PSY 324: Exam #3

Deductive reasoning

• If the argument is accepted, the conclusion must also be accepted. 

• General Premise → Specific Instance

Inductive reasoning

• If the argument is accepted, the conclusion is REASONABLE to accept. 

• Specific Observation → General Hypothesis

Validity

Required for soundness.

If we accept the premises, must the conclusion follow?

• Yes. The argument is valid.

Soundness

Are the premises true?

• We can accept that Betty is a dog.

• Is it true that all dogs like to go for walks outside? There are millions of dogs out there; I can’t be sure that’s true.
  

The argument is probably not sound.

Modus ponens

Affirming the antecedent.

It’s valid

It is often answered correctly

Modus tollens

Denying the consequent.

It is valid

It is tougher for people to grasp this idea.

Affirming/denying the antecedent/consequent

Base-rate neglect fallacy

Flaw in judgement. being tested for a disease example.

To have a true understanding, we need to know what proportion of the population actually has the disease – the base rate

Phonemes

The smallest unit of sound in language.
Example: T-u-r-k-ey (5 units)

Morphemes

The smallest unit of meaning. “Turkey” can be a morpheme when thinking of a farm animal. “Turkey|s” can have two morphemes by being more than one turkey. It is plural.

Motor theory

The system for producing, perceiving, and the physical motions of speech.

Includes species-specific.

Auditory theory

Speech perception is derived from the properties of the auditory system. It is not species-specific

Chomsky’s universal grammar

All languages share a fundamental basic property. Such as consonant and vowel sounds when spoken.

Pronunciation versus ordering

Pronunciation

Perhaps analogy (Glushko, 1979)

• “Loap” sounds like “trope” and both can be pluralized the same way

• However, what about “loaf” or “loan”?

Maybe it’s determined by the body itself (tongue, larynx, etc) or cultural norms

Ordering

Comprehension of language is dependent not only on the interpretation of phonemes and morphemes, but also on the ordering of words and their relations, or their syntax

Active versus passive sentences

“The dog is chasing the cat” - an active sentence. Participants responded faster to active sentences rather than passive sentences. 

Aspects of problems (4 of them)

1. Goals: the end game

2. Obstacles: hinders the achievement of the goal

3. Givens: anything that can relate to or is relevant to the problem. Conditions, restraints, environmental details, objects associated, etc.

4. Means of Trans-conditions: The problem has to be solvable. Meaning there must be a means or a way to solve it. Practical methods are included to solve the problem.

Problem space

Refers to the mental representation of the task environment wherein the problem-solving process occurs.

It is based on internal factors & outside of the self.

Task environment

Refers to the objective factors of a problem outside of cognitive processes

It is external objective factors related to the problem.

Problem state

Refers to a past, present, or future state in the problem-solving situation

Minimum of 2 states:: an initial state and a goal state or a solved state.

It can be referred to as a scenario. A complex problem that cannot be fixed in one step would be this term.

Problem operator

Refers to the means by which a problem solver can move from one state to another. Essentially taking actions to reach the goal state.

Requires means of trans-conditions.

Well-defined problems

• the conditions and aspects of the problem are clearly specified without ambiguities or uncertainties

  • Examples: games, mazes, arithmetic

Ill-defined problems

• the conditions and aspects are not clearly specified.  Ambiguities and uncertainties cloud the paths to the goal.

  • Example: attaining a desired career

Algorithms

• a systematic procedure which is certain to produce a solution to a problem

  • Involves searching all possible problem states to derive a solution

  • Perfectly reliable, but sometimes immensely time-consuming

  • Often intractable for the human mind

  • NOTE: Does not have to be mathematical in its description

Heuristics

• a method which can direct a partial search to find pertinent problem state information

  • More efficient than algorithms, but do not guarantee a solution

  • Example of hill climbing and means-end-analysis

  • Simply put: a short cut

Hill climbing

With this method, at any given point in a problem space, one must simply choose the next state closest to the goal.

Means-end analysis

• the problem solver reviews numerous problem operators available and identifies which one best reduces the distance to the goal.

