Semantic organization

1
Chapter 9

• Semantic organization

2
Organization and memory
Take a moment to memorize the following words.
3
Apple Dog Car Cat Orange
Horse Truck Pear Boat Cow
Peach Airplane Pig Plum Tractor
4
Recall Words
5
Did you recall

• The animals together?
• The fruits together?
• The vehicles together?

6
Did you recall

• The animals together?
• The fruits together?
• The vehicles together?

If you are like most people, you did. The
clustering of items from a common category is an
example of organization.
7
Types of organization

• Category clustering Grouping items together
  that come from the same conceptual category
• Associative clustering Grouping items together
  that are associatively related
• Subjective organization Consistently grouping
  unrelated items together on some idiosyncratic
  basis.

8
Category Clustering
Apple Dog Car Cat Orange Truck
Apple Orange Dog Cat Car Truck
back
9
Associative Clustering
Easter Snow Party Egg White Birthday
Easter Egg White Snow Birthday Party
back
10
Subjective Organization
Whiskey Bible
Cabbage Whiskey Stove Bible
Cabbage Stove
11
Why is organization important?

• It reflects the fact that our knowledge is highly
  organized.
• There is an important relationship between
  organization and memory performance.

12
Organized material is easier to learn and
remember.
Which of the following presentations of words
would be easier for you to remember?
13
Bronze
Slate
Common
Gold
Ruby
Lead
Minerals
Lead
Rare Brass Stones
Steel Granite Platinum Masonry
Precious Marble Iron Emerald
Limestone Alloys Copper Silver
Metals Diamond Lead
14
Minerals
Metals
Stones
Rare
Common
Alloys
Precious
Masonry
Platinum Silver Gold
Aluminum Copper Lead Iron
Bronze Steel Brass
Sapphire Emerald Diamond Ruby
Limestone Granite Marble Slate
15
Why does organization facilitate memory?

• Organization provides a systematic way to search
  memory a retrieval plan.
• Organization provides an effective strategy for
  generating information from memory.

16
The Organization of Knowledge
Associative Network Models
17
Historical Antecedents

• Ross Quillian developed a computer program that
  was meant to model certain aspects of human
  linguistic behavior
• Program required that he build in word meanings
• Word meanings were represented as a set of
  relationships among concepts.

18
A Loose Example Uncle
What is an uncle?
19
A Loose Example Uncle
UNCLE
Father
Mother
Brother
Sister
Brother
Sister
Husband
Husband
20
More on Quillians program

• There were a variety of relationships/links among
  concepts
• subordinate/superordinate links (Isa links)
• modifying links
• conjunctive links
• disjunctive links
• conditional links
• Program would compare pairs of concepts

21
A Simplification of the ModelCollins and
Quillians Hierarchical Model (TLC)

• Only considers subordinate/superordinate and
  modifying links.
• Results in knowledge structures that are
  hierarchically organized.

22
Breathes
Animal
Eats
Has skin
Feathers
Can Swim
Bird
Fish
Can Fly
Has Gills
Has Wings
Has Fins
Canary
Robin
Salmon
Trout
Sings
Red Breast
White
Pink
Yellow
Eats Worms
Edible
Edible
23
Modifiers are stored as high in the hierarchy as
possible. This feature is referred to as
cognitive economy, since it results in more
economical storage of information.
24
Hierarchical model explains how we are able to
answer basic questions such asWhat is a canary?
25
Breathes
What is a Canary?
Animal
Eats
Has skin
Feathers
Can Swim
Bird
Fish
Can Fly
Has Gills
Has Wings
Has Fins
Canary
Robin
Salmon
Trout
Sings
Red Breast
White
Pink
Yellow
Eats Worms
Edible
Edible
26
Hierarchical model allows us to draw
inferencesWhat is a nuthatch?
27
Hierarchical model allows us to draw
inferencesWhat is a nuthatch?
Its a bird. What else do you know about it now?
28
What else do you know about a nuthatch?
Breathes
Animal
Eats
Has skin
Feathers
Can Swim
Bird
Fish
Can Fly
Has Gills
Nuthatch
Has Wings
Has Fins
Canary
Robin
Salmon
Trout
Sings
Red Breast
White
Pink
Yellow
Eats Worms
Edible
Edible
29
A test of the model using a sentence verification
task

• Participants asked to verify factual sentences as
  being true or false
• Example Is the sentence A robin is a bird
  true or false?
• Since most will know the correct answer,
  verification time is recorded.
• Verification times for different sentences are
  compared

