Vous êtes sur la page 1sur 50

# Inductive and Deductive

Reasoning
Lecture 9
Prof. Storbeck

## Problem Solving and

Reasoning
Problem solving: set of cognitive processes
applied to overcome obstacles to a goal
Reasoning: cognitive processes used to
make inferences from knowledge and draw
conclusions.
Draw on categorization, imagery, decision
making, attention, LTM, WM, executive
processes, and language.

Problem
Problem has 3 parts
Goal state
Initial state
Operations

Problems
Well defined problem

Insight

## Insight invokes activity in

the right anterior superior
temporal gyrus and is
associated with increased
gamma bursts.
Kounios & Beeman,
2009

## Strategies and Heuristics

Algorithm
Heuristics

Heuristic
Heuristic, contrary to moving toward
goal state.

## Strategies and Heuristics

Algorithm
Heuristics
Random search
Trial and error
Behaviorist approach

## Strategies and Heuristics

Algorithm
Heuristics
Random search
Trial and error
Behaviorist approach

Hill-climbing
Water Jug Problem
Left-prefrontal

Initial State
8

0
0

Goal State
4

## Strategies and Heuristics

Algorithm
Heuristics
Random search
Hill-climbing
Water Jug Problem
Left-prefrontal

Means-ends analysis
Break down problem into
subgoals/subproblems
Tower of Hanoi

## Experts vs. Novices

Differences in organizing problems:
Novices = surface features
Experts = abstract or deeper level
concepts

Analogical Reasoning

## Cognitive Science relied on the

computer model to understand
memory
Biological viruses helped computer
scientists solve computer viruses.

Analogical Reasoning

1.
2.
3.
4.
5.

Retrieval
Mapping
Evaluation
Abstraction
Predictions

Military Problem
A small country was ruled from a strong fortress by a dictator. The
fortress was situated in the middle of the country, surrounded by farms
and villages. Many roads led to the fortress through the countryside. A
rebel general vowed to capture the fortress. The general knew that an
attack by his entire army would capture the fortress. He gathered his
attack. However, the general then learned that the dictator had planted
mines on each of the roads. The mines were set so that small bodies of
men could pass over them safely, since the dictator needed to move his
troops and workers to and from the fortress. However, any large force
would detonate the mines. Not only would this blow up the road, but it
would also destroy many neighboring villages. It therefore seemed
impossible to capture the fortress.

Transfer of Problems

Suppose you are a doctor faced with a patient who has a malignant
tumor in his stomach. It is impossible to operate on the patient, but
unless the tumor is destroyed the patient will die. There is a kind of
ray that can be used to destroy the tumor. If the rays reach it all at
once at a sufficiently high intensity, the tumor will be destroyed.
Unfortunately, at this intensity the healthy tissue the rays pass through
on the way to the tumor will also be destroyed. At lower intensities the
rays are harmless to healthy tissue, but they will not affect the tumor,
either. What type of procedure might be used to destroy the tumor
with the rays and at the same time avoid destroying the healthy tissue?

1) Solve Problem
problem (military), later solve

10%
30%
3) remember military + hint solve

## Three attempts to induce

transfer
1) subjects told to summarize story
in abstract terms
if you need a large force to accomplish
some purpose, but are prevented from
applying such a force directly, many
smaller forces

## 2) general principle provided

3) draw diagram
Beveridge & Parkins,
1987

## Three attempts to transfer

Transfer:
More transfer when schema more
abstract
More transfer when stories more similar
More transfer with diagram or principle
Abstract thinking separates
Experts from Novices!

Theories of Analogical
Reasoning
Structure Mapping Theory (SMT)
Search LTM for structural component
similarity
Evaluation of mapping

## Learning and Inference with Schemas

and Analogies (LISA)
Neural Networks
Target activates features
Features activates Source for analogy

## Kroger et al., 2002

Inductive Reasoning
Use of knowledge of specific known
unknown instances
Category-based inductions
General induction
Specific induction

No Inductive
Process can ever
be certain

Inductive Arguments
I have seen 99 swans and all of them are white.
Therefore, all swans are white.
The boiling point of water in the past has always been
212F.
Therefore, tomorrow the boiling point of water will be
212F.
There is intelligent life on Mercury.
There is intelligent life on Venus.
There is intelligent life on Jupiter.
Therefore, there is intelligent life on Mars.

