Object Recognition

Question 1

What regions of the brain are used in the “what” system?

Answer 1

The ventral visual-processing stream: Occipital, occipitotemporal and temporal regions.

Question 2

Why is the large receptive field of the ventral visual-processing stream helpful? What is it lacking?

Answer 2

It allows for object recognition regardless of size, but spatial information is lost.

Question 3

What visual-processing stream is sensitive to colour?

Answer 3

The ventral visual-processing stream. It allows us to separate foreground from background.

Question 4

How does the ventral visual-processing stream work?

Answer 4

It moves from the posterior of the brain to the anterior, working from simple visual process to complex.

Question 5

What is visual agnosia?

Answer 5

The inability to recognise objects in a visual modality.

Question 6

What is apperceptive visual agnosia?

Answer 6

When you see parts of an object put you can’t see it as a whole.

Question 7

What causes apperceptive visual agnosia?

Answer 7

Damage to the occipital areas, caused by strokes, anoxia or carbon monoxide poisoning.

Question 8

What is associative visual agnosia?

Answer 8

When you can see the whole object but you do not know what it is.

Question 9

What causes associative visual agnosia?

Answer 9

Bilateral damage to the inferior temporo-occipital junction and adjacent white matter.

Question 10

What is prosopagnosia? What causes it?

Answer 10

The inability to recognise or differentiate among faces. Damage to the right fusiform face gyrus is involved with this.

Question 11

What is perceptual invariance?

Answer 11

The ability to recognize objects regardless of orientation.

Question 12

What is sparse coding?

Answer 12

It is the theory that there are certain groups of cells that are coded to recognise a particular object (The Grandmother Cell theory)

Question 13

What is the issue with sparse coding as a theory?

Answer 13

It doesn’t account for the recognition of objects only seen later in life, and it doesn’t account for what happens to the cells after the object is removed from the viewers world.

Question 14

What is population coding?

Answer 14

It’s a theory that states that all brain cells are used in the recognition of all objects. There is a unique pattern for each object.

Question 15

How do they test the sparse vs. population theory?

Answer 15

Recording activity in random inferotemporal cells in monkeys, whilst showing them pictures from various different categories.

Question 16

What did Matsumoto find in 2005?

Answer 16

The study found that cells fired twice - once to establish category of the object and then again to determine the object itself.

Question 17

What is form-cue invariance?

Answer 17

The brain categorisation is constant regardless of the form of the cue that represents the object. (Apple, cartoon apple, eaten apple etc.)

Question 18

Briefly explain the Adaptation Method with regard to perceptual consistency.

Answer 18

The brain’s response to objects decreases the more it is exposed to the object. This is true even for cases where the object is shown from different angles.

Question 19

What part of the brain is in play with the adaptation method?

Answer 19

Lateral occipital complex

Question 20

What are the two theories for the adaptation method?

Answer 20

Theory 1: Your brain creates a 3D view from the 2D view it receives, and this view continues to grow with more and more exposure to the object.
Theory 2: There is no 3D image formed, but the brain extrapolates an image through systematic integration of viewer-centered representations.

Question 21

What is feature-based coding?

Answer 21

When you look at all the features of an object to make up the whole.

Question 22

What is configural coding?

Answer 22

When you look at the object as a whole.

Question 23

What kind of information does the left hemisphere of the ventral stream deal with?

Answer 23

Local perception and object features.

Question 24

What kind of information does the right hemisphere of the ventral stream deal with?

Answer 24

Global perception and holistic processing.

Question 25

What do first-order spatial relations entail?

Answer 25

What is on top and what is on the bottom.

Question 26

What do second-order spatial relations entail?

Answer 26

Distances between features and the size of the features.

Question 27

What is the inversion effect? What causes it?

Answer 27

It is when recognition is made difficult due to the picture being upside down. It happens because the inversion prevents configural coding so we are forced to use only feature-based coding.

Question 28

What do the specific neural modules within the ventral stream specialize in?

Answer 28

Fusiform face area - faces
Parahippocampal place area - places in local environment
Extrastriate body area - human bodies and body parts

Question 29

What are the four different regions involved in facial recognition and their functions?

Answer 29

1. Right Fusiform Gyrus and Parahippocampal Gyrus - face identification
2. Occipital Face Area - face detection and identification
3. Amygdala - processes facial emotions
4. Superior Temporal Sulcus - sensitive to social perception.