Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

YEAR 2 Perception > Object Recognition > Flashcards

Object Recognition Flashcards

1
Q

What is object recognition for humans?

A

Perception of familiar items

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

What is object recognition for computers?

A

Perception of familiar patterns

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

Between category discriminations

e.g. Is this a chair, is it not a chair?

Is this object recognition for humans or computers?

A

Computers

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

Within category discriminations

e.g. Whose face is this?

Is this object recognition for humans or computers?

A

Humans

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

Why is object recognition difficult?

A

Environment contains hundreds of overlapping objects

Yet perceptual experience is of structured coherent objects which we can recognise, use and usually name

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

True or False?

Apparent size and shape of an object does not change despite large variations in retinal image

A

True

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

Apparent ____ and ____ of an object does not change despite large variations in retinal image

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

What are 5 examples of variability in object recognition?

A
  1. Translation invariance
  2. Rotation invariance
  3. Size invariance
  4. Colour
  5. Partial occlusion
    and presence of other objects
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Define translation invariance

A

Recognising an object when it is not in the same location as before

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Recognising an object when it is not in the same location as before

This is known as…?

A

Translation invariance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Define rotation invariance

A

Recognising an object when it is not in the same rotation as before

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

Recognising an object when it is not in the same rotation as before

This is known as…?

A

Rotation invariance

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

Define size invariance

A

Recognising an object when it is not the same size as before

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Recognising an object when it is not the same size as before

This is known as…?

A

Size invariance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Define colour (variability)

A

Recognising an object when it is not the same colour as before

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Recognising an object when it is not the same colour as before

This is known as…?

A

Colour (variability)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Define partial occlusion
and presence of other objects

A

Where an object is blocked off partially by some other object/space

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Where an object is blocked off partially by some other object/space

This is known as…?

A

Partial occlusion
and presence of other objects

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What is intra-class variation?

A

Image variations occur between different images of one class

e.g. These objects are all chairs but they all look so different. (different shapes, structure, colour, height)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Image variations occur between different images of one class

e.g. These objects are all chairs but they all look so different. (different shapes, structure, colour, height)

This is known as…?

A

Intra-class variation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What are the 2 challenges of object recognition?

A
  1. Intra-class variation
  2. Viewpoint variation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What is viewpoint variation?

A

Being able to recognise an object even from different viewpoints

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Being able to recognise an object even from different viewpoints

This is known as…?

A

Viewpoint variation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

e.g. Recognising that the object is a coffee pot from the top view (looks like a square with a circle in the middle and a rectangle at the side)

This is an example of…?

A

Viewpoint variation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

How do we recognise the letter ‘A’ according to template theories?

A

When you see your letter A, you match it to what you have stored in your LTM memory (mini copy or template in LTM of all known patterns) and when you get a good enough match, you can recognise that as a letter A

simply = Multiple templates are held in memory, we compare stimuli to templates in memory for one with the greatest overlap until a match is found

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

When you see your letter A, you match it to what you have stored in your LTM memory (mini copy or template in LTM of all known patterns) and when you get a good enough match, you can recognise that as a letter A

This is known as…?

A

Template theories on 2D Pattern matching

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

How do we engage in 2D Pattern matching?

List 4 points

A
  1. Template theories
  2. Prototype theories
  3. Feature theories
  4. Structural descriptions
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

What are the 2 implications of template theories?

A
  1. Does not consider normalisation
  2. How many mini copies or templates of the object do we store in the LTM? Is it infinite?
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

Give 2 examples of template theory 2D pattern matching

A
  1. Barcodes
  2. Fingerprints
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

Multiple templates are held in memory, we compare stimuli to templates in memory for one with the greatest overlap until a match is found

This is known as…?

A

Template theories on 2D Pattern matching

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

What are the 3 limitations of template theories?

A
  1. Problem of imperfect matches
  2. Cannot account for the flexibility of pattern recognition system
  3. Comparison requires identical orientation, size, position of template to stimuli
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

How do we recognise the letter ‘A’ according to prototype theories?

A

Rather than storing lots of different templates, we have an average template stored in memory. When you see your letter A, you match it to the average template (prototype) stored in your memory.

Simply = Modification of template matching (flexible templates), Possesses the average of each individual characteristic, No match is perfect; a criterion for matching is needed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
33
Q

Rather than storing lots of different templates, we have an average template stored in memory. When you see your letter A, you match it to the average template (prototype) stored in your memory.

This is known as…?