30
ExampleIndicate as quickly as you can if the
following sentences are true or false.
31

32
A canary is a bird
33

34
A canary is a rock
35

36
A canary is an animal
37

38
A canary is a canary.
39
Which sentence do you believe would be verified
most quickly? Most slowly?

• A canary is a canary
• A canary is a bird
• A canary is an animal

40
Why?
41
Breathes
Animal
Eats
Has skin
Feathers
Can Swim
Bird
Fish
Can Fly
Has Gills
Has Wings
Has Fins
Canary
Robin
Salmon
Trout
Sings
Red Breast
White
Pink
Yellow
Eats Worms
Edible
Edible
42
The finding that sentence 1 is verified most
quickly and sentence 3 most slowly illustrates
the category size effect.
1 A canary is a canary 2 A canary is a
bird 3 A canary is an animal
43
Which of the following sentences are true?
44

45
A canary has feathers
46

47
A canary has gills
48

49
A canary is yellow
50

51
A canary has skin
52
Which sentence do you believe would be verified
most quickly? Most slowly?

• A canary is yellow
• A canary has feathers
• A canary has skin

53
Why?
54
Breathes
Animal
Eats
Has skin
Feathers
Can Swim
Bird
Fish
Can Fly
Has Gills
Has Wings
Has Fins
Canary
Robin
Salmon
Trout
Sings
Red Breast
White
Pink
Yellow
Eats Worms
Edible
Edible
55
The finding that sentence 1 is verified most
quickly and sentence 3 most slowly supports the
notion of cognitive economy.
1 A canary is yellow 2 A canary has
feathers 3 A canary has skin
56
Which of the following sentences are true?
57

58
A canary is a bird
59

60
A robin is a bird
61

62
A trout is a bird
63

64
A penguin is a bird
65
Which sentence do you believe would be verified
most quickly?

• A canary is a bird
• A penguin is a bird

66
Could the hierarchical model explain this finding?
67
Bird
Canary
Robin
Penguin
68
The finding that sentence 1 is verified more
quickly than sentence 2 illustrates the
typicality effect.
1 A canary is a bird 2 A penguin is a bird
The typicality effect seems to be inconsistent
with the hierarchical model since all members of
a category are thought to be equal
69
Sometimes there are reversals of the category
size effect. Such reversals also seem to be
inconsistent with the hierarchical model
70
Example
A chicken is a bird A chicken is an animal
71
Animal
Bird
Chicken
72
Sometimes the results are inconsistent with the
notion of cognitive economy.
73
Example
A salmon is pink A salmon can swim
74
Breathes
Animal
Eats
Has skin
Feathers
Can Swim
Bird
Fish
Can Fly
Has Gills
Has Wings
Has Fins
Canary
Robin
Salmon
Trout
Sings
Red Breast
White
Pink
Yellow
Eats Worms
Edible
Edible
75
Because of the findings that appeared to be
problematic for the hierarchical model, Collins
and Loftus (1975) modified and extended the
model.
76
Changes to Model

• Abandoned notion of strict hierarchy
• Incorporated notion of criteriality
• Added notion of spreading activation

77
Penguin
Fish
Swims
Animal
Salmon
Bird
Fly
Chicken
Canary
Pink
Yellow
Red
Back
78
Criteriality refers to the importance of a
particular link to the meaning of the concept.
79
Criteriality refers to the importance of a
particular link to the meaning of the
concept. It can be represented in a figure by
the length of a line (shorter line, higher
criteriality).
80
Penguin
Fish
Swims
Animal
Salmon
Bird
Fly
Chicken
Canary
Pink
Yellow
Red
81
By incorporating the notion of criteriality, the
model can account for the typicality effect
82
By incorporating the notion of criteriality, the
model can account for the typicality effect
Typicality effect Response to sentence 1 is
faster than sentence 2
1 A canary is a bird 2 A penguin is a bird
83
Penguin
Fish
Swims
Animal
Salmon
Bird
Fly
Chicken
Canary
Pink
Yellow
A canary is a bird A penguin is a bird
Red
84
The notion of criteriality, and the abandonment
of strict hierarchies, allows the model to
account for reversals in the category size effect
85
The notion of criteriality, and the abandonment
of strict hierarchies, allows the model to
account for reversals in the category size effect
Reversal in category size effect Response to
sentence 2 would be faster than response to
sentence 1
1 A chicken is a bird 2 A chicken is an
animal
86
Penguin
Fish
Swims
Animal
Salmon
Bird
Fly
Chicken
Canary
Pink
Yellow
A chicken is a bird A chicken is an animal
Red
87
The notion of criteriality, and the abandonment
of strict hierarchies, allows the model to
account for instances when cognitive economy
appears to fail
Example Responses to sentence 2 are faster
than responses to sentence 1 1 A salmon is
pink 2 A salmon can swim
88
Penguin
Fish
Swims
Animal
Salmon
Bird
Fly
Chicken
Canary
Pink
Yellow
A salmon is pink A salmon can swim
Red
89
The concept of spreading activation
90
Penguin
Fish
Swims
Animal
Salmon
Bird
Fly
Chicken
Canary
Pink
Yellow
Red
91
The concept of spreading activation

