that fall below certain threshold they deem to be the max. done when given mult. alternatives (4 diff apartments) when deciding on which was best utility and value, simply eliminate alternatives that do not fit certain criteria (one that is over $1,000)
may prevent form picking the best alt., b/c never get chance to see how the eliminated alt weights on other dimensions.
errors fail to utilize the value and utility of each alternative (key to normative approach) also do not meet the prereq for normative theory-being willing to make tradeoffs btw dimensions (one dimension is end-all be-all for decision)

Front 4

8.4
What are framing effects (discuss current state, losses vs gains, the issue of risk) Give an example not used in lecture/test. be clear about what aspect of framing effect your ex. illustrates

Back 4

framing effects-when decision influenced either by the background context of the choice (current state) and/or the way in which the question is worded (losses vs. gains)
current state-influences the decisions ppl make ex. someone paid $5 for cheeseburger, got it and it was moldy and had to throw it away and see it as a loss of $5 and not buy another. if got to restaurant took out their $5 to find it had been ripped/gone through dryer likely to see the loss as irrelevant to the situation of buying a cheeseburger (either way losing $5, not eating the cheeseburger)
wording-people gen choose alternatives that are framed as risk-avoidant for gains and risk-seeking for losses. deep structure of these problems are the same (statistically similar) but the surface structure (whether framed as gain or loss) is different. when framed in terms of a loss, ppl take a chance (trying to minimize negative outcomes)
if framed in a gain, they are less likely to want to take a risk

Front 5

8.5
Describe the representativeness heuristic. discuss how base rate neglect and the conjunction fallacy are related to representativeness heuristic. give an example of one of these (not use in lecture/text)

Back 5

rep heuristic-when ppl assume commonality btw objects of similar appearance, when their is no real evidence pointing to this (statistics show the option is less probable)
both base rate neglect and the conjunction fallacy are errors ppl have from using rep heuristic
base rate fallacy-when judge an unknown item/group member to be part of the base rate, or how often the item occurs in the population ex. told there are 90% college students in room, 10% politicians description: person doesn't like reading, spends time day dreaming, loves to surf a) surf bum b) student c) politician ppl choose surf bum b/c ase their decision on rep info, not how often that item occurs in the population (0)
conjunction fallcay-ppl judge prob of the conjunction of 2 events to be greater than the prob on a constituent event-even when the chance of 2 items being in conjunction w/ each other (statistically) is not higher than either single event
ppl ignore the stats and judge them to be more likely b/c of previous info received that would seem to create a correlation between the 2 events

Front 6

8.6
What is an availability heuristic? Discuss the issues of recency and familiarity. Give an example of availability heuristic due to one of these.

Back 6

Availability heuristic: ppl estimate the req/prob in terms of how easy it is to think of relative examples of something. more recent the info/item is in the memory, recall is easier (memory fades with time)
familiarity when ppl judge an accurance more likely the more familiar the item is. ex. familiarity as an availability heuristic would be for someone living close to a college campus to judge that most young 20 year olds drink a lot and are good at beer pong, because they see/are familiar w/ many 20 year olds drinking a lot and playing beer pong when not all 20 year olds drink a lot and play a lot of beer pong

Front 7

8.7
What is anchoring and adjustment (explain the psychological processes and how it effects numeric estimates) give an example. What are confidence intervals and how could anchoring effect their estimation?

Back 7

anchoring and adjustment-person begins w/ an approximation about something (anchor) and then makes adjustments to that number on the basis of additional info
emphasizes top-down processing: info gathered and knowledge used at the first stage (anchoring) affects how the later data is processed. ppl tend to endorse their current hypothsis, rather than (adjust) question them.
ex Ss given 1. 8X7X6X5..X1 or 2. 1X2X3X4...X8 Ss gave much larger number (2250) thatn 2 (512) Both groups anchored their knowledge on the initial info and did not adjust to the later digits by looking at all the later digits (answer should be 40,000)
confidence intervals are ranges w/in which ppl expect a number to fall a certain % of time ex. estimate 98% confidence any 1 college will have 3,000-4,000 ppl
most of the time, these confidence levels are incorrect (60%)
ppl make an anchor based on beliefs they have and then fail to take more info into account to make a correct adjustment.

Front 8

8.8
How does real life decision-making differ from the assumptions of normative models (name 4 and be explicit about what those normative assumptions are-mention/define situation assessment) How do Naturalistic theories of decision making say experts make decisions?

Back 8

In real life, ppl do not make decisions based on a rational (normative) approach. normative model assumes to a perfect decision is required, alternatives are dimensions are static, there is plenty of time to come up w/ alternatives, and all the info is available simultaneously.
Oransanot and connolly use fire chief making decision about event to illustrate how real life decisions contradict assumptions of the normative model:
rare to have a perfect decision-would take a long time, and ppl must genarlly make decisions quickly (in fire man case, before ppl die)
things in real life are not static, they are dynamic. alternatives and dimensions included in the decision shift and change while the decision is being made (had to move from trying to save building, to trying to save fire men and families)
sometimes there is very little time to make important decisions (must decide quickly)
similarly, all the info relating to the decision is not all present at one time. certain parts may be known, sometimes this is crucial info, sometimes not.
in real life, ppl must make a quick situation assessment where the figure out what alternatives and dimensions are as they gather info.
real life events often to not allow for ppl to follow the normative approach in their decision making.
in fire example, chief made a quick decision w/o taking much time to evaluate other evidence/alternatives. in cases with experts, their decisions are affected by recognition primed decision making. this is when the expert recognizes how the situation matches a previous one and then takes the previous solution to be used (w/o taking the time to consider alternatives)

Front 9

8.9
Describe over-confidence in decision-making and the proposed reasons for it. Use an example of "planning fallacy" from your own experience. Be sure to explain why this is an exaple of decision-making. Which of the 5 reasons might have been factors in the situations you describE?

Back 9

over confidence: ppl's confidence judgments are higher than they should be, based on their actual performance on the task
affects the outcomes of the decisions ppl make (may over estimate future performance, possible hypothesis, etc.)
overconfidence plays role in planning fallacy where ppl underestimate the amount of time/money required to complete a project and estimate the task will the relatively easy to complete.
5 reasons ppl have over confidence: 1. ppl unaware their knowledge based on uncertain assumptions and info is from unreliable sources. 2. examples confirming our hypothesis are available, while we resist counterexamples 3. ppl have difficulty recalling other possible hypothesis (depends on memory) 4. even if ppl manage to recall other possible hypothesis, they are not treated seriously 5. when ppl make decision as a group, they sometimes engage in group think.
i committed a planning fallacy when writing a final paper for my Psych 306 class. underestimated the time it would take me to finish and thus, did not have enough time to do a through job on the paper. b/c i did not correctly estimate the time and effort i would have to put in it, I got a lower grade than I was expecting. I followed reason one, where my knowledge about how the paper would go was based on the assumption the paper would be easy and i was a good student (thus could finish it well) I also confirmed this hypothesis by looking at other times I had written papers in the same amount of time and they had turned out well.