A

Prototype theories on 2D Pattern matching

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
34
Q

Modification of template matching (flexible templates), Possesses the average of each individual characteristic, No match is perfect; a criterion for matching is needed

This is known as…?

A

Prototype theories on 2D Pattern matching

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
35
Q

Describe Franks & Bransford’s (1971) evidence for prototype theory

List 4 points

A
  1. Presented objects based on prototypes
  2. Prototype not shown
  3. Yet participants are confident they have seen the prototype
  4. Suggests the existence of prototypes

Simply = By seeing all these examples that all surround this prototype, you think as though you have seen the prototype even though you have not seen it at all

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
36
Q

Which 2D pattern matching theory does this apply to?

Pattern consists of a set of features or attributes

a. Template theories
b. Prototype theories
c. Feature theories
d. Structural descriptions

A

c. Feature theories

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
37
Q

Which 2D pattern matching theory does this apply to?

Modification of template matching (flexible templates)

a. Template theories
b. Prototype theories
c. Feature theories
d. Structural descriptions

A

b. Prototype theories

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
38
Q

Which 2D pattern matching theory does this apply to?

Possesses the average of each individual characteristic

a. Template theories
b. Prototype theories
c. Feature theories
d. Structural descriptions

A

b. Prototype theories

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
39
Q

Which 2D pattern matching theory does this apply to?

Mini copy or template in LTM of all known patterns

a. Template theories
b. Prototype theories
c. Feature theories
d. Structural descriptions

A

a. Template theories

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
40
Q

Which 2D pattern matching theory does this apply to?

Multiple templates are held in memory

a. Template theories
b. Prototype theories
c. Feature theories
d. Structural descriptions

A

a. Template theories

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
41
Q

Which 2D pattern matching theory does this apply to?

No match is perfect; a criterion for matching is needed

a. Template theories
b. Prototype theories
c. Feature theories
d. Structural descriptions

A

b. Prototype theories

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
42
Q

Which 2D pattern matching theory does this apply to?

Compare stimuli to templates in memory for one with greatest overlap until a match is found

a. Template theories
b. Prototype theories
c. Feature theories
d. Structural descriptions

A

a. Template theories

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
43
Q

Which 2D pattern matching theory does this apply to?

Problem of imperfect matches

a. Template theories
b. Prototype theories
c. Feature theories
d. Structural descriptions

A

a. Template theories
b. Prototype theories

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
44
Q

Which 2D pattern matching theory does this apply to?

Cannot account for the flexibility of pattern recognition system

a. Template theories
b. Prototype theories
c. Feature theories
d. Structural descriptions

A

a. Template theories

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
45
Q

Which 2D pattern matching theory does this apply to?

Comparison requires identical orientation, size, position of template to stimuli

a. Template theories
b. Prototype theories
c. Feature theories
d. Structural descriptions

A

a. Template theories

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
46
Q

How do we recognise the letter ‘A’ according to feature theories?

A
  • Pattern consists of a set of features or attributes
  • A = 2 straight lines and a connecting crossbar.
  • But also need to know the relationship between features
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
47
Q
  • Pattern consists of a set of features or attributes
  • A = 2 straight lines and a connecting crossbar.
  • But also need to know the relationship between features

This is known as…?

A

Feature theories on 2D pattern matching

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
48
Q

How do we recognise the letter ‘A’ according to structural descriptions?

A
  • Describe the nature of the components of a configuration and the structural arrangement of these parts” (Bruce & Green, 1990)
  • Capital letter T = 2 parts; 1 horizontal; 1 vertical; vertical supports horizontal; vertical bisects horizontal
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
49
Q
  • Describe the nature of the components of a configuration and the structural arrangement of these parts” (Bruce & Green, 1990)
  • Capital letter T = 2 parts; 1 horizontal; 1 vertical; vertical supports horizontal; vertical bisects horizontal

This is known as…?

A

Structural descriptions on 2D pattern matching

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
50
Q

How do we recognise 3D objects?

List 2 steps

A
  1. Firstly we must interpret the input to the visual system as coherent structures, segregated from one another and from the background (early image processing)

simply = Must know how to segregate an object as being a distinct one object that goes together and is distinct from the background

  1. Must be processed to give a description– which can then be matched to the descriptions of visual objects stored in memory.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
51
Q

What does Marr’s Computational Approach aim to explain?

A

3D Object recognition

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
52
Q

3D Object recognition can be explained by which approach?