• provides a model of thought
• allows one to account for semantic priming

92
Spreading Activation as a Model of Thought
One can think about thinking as the spread of
activation through the semantic network.
93
An illustration
Start thinking about summer and see where your
thoughts take you.
94
An illustration
Start thinking about summer and see where your
thoughts take you. Does the drift of your
thoughts make sense in the context of spreading
activation?
95
Another illustration
The Word Association Game!
96
Semantic Priming
97
Another semantic priming paradigm

• Participant receives a series of trials
• On each trial the participant is shown a category
  label (e.g. fruit) with a word (e.g. apple)
• The participants task is to indicate if the word
  is an example of the category.
• Response times are recorded

98
Example
Does the following word belong to the category
TOOL?
99
Example
Does the following word belong to the category
TOOL?
WRENCH
100
Does the following word belong to the category
FRUIT?
101
Does the following word belong to the category
FRUIT?
LAKE
102
Does the following word belong to the category
COLOR?
103
Does the following word belong to the category
COLOR?
RED
104
Does the following word belong to the category
COLOR?
105
Does the following word belong to the category
COLOR?
YELLOW
106
Semantic priming would occur if the response to
COLOR - YELLOW was faster than COLOR - RED
107
Semantic priming would occur if the response to
COLOR - YELLOW was faster than COLOR - RED How
would the notion of spreading activation explain
this?
108
Penguin
Fish
Swims
Animal
Salmon
Bird
Fly
Chicken
Canary
Pink
Yellow
COLOR -- RED COLOR -- YELLOW
Color
Red
109
Other Theories of Semantic Memory

• Propositional Network Theory
• Feature Theory

110
Propositional Network Theories

• Similar to associative network theories in that
  it consists of a network
• Different in that the basic unit of semantic
  memory is a proposition.
• A proposition is the smallest unit of knowledge
  that can be verified as true or false.

111
Example
A canary is a bird can be verified as being
true, but none of its components (e.g., canary
or bird) can be verified as true. Therefore,
a canary is a bird is a single proposition.
112
While a canary is a bird is a sentence, it is
not true that a sentence and a proposition are
the same thing.
113
While a canary is a bird is a sentence, it is
not true that a sentence and a proposition are
the same thing. The sentence a canary is a
yellow bird actually contains two propositions A
canary is a bird The bird is yellow
114
Propositions can be represented in a figure
Is a
Relation
Canary
Bird
Subject
Object
115
Propositions are linked together to form a network
Canary
Yellow
Subject
Object
Object
Subject
Bird
1
2
Relation
Relation
Is
Is a
116
Psychological reality of propositions
117
Psychological reality of propositions
Memorize the following sentence Children who
are slow, eat bread that is cold
118
In a moment I am going to give you a word from
the sentence. What I want you to do is write
down the first word from the sentence (other than
the word I give you , of course) that comes to
your mind.
119
In a moment I am going to give you a word from
the sentence. What I want you to do is write
down the first word from the sentence (other than
the word I give you , of course) that comes to
your mind.
Ready?
120
slow
121
Did you write down children?
122
Did you write down children?
People are more likely to write down children
than they are to write down eat or bread even
though eat and bread are closer to slow in
the sentence.
123
An explanation
The word slow modifies children. They are part
of the same proposition. The words eat and
bread belong to different propositions.
124
Feature Theory
Concepts are represented in memory as a set of
features or characteristics.
125
Example Bird
Feathers Wings Fly Eggs Beaks Talons Calls
126
Those features which are most critical to the
meaning of the concept are referred to as
defining features. Those features that are less
critical are referred to as characteristic
features.
127
Example Bird
Feathers Wings Fly Eggs Beaks Talons Calls
Defining
Characteristic
128
The process of sentence verification
The sentence a canary is a bird would be
verified by comparing the feature list for
canary with the feature list for bird
129
Comparison of feature lists is a two stage process

• Stage 1 Compare two feature lists to determine
  overall similarity
• If overall similarity exceeds an upper criterion,
  the person will respond yes quickly.
• If overall similarity falls below a lower
  criterion, the person will respond no quickly.
• If overall similarity falls between two
  criterion, second stage begins.
• Stage 2 A slower, more deliberate, comparison
  of defining features to determine truth.

