SSR - Lecture 2

Linguistic issues in arguments that make propositions unclear (4)

• ambiguity

• vagueness

• implicit relativity

• metaphors

types of ambiguity (2)

• lexical ambiguity

• syntactic ambiguity

lexical ambiguity

• word or phrase with multiple different extensions (e.g. ball)

syntactic ambiguity

• sentence that can be interpreted in terms of different clear propositions because of its structure (e.g. “add the coriander, finely chopped (if desired)

vagueness

• statement with inexact or unclear meaning

primary connotation

• something falls within extension of term if, and only if, these conditions are met

secondary connotation

• characteristics that things within extensions usually exhibit, but not necessarily

What does a metaphor work on the basis of?

• secondary connotation

Which effect do more specific quantifiers have?

• more difficult to disprove

→ beware of intentional use of vague quantifiers

Distinction of generalizations

• hard generalization vs. soft generalizations

Hard generalizations

• absolute, exceptionless statements about a group

Soft generalizations

• describe trends, tendencies, or probabilities within a group → leave more room for exceptions

3 benchmarks of (ir)rationality

• logical benchmark

• probabilistic benchmark

• rational decision benchmark

logical benchmark (2)

• consistency of belief

• making deductively valid inferences

probabilistic benchmark (2)

• consistency of degrees of belief

• updating beliefs in accordance with rules of probability

rational decision benchmark (2)

• consistency of preference and choice

• deciding in manner that maximizes expected utility

Modus Ponens

• P1: if p → q

• P2: p

• C: q

→ valid

Affirming the consequent

• P1: if p → q

• P2: q

• C: p

→ invalid (fallacy)

Denying the antecedent

• P1: if p → q

• P2: no p

• C: no q

→ invalid (fallacy)

Modus Tollens

• P1: if p → q

• P2: no q

• C: no p

→ valid

conversational implicature

• (that which is usually implied in normal conversation) can be misleading when doing logical assessment

belief bias

• existing convictions/beliefs influence evaluation of structure of argument

What should we do to be logical and what do we do instead? (2)

• strive for falsification → confirmation bias

• consider only relevant information → rely on conversational implicature and belief bias

The Is-this-true? (3 steps)

1. How unexpected is new information if this is true?

2. How unexpected is new information if this is false?

3. Did you think this was true before the new information?

What should we do in regard to probability and what do we do instead? (2)

• use sample sizes → sample size neglect

• calculate conditional probabilities → base rate neglect

How do you calculate the expected value?

• expected value = probability x objective value

Subjective Expected Utility

• decision made by combining how much person cares about outcome (utility) and how likely they think that outcome is (subjective probability)

• subjective expected utility = utility x subjective probability

Allais paradox

• Most people choose A in exp 1 AND B in exp 2

• Expected outcome 1A (1milj) vs 1B (1.39milj) → So exp value 1B is 390.000 higher

• Expected outcome 2A (110k) vs 2B (500k) → So exp value 2B is 390.000 higher

What should people do to make rational decisions and what do they do instead?

• maximize expected utility → sensitivity to logically irrelevant contextual variation

What happens in normative judgement (3)

• deductively valid arguments

• calculating conditional probabilities

• maximizing expected utility

What happens in intuitive judgement (3)

• how easy does this come to mind?

• does example match category?

• use info given as anchor and adjust

Which heuristics and biases are there? (5)

• availability heuristic

• hindsight bias

• representativeness heuristic

• conjunction fallacy

• anchoring and adjustment

availability heuristic

• makes people judge the likelihood or frequency of an event based on how easily an example comes to mind

hindsight bias

• common tendency to believe that past events were more predictable than they actually were before they happened

representativeness heuristic

• shortcut that helps people make quick decisions by comparing a new situation, person, or object to an existing prototype or stereotype in their minds

→ e.g. gambler’s fallacy (= after RRRRRR on roulette wheel most people would put money on B), hot hand (= player has hot hand and is on winning streak → likely to score again)

conjunction fallacy

• cognitive bias where people mistakenly believe that two specific conditions are more likely to happen together than a single general condition

anchoring and adjustment

• people anchor estimates on values that are suggested to them, but are too conservative in adjusting their estimate based on other information than anchor