A

Marr’s Computational Approach

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
53
Q

What are the 3 steps to Marr’s Computational Approach

A
  1. Primal Sketch
  2. 2 1⁄2-D Sketch
  3. 3-D Representation
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
54
Q

What is a primal sketch in Marr’s Computational Approach?

A

2-D description includes changes in light intensity, edges, contours, blobs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
55
Q

What is a 2 1/2 D Sketch in Marr’s Computational Approach?

A

Includes information about depth, motion, shading. Representation is observer-centered.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
56
Q

What is a 3D representation in Marr’s Computational Approach?

A

A representation of objects and their relationships, observer-independent

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
57
Q

2-D description includes changes in light intensity, edges, contours, blobs

This is known as…?

A

Primal sketch

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
58
Q

Includes information about depth, motion, shading. Representation is observer-centered.

This is known as…?

A

2 1/2 D Sketch

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
59
Q

A representation of objects and their relationships, observer-independent

This is known as…?

A

3D representation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
60
Q

What are the 4 questions asked when recognising objects?

A
  1. What elements are used in the description? (primitives)
  2. How is the relationship between these elements specified?
  3. How is the overall description invariant across views?

simply = How can you recognise objects from different viewpoints? How is it that we can create a description that works from different viewpoints

  1. What about viewpoint dependence?

simply = Does it depend on the task that you are doing?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
61
Q

What did Marr & Nishihara (1978) express about objects in relation to cylinders?

A

Every object can be broken down into a series of cylinders and we must specify the relationship between cylinders to create a structural description of the object

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
62
Q

Every object can be broken down into a series of cylinders and we must specify the relationship between cylinders to create a structural description of the object

Who proposed this?

A

Marr & Nishihara (1978)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
63
Q

The relationship between these cylinders creates a structural description of the object

This works particularly well what kinds of objects?

A

Biological type objects
e.g. human body

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
64
Q

Marr & Nishihara (1978) expressed structural relations by …?

A

A hierarchical organisation of cylinders

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
65
Q

According to Marr & Nishihara (1978), structural relations are based on a hierarchical organisation of cylinders

What does this mean?

A

Each cylinder has an axis and the way in which others are joined is expressed as coordinates

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
66
Q

The position of each cylinder is described relative to its own axis, resulting in a description which is invariant across viewpoints.

This is useful to determine what types of objects?

A

Argued to be useful in determining who is a human or gorilla, depending on the length of the arms, etc.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
67
Q

The position of each cylinder is described relative to its own axis, resulting in a description which is invariant across viewpoints.

This is not useful to determine what types of objects?

A

Other objects that are not biological/bodily types of objects

e.g. a piece of paper cannot be broken down into cylinders

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
68
Q

Works well for biological objects but not other objects

Which model does this apply to?

A

Marr & Nishihara’s (1978) cylinder model

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
69
Q

Who provided an alternative model to Marr & Nishihara (1978)?

A

Biederman (1987;1989)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
70
Q

Which theory proposes that objects are composed of basic shapes?

A

Biederman’s Recognition-by-components theory

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
71
Q

What does Biederman’s Recognition-by-components theory propose about objects?

A

Objects composed of basic shapes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
72
Q

According to Biederman’s Recognition-by-components theory, objects are composed of basic shapes known as…?

A

GEONS

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
73
Q

What are GEONS?

A

Geometrical ions

Basic shapes that make up an object

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
74
Q

Give 4 examples of GEONS

A
  1. Blocks
  2. Cylinders
  3. Arcs
  4. Wedges
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
75
Q
  1. Blocks
  2. Cylinders
  3. Arcs
  4. Wedges

These are examples of…?

A

GEONS

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
76
Q

Approximately how many different volumetric shapes (GEONS) are there based on Biederman’s Recognition-by-components theory?

A

36

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
77
Q

Which model/theory is viewpoint invariant?

A

Biederman’s Recognition-by-components theory

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
78
Q

Which model/theory proposes that objects are made up of cylinders?

A

Marr & Nishihara (1978)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
79
Q

Which model/theory proposes that objects are made up of GEONS?

A

Biederman’s Recognition-by-components theory

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
80
Q

Which model/theory proposes that there are small numbers of structural relationships?

A

Biederman’s Recognition-by-components theory

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
81
Q

Biederman’s Recognition-by-components theory claims that there is a small number of structural relationship

Give 4 examples

A
  1. Relative size
  2. Verticality
  3. Centring
  4. Relative size of surfaces at join
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
82
Q

Viewpoint-invariant theory of recognition is known as…?