  • This time, the best operator is to get on an airplane (since it reduces 3,000 miles of distance more effectively than walking or driving), which requires a trip to the airport.

Functional fixedness

an inability to use objects for purposes other than their intended functions or designs.

Example: sticking a candle to a wall

Normative models

• these attempt to predict how humans ought to behave or what would be the most optimal, also called rational models. 

• Ought to behave

• Includes expected value theory and expected utility theory

Descriptive models

• these attempts to predict actual human behavior without regard for what is optimal. 

• Actually behave 

• Includes prospect theory and regret theory

Expected value theory

• follows the math

• was an early model in economics designed to predict how people determine monetary value

Expected utility theory

• consistency

• This theory follows the foundation of expected value theory. In this case, utility is still accounted for in a calculation. 

• This term refers to the specific goals a person has when making a decision.

• can’t be tested directly, so it is tested by comparison.

Prospect theory

• part of descriptive.

• an adjusted expected utility with the hopes of accounting for the various non-normative tendencies observed in human decision-making data. 

• it is subjective/psychological probabilities

Regret theory

• provides another way researchers have attempted to account for certainty effects. 

• The latter choice may seem more rational. The threat of regret is considerable to the decision-maker.

Transitivity principle

• Part of the normative model

•  if A is preferred over B, and B is preferred over C, then A is preferred over C. 

  • Or also written as: if a>b and b>c, then a>c

Non-contradiction principle

• Part of the normative model

• if A is preferred over B, B is not preferred over A. Do these rules apply to human decisions?

  • Or also written as: if a>b, don’t b>a

Violations of normative principles

• Violations of transitivity

  • Sometimes this does not seem to apply

  • Consider candidates in an election – complex dynamics may influence our perspectives of how candidates compare with each other

  • Example: If I prefer Mary Smith over Bill Williams, and I prefer Bill Williams over Jane Jones, that doesn’t necessarily mean I prefer Mary Smith over Jane Jones
    

• Violations of non-contradiction

  • Our preferences between two or more options may not be absolute; certain contextual details may influence the preference.

  • Example: If on Monday night I prefer spaghetti over tacos for dinner, that doesn’t imply that I will prefer spaghetti over tacos in all possible situations

Risk-seeking versus risk-averse (and framing effects)

• When a problem is framed in terms of losses, people tend to be risk-seeking.

  • When program A describes “400 people will die”, this framing encourages greater willingness to gamble for a better result.
    

• When a problem is framed in terms of gains, people tend to be risk-averse.

  • When program A describes “200 people will be saved”, this framing deters willingness to gamble at risk of losing those gains.

Availability heuristic

Refers to the tendency of people to make judgments or decisions based upon what most readily comes to mind.

Example: writing down as many words that begin with the letter “K” in 30 seconds and another 30 seconds of writing down words with “K” as the third letter.

Representativeness heuristic

Refers to cases in which an event is judged to be more likely if its features or properties are more like its category (i.e. more representative of the category).

Example: where is a person more likely to struggle with thirst? near a lake or in a desert?

Anchoring and adjustment heuristic

a proposed heuristic by which people make estimates on the basis of some starting or baseline value, and adjust from that point with new information.

Conjunction fallacy

Example: Which of the following is more likely?

◦ Janet raises cattle on a farm in the country.

◦ Janet recently applied to work for NASA and raises cattle on a farm in the

country

Many are inclined to choose the second option. This is the conjunction fallacy, which is attributed to the representativeness heuristic.

Optimizing

The most basic approach to a decision with a vast array of choices is to examine every one of them and then select the best option.

Example: university departments may try to optimize when selecting among graduate student applicants.

Satisficing

Refers to cases in which a decision maker will decline to investigate every single option available.

◦ A smaller subset is reviewed instead

◦ The best option among them is chosen assuming it is satisfactory even if not the best

Example: Instead of searching the entire barrel for the best possible apple, someone may find a satisfactory apple near the top

Elimination by aspects

procedure (Tversky, 1972) describes a process by which a decision can be reduced to its various components.

◦ Some aspect is considered, and any option not meeting a criterion is

eliminated.

◦ This repeats until all aspects are exhausted and/or only one option remains.