Dorsal _____ pathway= ________ and _______ ( _?type_ processing)

Ventral _____ pathway = _______ ( _?type_ processing)

Dorsal parietal pathway= where and how (spatial processing)

Ventral temporal pathway= what (object processing)

Evidence for what-where distinction in the brain: lesion studies...

Monkeys were trained to perfrom either _____ task or ______ task.

After training, lesion made in either __1__ lobe or __2__ lobe.

___1___ resulted in...?

___2___ resulted in...?

Landmark discrimination or object discrimination.

Temporal or parietal lobe.

Temporal lobe lesions showed much impairment on object discrimination task and little on landmark.

 Parietal lobe lesions showed much impairment on landmark discrimination task and little on object.

Evidence for what-where distinction in the brain: Electrophysiological studies...

______ neurons suited to "where" detection.

_% of neurons have receptive fields outside of the fovea.

 Respond (same/differently) to large and small stimuli... Why is that helpful?

Parietal

60%

Same... Perfect for detecting stimuli that have just entered field of view.

What and where processing differ in their dependence on _______ processing.

Evidence for what-where distinction in the brain: Electrophysiological studies...

______ neurons suited to "what" detection.

Receptive fields include _____ and _____.

Most respond to (complex objects/lines and spots)?

Respond to ______ (type of stimuli and orientation)

Temporal

 Fovea and both visual fields.

Complex Objects

Very specific stimuli, across different orientations (view-invariant)

Evidence for what-where distinction in the brain: Neuropsychological Studies

Patient D.F.: _____ visual acuity

impaired "_____", intact "_____"

Can:_____

Cannot:_____

When asked to match orientation of card to that of the slot (perception, recall, and action condition)... Impaired or intact?

normal

impaired ventral "what", intact "where"

Can: describe object if given name... name and describe objects when placed in the hand

 Cannot: recognize objects from pictures or displayed objects... describe displayed objects

Perception Condition: Impaired

Recall and Action Condition: Intact 

Disorders

_____: can recognize objects (what) but cannot use visual information to guide their action toward location processing (where)

______: impaired object recognition in the visual modality (what) but intact location processing (where)

Optic Ataxia

Visual Agnosia

Evidence for what-where distinction in the brain:

Brain Imaging Studies...

Dorsal and ventral pathways communicate (rarely/extensively)?

This kind of information can be used to identify components of a visual scene.

_____ cortex is used for spatial localization as well as other functions (selective attention).

Communicate Extensively

Spatial information or depth

Parietal Cortex

_____ _____: the ability to recognize an object despite all the variability in the image

3 Forms of Variability 

Object constancy

Variability: illumination and shadows, occlusion, different viewpoints

Theories of Object Recognition

View-Dependent: 3 Factors

View-Invariant: 3 Factors

View-dependent:

1. Recognize an object by matching it to a stored representation from the perceived view

2. Requires that we have multiple representations of an object, each from a different view

3. Places a large burden on perceptual memory

View-Invariant:

1. Use basic sensory information to determine the main features of an object from any view

2. Example of Bike, the major axis of the bike runs along its length while the handle bars form the minor axis

3. The main features are visible from almost all views.

fMRI data supporting each theory

fMRI response is _____ for repeated than for novel stimuli (repetition suppression)

Match: view invariant theories and view dependent theories

_____: predict repitition suppression in object processing areas regardless of whether the same view is repeated

______: predict repitition suppression in object processing areas only when the exact view is repeated.

smaller

view invariant theories 

view dependent theories

Hierarchical Coding Hypothesis _________

Grandmother Cells ________, Pros / Cons

Elementary features are combined to create units that recognize complex objects

An individual cell might respond only to your grandmother or to a certain object

Pros: Fits with the findings that temporal lobe neurons respond to highly specific stimuli like hands and faces

Cons: What if this cell died? How does this cell respond to grandma as she ages? We would need a bunch of cells to represent all the objects and people we recognize!

Ensemble Coding

4 Feature Detectors

3 Strengths

Object recognition is enabled by the simultaneous activation of several complex feature detectors.

4: shape, color, glasses, facial markings

3: Explains why we confuse visually similar objects (they activate many of the same cells).

No need to worry about a single cell that represents your grandma dying.

Cells in inferior temporal cortex that respond to hands respond to many hands and not just a single hand.

Visual Agnosia

Apperceptive Agnosia:intact color, brightness, acuity

Associative Agnosia

Integrative Agnosia?

Which has more severe perceptual deficits?

Apperceptive Agnosia: perceptual impairment despite preserved basic visual functions

Associative Agnosia: selective impairment in object recognition despite adequate perception

Intergrative Agnosia: inability to integrate the part of an object to a coherent whole

Problems in perception present in both but are worse in apperceptive agnosics.

Theory of Apperceptive Agnosia

 -Problem of ______ and how that causes apperceptive agnosia.

Object Constancy

Difficulty recognizing a degraded object could stem from inability to extract its main perceptual features, an inability that normally allows us to recognize objects from differing views, under different lighting, etc.

Theory of Associative Agnosia, a problem with...?

They have trouble ______ stimuli but can _____ them.

Drawing as a memory deficit...?

Could be a problem with recognizing the meaning of visually presented objects.

Identifying... Copy

Patients often cannot recognize their own drawings on a later test

Patient J.B.R... deficit... worse for...?

Hypothesis 1 and 2

Exceptions

Severe associative agnosia, worse for living than for non-living objects

Hypothesis: categorical organization of semantic knowledge, brain damage destroys one category but not another.

 nonliving objects are easier to recognize because they activate additional forms of manipulation (how it is manipulated and how it feels) (scissors activating sensorimotor and visual areas)

More likely to see deficits for living over non-living objects but the opposite has been observed... never double-dissociations between patients in the same study, it is possible that the living objects were harder to distinguish than the nonliving objects in this single study. 

Face Recognition: 3 similar features

Are faces special: 3 pieces of evidence

Faces and Objects activate distinct brain regions

Recognition: two eyes, nose, mouth yet we recognize faces better than we recognize other stimuli

Do face perception and object perception rely on different brain regions?

1 Prosopagnosia: selective deficit in recognizing face, recognition of other objects is good, can recognize people via other features such as hair and clothes, can recognize voices, brain damage to occipital-temporal cortex, can also be cogenital

2 face selective cells found in the monkey superior temporal sulcus (fire when faces are seen)

3 a face-selective region in the fusiform gyrus (FFA) in the human temporal lobe (fMRI showed)

Can face and object processing operate independently of one another?

Double Dissociation

Face Inversion Effect and Holistics

Prosopagnosia: object processing operates somewhat independently of impaired face processing

Associative Agnosia: face processing operates somewheat independently of impaired object processing

Better at recognizing faces when not inverted, less true for other objects (houses) which suggests that faces are processed more holistically than other objects.  Faces are processed very holistically and there is poor memory when processing parts of faces.