A

Biederman’s Recognition-by-components theory

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
83
Q

Recognition using 3D component parts (e.g. 36 GEONS), with structural relations between the parts

Which model/theory proposes this?

A

Biederman’s Recognition-by-components theory

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
84
Q

Define viewpoint-invariant

A

The object can be identified from any angle

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
85
Q

The object can be identified from any angle

This is known as…?

A

Viewpoint-invariant

86
Q

According to Biederman (1987; 1989), what is helpful in segmenting visual image into parts?

A

Concave parts of an object’s contour

87
Q

According to Biederman (1987; 1989), concave parts of an object’s contour is helpful in…?

A

Segmenting visual image into parts

88
Q

Geons are specified in terms of…?

A

‘Non-accidental’ properties

89
Q

Are specified in terms of ‘non-accidental’ properties

This is known as…?

90
Q

GEONS are specified in terms of ‘non-accidental’ properties

What are they? List 5

A
  1. Curvature
  2. Parallel
  3. Co-termination
  4. Symmetry
  5. Co-linearity
91
Q
  1. Curvature
  2. Parallel
  3. Co-termination
  4. Symmetry
  5. Co-linearity

These are properties of…?

92
Q

How do we recognise an object using GEONS?

List 4 points

A
  1. Pick out one GEON from another in terms of points of concavity
  2. Look at the relationships between the GEONS, which gives you structural descriptions of an object
  3. Once you have created that structural description, you try to match it to all of the structural descriptions you have stored in your memory
  4. Once you find a good enough match, you recognise it as an object (e.g. lamp)
93
Q

How do we recognise a cylinder using GEONS?

List 2 points

A
  1. Cylinder posses curved edges & two parallel edges connecting the curved edges
  2. Regularities in the visual image thought to reflect actual (non-accidental) regularities in the world

e.g. 2D symmetry in the visual image indicates symmetry in 3D object.

94
Q

GEONS are specified in terms of ‘non-accidental’ properties such as curvature

What is curvature?

A

Points on a curve

95
Q

GEONS are specified in terms of ‘non-accidental’ properties such as parallel

What is Parallel?

A

Set of points in parallel

96
Q

GEONS are specified in terms of ‘non-accidental’ properties such as co-termination

What is co-termination?

A

Edges terminating in a common point

97
Q

GEONS are specified in terms of ‘non-accidental’ properties such as symmetry

What is Symmetry?

A

Versus asymmetry

98
Q

GEONS are specified in terms of ‘non-accidental’ properties such as co-linearity

What is co-linearity?

A

Points in a straight line

99
Q

List these steps of recognising object using GEONS in the correct order, according to Biederman (1987)

DETERMINATION OF COMPONENTS

EDGE EXTRACTION

DETECTION OF NON-ACCIDENTAL PROPERTIES

MATCHING OF COMPONENTS TO OBJECT REPRESENTATIONS

PARSING OF REGIONS OF CONCAVITY

A
  1. EDGE EXTRACTION
  2. DETECTION OF NON-ACCIDENTAL PROPERTIES

and

PARSING OF REGIONS OF CONCAVITY

  1. DETERMINATION OF COMPONENTS
  2. MATCHING OF COMPONENTS TO OBJECT REPRESENTATIONS
100
Q

According to Biederman’s (1987; 1989) model, what makes objects more difficult to recognise?

A

Forms of degradation which disrupt the basis for identifying geons

101
Q

According to Biederman’s (1987; 1989) model, what effect do forms of degradation which disrupt the basis for identifying geons have on object recognition?

A

It makes objects more difficult to recognise

102
Q

According to Biederman’s (1987; 1989) model, why do forms of degradation lead to difficulty in recognising objects?

A

They disrupt the basis for identifying geons

103
Q

According to Biederman’s (1987; 1989) model, what disrupt the basis for identifying geons?

A

Forms of degradation

104
Q

Describe Biederman’s (1987) study on forms of degradation

List 2 points

A
  1. Biederman (1987) deleted edges at points where easily reinstated or difficult to determine
  2. Stimuli was presented for 100, 200 or 750 msec with 25%, 45% or 65% contours removed
105
Q

Describe the results of Biederman’s (1987) study on forms of degradation

List 2 points

A
  1. Ps were relatively good at recognising objects with deleted edges at points where easily reinstated or recognisable objects (partial degradation)
  2. Ps were slow and inaccurate at recognising objects with deleted edges at points where it was difficult to determine or non-recognisable objects
106
Q

Ps were relatively good at recognising objects with deleted edges at points where …?

a. Easily reinstated
b. Difficult to determine

A

a. Easily reinstated

107
Q

Ps were slow and inaccurate at recognising objects with deleted edges at points where …?

a. Easily reinstated
b. Difficult to determine

A

b. Difficult to determine

108
Q

According to Biederman (1987), what does deletion of component affect?

