College 3 Flashcards

You may prefer our related Brainscape-certified flashcards:
1
Q

Life is unpredictabe

A

We are dependent on estimates and rules of thumb.

Our intuition isn’t always right.

We don’t apply statistics & math correctly.

Life is like a box of chocolates; you never know which one you’re gonna get.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

When do we notice a decision the most?

A

When it goes wrong (e.g., a shipwreck or after a breakup you think “how did I now see this before”)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Biases used in decisions and judgments

A

Bias in finding information.
–> In line with theory (tunnel theory)

Bias in sample.
–> E.g., small size, selective sample (e.g., only friends)

Bias in base rates.

Bias in using or “averaging” the information.
–> You gather a lot of information but use statistics wrong

Biases in seeing co-variation.
–> Illusory correlation

Biases in ‘weighing’ the data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

The law of large numbers

A

The more you repeat an experiment, the more likely the outcome of this experiment is in line with the true expected value.

Outcomes become closer to the expected value with more trials.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Gambler’s fallacy

A

If a particular event occurs more frequently than normal during the past, it is less likely to happen in the future, even when events are statistically independent.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Regression effect

A

Regression to the mean does exist, but…

Extreme effect will, on average, be less extreme at another point in time. With more observations it will more likely be less extreme.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Regression to the mean

A

Effect from the theory of large numbers.

Effect applies to stable context.

In unstable context, extreme observation can be indicative of change (e.g., new chef).

Whether you punish or praise someone that comes late to their job/school doesn’t work, because it is random. You don’t have as much of an effect as you think you have.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Anchoring and adjustment

A

People give too much weight to the first bit of data for their quantitative estimates.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Anchoring

A

Assimilation of quantitative estimates to an available comparison figure.

  • Also, with numbers that are irrelevant to the decision.

Explanations

  • Initial hypothesis: answer = anchoring value; then people adjust too little.
  • Anchors make different types of information accessible, that are then used in the judgement.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Lexical Decision Task

A

Task

  • Task to measure cognitive accessibility
  • Existing word => “yes” button (as fast as possible)
  • Non-existing word => “no” button (as fast as possible)

Findings

  • More accessible words are recognized more quickly as being existing words.
  • In the 20,000 Euros condition participants recognized cheaper car brands (Opel, Golf) faster.
    o Greater chance that this information is retrieved from memory and used to make an estimate.
    o This is why the estimate was lower.
  • In the 40,000 Euros condition participants recognized expensive car brands (BMW, Mercedes) faster
    o Etc.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What helps against anchors?

A

Shaking your head helps a bit.

Leaving the negotiation (not too sure if this’ll work).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Availability heuristic

A

The more vivid, the more you overestimate it.

  • Easier to come up with example of event => event is estimated to be more probable.

Subjective ease counts, not number of retrieved examples.

  • Experiment where subjects had to give examples of assertiveness 12 vs. 6. The people that had to come up with 6 examples were found to think of themselves as more assertive than people that had to come up with 12 examples. This is probably because it’s hard to come up with 12 examples.

Often leads to good estimates but familiarity and vividness of information can bias estimates.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Simulation heuristic

A

The generation (mental simulation) of events.

  • Guides expectations, motivation, behaviour

Particularly with missed opportunities.

Counterfactual thinking.

  • Simulation of alternative results (“if only I had…”)
  • The easier it is => the greater the disappointment
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Experiment by Medvec, Madey & Gilovic (1995) on medal winners

A

Participants rated emotions of medal winners during prize ceremonies

  • Bronze winners look happier than silver winners (i.e., “gold losers”)
  • Silver winners can imagine what it would have been like to win (1 step), bronze winners can image what it would have been like to not have been on the podium at all (1 step).
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Representation heuristic

A

The more characteristics A (librarian) shares with B (Joris his behaviour), the more likely people think that A and B are associated.

  • B can be the consequence of A
    o Because he is a librarian he likes to read
  • B can be an exemplar of category A
    o Boring, tidy and enjoying reading have a lot in common with a librarian. Observed probability is greater that Joris works in a library than in construction.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Which option is most likely?
A BBRBRRBRBB
B BBBBBRRRRR
C BBBBBBBBBB

A

Each option is equally likely, but option A is more representative.

17
Q

We make stories which…

A

Can lead to accurate judgments.

Can lead to biases. e.g., if initial probabilities are ignored (base-rate fallacy).