Inductive Arguments
Not valid or invalid
Vary in strength
A grab bag of tricks
Not just particular to general
E.g., Chimpanzees like eating onions.
Therefore,gorillas like eating
onions.

## But if we do go from particular to

general

Inductive Reasoning
Scientist use to generate predictions
Inductive Reasoning with a game
Deck of cards varied on 4 dimensions
with 3 attributes

## Color (white, black, blue)

Number of items on card (one, two, three)
Shape of item (circle, cross, square)
Number of borders (one, two, three)
3 X 3 X 3 X 3 = 81 cards, instances

Inductive Reasoning
Simples rules easy to get (red)
Conjunctive rules next easiest (red +
square)
Disjunctive rules harder (red or
square)
Negative rules (not red) were still
harder
Disjunctive negative rules most hard
(not red or cross).

Category-Based Inductions
Induction based on category and
related features.
Similarity
Typicality
Group homogeneity

Similarity-Coverage Model
Typicality for category-based induction
More typical, more readily feature is
mapped to conclusion:
Dogs have a liver.
Cats have a liver.
Conclusion: Mammals have a liver.
Dogs have a liver.
Whales have a liver.
Conclusion: Mammals have a liver.

Brain!

## WCST & fMRI

Monchi et al. 2001

## Seger et al. 2000

Inductive Reasoning
Implausible arguments activate areas
noted for error detection!

Deductive Reasoning
Syllogism

Deductive Arguments
All men are mortal.
Socrates is a man.
Therefore, Socrates is mortal.

premise
premise
conclusion

## All artists are beekeepers.

All beekeepers are chemists.
All artists are chemists.
NOT just general to particular
IF PREMISES ARE TRUE THEN CONCLUSION MUST
BE TRUE BY VIRTUE OF ITS FORM

Deductive vs Inductive
All plurbs are chenny.

## I saw 100 plurbs and all

are chenny.
Fred is a plurb.
Fred is a plurb.
Conclusion? Fred is Conclusion? Fred is
Deductive:
Inductive:
True by virtue of form alone. May be stronger or
weaker.
Meaning of words not matter.Meaning of words
matters.

Categorical Syllogisms
Relations between two categories
Premise 1: All A are B
Premise 2: C is an A.
Conclusion: C is B.

Categorical Syllogisms

## Universal Affirmation: All A are B.

Universal Negative: No A are B.
Particular Affirmative: Some A are B.
Particular Negative: Some A are not
B.

## Can be represented as Venn

diagrams.

Conditional Syllogisms
First premise If p, then q
Second premise to take one of four forms:
Affirmation of the antecedent: p is true
Denial of the antecedent: p is not true
Affirmation of the consequent: q is true
Denial of the consequent: q is not true.
If the automobile is a Porsche, then it is reliable.
The Boxster is a Porsche.
Conclusion: The Boxster is reliable.

A

7 D

## If card has a vowel on one side,

then it has an even number on the
other side
Goal: to determine whether the
rule is true or false with least
number of cards flipped.

## Errors in Deductive Thinking

Form
Premise 1: No As are Bs
Premise 2: No Bs are Cs
Conclusion: No As are Cs

Content
Premise 1: All things that have a motor
need oil.
Premise 2: Automobiles need oil.
Conclusion: Automobiles have motors.

Belief-bias Effect
Premise 1: No cigarettes are
inexpensive
Premise 2: Some addictive things are
inexpensive
not cigarettes.

Belief-bias Effect
Premise 1: No cigarettes are inexpensive
Premise 2: Some addictive things are
inexpensive
Conclusion: Some additive things are not
cigarettes.
Premise 1: No addictive things are inexpensive
Premise 2: Some cigarettes are inexpensive
Conclusion: Some cigarettes are not addictive

Theories of Deductive
Reasoning
Theory of Mental Models
Mental model constructed
A tentative conclusion is generated,
evaluated
The conclusion must be validated
Accounts for Form and Content errors.