A

Matching stage

simply = It reduces the number of components to match to

109
Q

What reduces the number of components to match to?

A

Deletion of components of an object

110
Q

According to Biederman (1987), what type of deletion makes it more difficult to determine components?

A

Midsegment deletion

111
Q

True or False?

At brief exposures
(65ms) partial objects are better recognised

112
Q

At brief exposures
(65ms) _____ objects better recognised

A

Partial objects

113
Q

True or False?

At longer exposures (200ms) midsegment deletion led to
more errors

A

False

At longer exposures (200ms) midsegment deletion led to
less errors

114
Q

At longer exposures (200ms) ____ led to less errors

A

Midsegment deletion

115
Q

Describe Vogel et al.’s (2001) results supporting Biederman’s model

List 3 points

A
  1. Found some cortical neurons in monkeys sensitive to geons.
  2. Assessed response of individual neurons in the inferior temporal cortex to change in geon or change in size of object
  3. Some neurons responded more to geon changes, providing support for geons
116
Q

According to Vogel et al. (2001), what neurons in monkeys are sensitive to geons?

A

Cortical neurons

117
Q

According to Vogel et al. (2001), some neurons responded more to…?

a. Change in size of object
b. Change in geons

A

b. Change in geons

118
Q

According to Vogel et al. (2001), some neurons responded more to geon changes

What claim does this support?

A

Support for the existence of geons

119
Q

What are the 3 evaluations of Biederman’s (1987; 1989) model?

A
  1. Why 36 geons?
  2. Experimental results were consistent with the model but doesn’t provide critical test
  3. It doesn’t explain how descriptions are matched to those stored in memory
120
Q

What are 2 advantages of Biederman’s (1987; 1989) model?

A
  1. Recognises the importance of the arrangement of the parts

simply = It is not just the components that are important but also the relationships between the components

  1. Parsimonious: Small set of primitive shapes
121
Q

It is a flexible and comprehensive system for describing objects

This applies to which theory/model?

A

Biederman’s (1987; 1989) model

122
Q

It doesn’t explain how descriptions are matched to those stored in memory

This applies to which theory/model?

A

Biederman’s (1987; 1989) model

123
Q

Experimental results were consistent with the model but doesn’t provide critical test

This applies to which theory/model?

A

Biederman’s (1987; 1989) model

124
Q

Only tells us about how we create structural descs of the object and not how we match it to the descs stored in our memory

This applies to which theory/model?

A

Biederman’s (1987; 1989) model

125
Q

Recognises the importance of the arrangement of the parts

simply = It is not just the components that are important but also the relationships between the components

This applies to which theory/model?

A

Biederman’s (1987; 1989) model

126
Q

Parsimonious: Small set of primitive shapes

This applies to which theory/model?

A

Biederman’s (1987; 1989) model

127
Q

What are the 5 disadvantages of Biederman’s (1987; 1989) model?

A
  1. Structure is not always key to recognition: Peach vs. Nectarine

simply = Basing object recognition on edges (point of concavity)

They both have the same edge, so how do we discriminate between them?

  1. Which geons?

simply = Does not explain which geons are used for recognising different objects

  1. Does not explain within category discrimination (which chair?)
  2. De-emphasise the role played by context in object recognition (affects later stages of object recognition)
  3. Simplifies the contribution of viewpoint- dependence

simply = It is easier to recognise objects from one viewpoint to another but this perspective does not take this into account

128
Q

Structure is not always key to recognition: Peach vs. Nectarine

simply = Basing object recognition on edges (point of concavity)

They both have the same edge, so how do we discriminate between them?

This applies to which theory/model?

A

Biederman’s (1987; 1989) model

129
Q

Does not explain which geons are used for recognising different objects

This applies to which theory/model?

A

Biederman’s (1987; 1989) model

130
Q

Does not explain within category discrimination (which chair?)

This applies to which theory/model?

A

Biederman’s (1987; 1989) model

131
Q

De-emphasise the role played by context in object recognition (affects later stages of object recognition)

This applies to which theory/model?