  • Initial probability is much greater that a man works in the construction industry than in a library.
18
Q

Conjunction fallacy

A

Probability of co-occurring events is estimated to be greater than the probability of the separate events.

Pr(A) > Pr(A + B)

19
Q

Less is more

A

Experiment: There are two sets of dishes, A has 40 pieces and B has 24 pieces.

People were asked to estimate the price of the sets.

When shown together: A = 32 dollars, B = 30 dollars.
When shown separate: A = 23 dollars, B = 33 dollars

20
Q

Which leaves a more positive impression, PP or PPMM (P = positive feature, M = moderately positive feature)?

A

PP leaves a more positive impression than PPMM.

21
Q

Which leaves a more positive impression, NN or NNMM (N = negative feature, M = moderately negative feature)

A

NNMM leaves a more positive impression than NN.

22
Q

Dilution effect

A

Connected to less is more.

Experiment
If someone gives information about someone and asks if he’s an alcoholic, the answer is ‘quite sure’ 7/10.
If you give them the same information but with a couple of things added to it, the answer is ‘not so sure…’.
The added information is additional ‘non-diagnostic’ information, but it still makes them less sure.

23
Q

What happens when diagnostic information is diluted with non-diagnostic information?

A

Judgement becomes more moderate and less certain.

Conjunction is often ‘more diagnostic’.

Especially in observers who are motivated to form an accurate impression.

24
Q

Decoy effect

A

If you have to choose between a small popcorn (4 dollars) and a large popcorn (7 dollars) you will most likely choose the small popcorn.

But, by adding a medium option (6,50 dollars) you will be quicker to buy the large popcorn, because the difference between medium and large is quite small, and you are lured to large by wanting to buy medium. The medium option will be called the decoy.

  • Consumers change preference due to a dominated alternative.
  • A dominates B in terms of price but not in volume.
  • C dominates B in volume, and relative price.
25
Q

Are heuristics bad?

A
  1. Yes, they are bad
  • Biases have far-reaching consequences
    o Politics, education, stereotyping, health
    o Computer also use rules
     More consistently!
    o Humans can be trained to use rules more consistently
  1. Meh, they are not so bad.
  • People often do make good decisions
  • Biases are created in the lab
    o It’s no so bad in ‘real life’
    o Information is often unreliable, incomplete
  • Experts take other things into consideration (compared to computers)
    o Changing circumstances, criteria
  • Often unclear what exactly constitutes a good decision
  1. Just don’t think! (… sometimes)
  • Some people think intuitive decision making is good
  • Unconscious thought (complex decisions)
26
Q

Dijksterhuis (2004) on rating student apartments

A

Rate 4 student apartments

  • 12 pieces of info (positive/negative) for each apartment (e.g., “attractive neighborhood” or “unfriendly landlord”)
  • One apartment is best (8 pos., 4 neg.)
  • One apartment is worst (4 pos., 8 neg.)
  • Two apartments are in between (6 pos., 6 neg.)

Three conditions

  1. Rate immediately (base condition: little or no thoughts take place)
  2. Think about it for three minutes, then rate (conscious thoughts)
  3. Filler task to prevent conscious thoughts, then rate (unconscious thoughts)

Evaluation of best apartment compared to other apartments.

27
Q

What is the conclusion of the Dijksterhuis (2004) experiment on rating student aparments?

A

If there is some unconscious thinking going on while working on a filler task, you make the best decision on a complex situation.

28
Q

Dijksterhuis et al. (2006) involving the IKEA and Bijenkorf

A

He sent people to the IKEA or Bijenkorf, he then asked how satisfied they were with their decision.

Results:
People were more satisfied after the IKEA purchase when they were making the decision unconscious. People were more satisfied after the Bijenkorf purchase when they were making a conscious decision.

29
Q

What is the conclusion of the Dijksterhuis et al. (2006) experiment involving the IKEA and Bijenkorf?

A

The conclusion from Dijksterhuis (2004) is particularly true for complex decisions (too difficult to think about consciously).

30
Q

“Mistakes in heuristics are due to system 1.”

A

Even experts can fool themselves.

When you’re motivated, a bias can happen.

31
Q

“Anchors can easily be countered.”

A

No

32
Q

“2nd place winners are happier than 3rd place winners.”

A

No

33
Q

“Taking more courses makes you look smarter.”

A

Marco personally thinks it does.