130
Stage 2 processing required
No
Yes
Upper Criterion
Lower Criterion
Low
High
Similarity
131
Feature set theory provides explanations for

• Category size effect (and reversals)
• Typicality effect

132
Schema Theory
133
Most basic concept of schema theory
Knowledge is stored in schemas
134
Whats a schema?
For the most part, schemas are vaguely defined.
For our purpose, we will define a schema as an
organized body of knowledge about something.
135
Examples
What do you know about the following? a car a
chair a school
136
Script, a special type of schema
A script contains your knowledge of how to
accomplish something, such as order dinner at a
restaurant or get dressed.
137
The importance of scripts a quick illustration
Does this make sense?

The procedure is quite simple. First you arrange
items into different groups. Of course one pile
may be sufficient depending on how much there is
to do. If you have to go somewhere else due to
lack of facilities that is the next step
otherwise, you are pretty well set. If is
important not to overdo things. That is, it is
better to do too few things at once than too
many. In the short run this may not seem
important, but complications can easily arise. A
mistake can be expensive as well. At first, the
whole procedure will seem complicated. Soon,
however, it will become just another facet of
life. It is difficult to foresee any end to the
necessity for this task in the immediate future,
but then, one never can tell. After the
procedure is completed one arranges the materials
into their appropriate places. Eventually, they
will be used once more and the whole cycle will
then have to be repeated. However, that is part
of life.
138
Would it make sense if you knew the passage was
about washing clothes?
139
Two experiments which illustrate the effect of
schemas on memory

• Bartletts War of the Ghosts study.
• Brewer and Treyenss study of memory for places.

140
Bartletts study

• Read The War of the Ghosts to his participants
• Had them recall the story at intervals ranging
  from immediately to years
• Found systematic changes in the recall protocols
• Details tended to be left out
• Changes in detail that reflected culture of
  participants

141
Bartlett interpreted these results in the context
of his notion of a schema.
142
Bartlett interpreted these results in the context
of his notion of a schema. 1) Knowledge is
stored in schemas
143
Bartlett interpreted these results in the context
of his notion of a schema. 1) Knowledge is
stored in schemas 2) Knowledge is abstracted from
experience (provides explanation for loss of
detail)
144
Bartlett interpreted these results in the context
of his notion of a schema. 1) Knowledge is
stored in schemas 2) Knowledge is abstracted from
experience (provides explanation for loss of
detail) 3) Experience is interpreted in the
context of theses schemas (provides explanation
for memory disortions)
145
Brewer and Treyenss study
Participants were placed in an office under the
pretext that they were waiting for the
experimenter
146
Brewer and Treyenss study
Participants were placed in an office under the
pretext that they were waiting for the
experimenter After a short period (35 sec)
participants were moved to another room and asked
to remember what they had seen in the office.
147
Heres the office
148
Some of the items in the office are ones you
would not expect to see (e.g. skull)
149
Some of the items in the office are ones you
would not expect to see (e.g. skull) Some items
you would expect to see were not present (e.g.
books)
150
Results

• There was a tendency for participants to recall
  items that they would expect to find even when
  the item was not there
• There was a tendency to forget objects one would
  not expect to find.
• There was a tendency normalize the office (the
  table was thought to be rectangular the note pad
  was on the desk)

151
Conclusions

• New information is integrated with old-schema
  based information
• Schema directs the retrieval of information from
  memory

152
Schema theory suggests are four processes
important for memory

• Selection
• Abstraction
• Interpretation
• Integration

153
1. Selection
Not all information is encoded. What is selected
for encoded is determined by the schema that has
been activated.
154
2. Abstraction
Memorial representations are abstract. Verbatim
information is lost.
155
3. Interpretation
Information is interpreted on the basis of
existing knowledge.
156
4. Integration
Information is integrated into coherent schemas