A

Biederman’s (1987; 1989) model

132
Q

Simplifies the contribution of viewpoint- dependence

simply = It is easier to recognise objects from one viewpoint to another but this perspective does not take this into account

This applies to which theory/model?

A

Biederman’s (1987; 1989) model

133
Q

What do viewpoint invariance theories argue?

A

The ease of object recognition is not affected by the observer’s viewpoint

simply = being able to recognise an object is not affected by viewpoint

134
Q

What do viewpoint dependent theories argue?

A

Assume changes in viewpoint reduce the speed and / or accuracy of object recognition

135
Q

Assume changes in viewpoint reduce the speed and / or accuracy of object recognition

a. view point invariance theories

b. view point-dependent theories

A

b. view point-dependent theories

136
Q

The ease of object recognition if not affected by the observer’s viewpoint

simply = being able to recognise an object is not affected by viewpoint

a. view point invariance theories

b. view point-dependent theories

A

a. view point invariance theories

137
Q

According to view point invariance theories (Biederman, 1987) ease of object recognition if not affected by the observer’s viewpoint

Why?

A

Because you are creating an invariant desc which is invariant across views

138
Q

Give one example of viewpoint invariance theories

A

Biederman (1987

139
Q

Give two examples of viewpoint invariance theories

A
  1. Tarr (1995)
  2. Tarr & Bulthoff (1995; 1998)
140
Q

e.g. you normally see a bowl as being upright but if you see it being tilted to the side on it’s back, you might take longer to recognise it as a bowl

a. view point invariance theories

b. view point-dependent theories

A

b. view point-dependent theories

141
Q

According to viewpoint dependent theories, object representations are…?

A

Collections of views that depict the appearance of objects from specific viewpoints

142
Q

Object representations are collections of views that depict the appearance of objects from specific viewpoints

a. view point invariance theories

b. view point-dependent theories

A

b. view point-dependent theories

143
Q

Evidence suggests that viewpoint invariant mechanisms are used sometimes in

a. object recognition
b. other times other than object recognition

A

a. object recognition

144
Q

Evidence suggests that viewpoint dependent mechanisms used at…?

a. object recognition
b. other times other than object recognition

A

b. other times other than object recognition

145
Q

Evidence suggests that ____ used sometimes in object recognition

a. viewpoint dependent mechanisms

b. viewpoint invariant mechanisms

A

b. viewpoint invariant mechanisms

146
Q

Evidence suggests that ____ used used at other times other than object recognition

a. viewpoint dependent mechanisms

b. viewpoint invariant mechanisms

A

a. viewpoint dependent mechanisms

147
Q

Which mechanism is better for within category discriminations (e.g. different kinds of car)?

a. viewpoint dependent mechanisms

b. viewpoint invariant mechanisms

A

a. viewpoint dependent mechanisms

148
Q

Viewpoint dependent are more important for _____ discriminations

a. Within category
b. Different category
c. None of the above

A

a. Within category

149
Q

True or False?

The key question is no longer if object recognition is viewpoint-dependent or viewpoint independent, but rather when, i.e. under which circumstances

150
Q

Which mechanism is more useful for complex within category decisions?

a. viewpoint dependent mechanisms

b. viewpoint invariant mechanisms

A

a. viewpoint dependent mechanisms

151
Q

Which mechanism is more useful for easy categorical decisions?

a. viewpoint dependent mechanisms

b. viewpoint invariant mechanisms

A

b. viewpoint invariant mechanisms

152
Q

Which mechanism is more useful for this task:

e.g. This is my laptop, this is your laptop

a. viewpoint dependent mechanisms

b. viewpoint invariant mechanisms

A

a. viewpoint dependent mechanisms

153
Q

Which mechanism is more useful for this task:

e.g. Is it a chair, is it a table?

a. viewpoint dependent mechanisms

b. viewpoint invariant mechanisms

A

b. viewpoint invariant mechanisms

154
Q

True or False?

Object representations are only viewpoint-dependent

A

False

Object representations are neither viewpoint-dependent nor viewpoint- invariant

There is a middle ground where both of those things play a role

155
Q

What are the issues with object recognition theories?

List 2 points

A
  1. Any theory of object recognition must address the binding problem = how do we integrate different kinds of information to produce object recognition?
  2. When presented with several objects how do we decide which features or geons belong to which object?
156
Q

What should happen once structural description of an object is formed?

A

It must be matched to stored representations

157
Q

If there is a match between stored representations and structural description of an object, then the object is …?

A

‘Recognised’

158
Q

If there is a match between ____ and ____, then the object is recognised

A
  1. Stored representations
  2. Structural description of an object
159
Q

What are the 5 stages of the recognition process, according to Humphreys et al (1988)?

A
  1. OBJECT
  2. STRUCTURAL DESCRIPTION
  3. SEMANTIC REPRESENTATION
  4. NAME REPRESENTATION
  5. NAME
160
Q

What patients may struggle to create a structural description of an object?

A

Agnosic patients

161
Q

Agnosic patients may struggle to create a structural description of an object

What are the consequences of this?

A

They can’t match the structural descriptions to the representations stored in memory and the objects don’t seem familiar to them

162
Q

What is the main limitation of Humphreys et al.’s (1988) stages of object recognition?

A

Oversimplification

‘Later’ processes may start before earlier ones have been completed

163
Q

What is the main advantage of Humphreys et al.’s (1988) stages of object recognition process?

A

Provides general support for model from patients with object recognition difficulties

e.g. Associative agnosia
Patient HJA; Patient JB

164
Q

Provides general support for model from patients with object recognition difficulties

e.g. Associative agnosia
Patient HJA; Patient JB

Which theory/model does this apply to?

A

Humphreys et al.’s (1988) stages of object recognition process

165
Q

Oversimplification

‘Later’ processes may start before earlier ones have been completed

Which theory/model does this apply to?

A

Humphreys et al.’s (1988) stages of object recognition process

166
Q

What are the 4 stages of object recognition process, according to Humphreys et al.’s (1988) alternative model (‘Cascade’ model’)

A
  1. STRUCTURAL DESCRIPTION SYSTEM
  2. SEMANTIC REPRESENTATIONS
  3. NAME REPRESENTATIONS
  4. NAME
167
Q

According to Humphreys et al.’s (1988) alternative model (‘Cascade’ model’) of object recognition, ____, ____ and ____ interact

A
  1. Structural
  2. Semantic
  3. Name stages
168
Q

What are the 3 characteristics of Humphreys et. al.’s (1988) ‘Cascade’ model’?

A
  1. Structural, semantic and name stages interact
  2. Both within and between stages
  3. Makes different predictions about how subjects will perform in object naming task.

Problems at one stage will have ‘knock on’ effect

169
Q

Which theory/model does this apply to?

Makes different predictions about how subjects will perform in object naming task.

Problems at one stage will have ‘knock on’ effect

A

Humphreys et al.’s (1988) ‘Cascade’ model

170
Q

Which theory/model does this apply to?

Structural, semantic and name stages interact both within and between stages

A

Humphreys et al.’s (1988) ‘Cascade’ model

171
Q

List these stages in the correct order, according to Humphreys et al.’s (1988) ‘Cascade’ model

(Apple) edible crunchy sweet
Apple
(Apple) round stalk
Object = ‘Apple’

A
  1. (Apple) round stalk
  2. (Apple) edible crunchy sweet
  3. Apple
  4. Object = ‘Apple’
172
Q

There are ___ and ___ evidence for a separation of structural, semantic and naming processes in recognition

A
  1. Anecdotal
  2. Empirical
173
Q

Anecdotal and empirical evidence for a separation of ____, _____ and ____ in recognition

A
  1. Structural
  2. Semantic
  3. Naming processes
174
Q

_____ propose processing across these stages operates in cascade rather than independently

A

Humphreys et. al. (1988)

175
Q

Humphreys et. al. (1988) propose processing across these stages operates in ____ rather than independently

A

Cascade

simply = Interaction between the different stages

176
Q

Humphreys et. al. (1988) propose processing across these stages operates in cascade rather than independently

Give an example of this

A

e.g. Patient JB.

Naming visually confusable objects (birds, animals) had knock on effects, making it more difficult to identify their category

simply = If you get stuck on one stage, you can’t move on to later stages

177
Q

e.g. Patient JB.

Naming visually confusable objects (birds, animals) had knock on effects, making it more difficult to identify their category

simply = If you get stuck on one stage, you can’t move on to later stages

Which theory/model supports this?

A

Humphreys et. al. (1988) cascade model

178
Q

Define agnosia

A

Failure of knowledge or recognition

179
Q

Failure of knowledge or recognition

This is known as…?

180
Q

In visual agnosias, ___ and ___ remain intact

A
  1. Feature processing
  2. Memory
181
Q

When feature processing and memory remain intact, and recognition deficits are limited to the visual modality

This is known as…?

A

Visual agnosias

182
Q

Inability to recognise objects from seeing that object

This is known as…?

A

Visual agnosias

183
Q

Define visual agnosia

A

Inability to recognise objects from seeing that object

184
Q

In visual agnosias, recognition deficits are limited to …?

A

The visual modality

185
Q

True or False?

In visual agnosia, alertness, attention, intelligence and language are affected

A

False

In visual agnosia, alertness, attention, intelligence and language are unaffected

186
Q

In visual agnosia, what may substitute for vision in allowing objects to be recognised?

A

Other sensory modalities (touch, smell)

187
Q

Define Apperceptive agnosia

A

Problems with early processing (shape extraction)

188
Q

Problems with early processing (shape extraction)

This is known as…?

A

Apperceptive agnosia

189
Q

All objects look similar, unable to tell the difference, low level perceptual deficit

This is an example of…?

A

Apperceptive agnosia

190
Q

What causes Apperceptive agnosia?

A

Perceptual deficit

191
Q

Describe how perceptual deficit causes Apperceptive agnosia

List 4 points

A
  1. Perceptual deficit affects visual representations directly
  2. Components of visual percept are picked up, but can’t be integrated
  3. Effects may be graded, often affected
  4. Results in unusual views of objects
192
Q
  1. Perceptual deficit affects visual representations directly
  2. Components of visual percept are picked up, but can’t be integrated
  3. Effects may be graded, often affected
  4. Results in unusual views of objects

What type of agnosia is this?

A

Apperceptive agnosia

193
Q

Define Associative agnosia

A

Problems with later processing (recognition)

194
Q

Problems with later processing (recognition)

A

Associative agnosia

195
Q

What causes Associative agnosia?

A

Lack of information about the percept

simply = lack of info, high level perceptual deficit

196
Q

Visual representations are intact, but cannot be accessed or used in recognition. Lack of information about the percept.

What type of agnosia is this?

A

Associative agnosia

197
Q

Feature processing and memory remain intact

a. Associative agnosia
b. Visual agnosias
c. Apperceptive agnosia
d. None of the above

A

b. Visual agnosias

198
Q

Perceptual deficit, affects visual representations directly

a. Associative agnosia
b. Visual agnosias
c. Apperceptive agnosia
d. None of the above

A

c. Apperceptive agnosia

199
Q

Visual representations are intact, but cannot be accessed or used in recognition

a. Associative agnosia
b. Visual agnosias
c. Apperceptive agnosia
d. None of the above

A

a. Associative agnosia

200
Q

Recognition deficits are limited to the visual modality

a. Associative agnosia
b. Visual agnosias
c. Apperceptive agnosia
d. None of the above

A

b. Visual agnosias

201
Q

Lack of information about the percept

a. Associative agnosia
b. Visual agnosias
c. Apperceptive agnosia
d. None of the above

A

a. Associative agnosia

202
Q

Other sensory modalities (touch, smell) may substitute for vision in allowing objects to be recognised

a. Associative agnosia
b. Visual agnosias
c. Apperceptive agnosia
d. None of the above

A

b. Visual agnosias

203
Q

Components of visual percept are picked up, but can’t be integrated

a. Associative agnosia
b. Visual agnosias
c. Apperceptive agnosia
d. None of the above

A

c. Apperceptive agnosia

204
Q

Alertness, attention, intelligence and language are unaffected

a. Associative agnosia
b. Visual agnosias
c. Apperceptive agnosia
d. None of the above

A

b. Visual agnosias

205
Q

Effects may be graded, often affected: unusual views of objects

a. Associative agnosia
b. Visual agnosias
c. Apperceptive agnosia
d. None of the above

A

c. Apperceptive agnosia

206
Q

Inability to recognise objects from seeing that object

a. Associative agnosia
b. Visual agnosias
c. Apperceptive agnosia
d. None of the above

A

b. Visual agnosias

207
Q

Lack of info, high level perceptual deficit

a. Associative agnosia
b. Visual agnosias
c. Apperceptive agnosia
d. None of the above

A

a. Associative agnosia

208
Q

All objects look similar, unable to tell the difference, low level perceptual deficit

a. Associative agnosia
b. Visual agnosias
c. Apperceptive agnosia
d. None of the above

A

c. Apperceptive agnosia

209
Q

Agnosia is useful for studying …?

A

Object recognition

210
Q

True or False?

Agnosia may be restricted to specific categories

211
Q

True or False?

Agnosia can only be found alone

A

False

Agnosia may be found alone or with problems with faces.

YEAR 2 Perception (6 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions