Chapter 6: Object recognition Flashcards
What is object constancy?
The ability to recognize objects in multiple situations
What are the two pathways for visual (and auditory) perception and what aspect do they cover?
1) Dorsal: where (superior longitudinal fasciculus): respond to large areas of space
2) Ventral: what (inferior longitudinal fasciculus): respond to specific areas, especially fovea
What is the difference between visual agnosia and optic ataxia and which streams have deficits for each case?
Agnosia = not able to recognize objects due to lesion in ventral stream
Optic ataxia = not able to use visual cues to guide actions in space due to lesion in dorsal stream
What area in the ventral stream is especially important for object recognition?
LOC: lateral occipital cortex
What is cue invariance?
The insensitivity to specific visual cues that define an object (drawing, picture, statue of a dog is always a dog)
What causes an increase in bloodflow in the LOC?
Novel and familiar objects. The LOC cares the most about shape, not texture
What is repetition suppression (RS)?
The neural response to a stimulus is lower in the second presentation compared to the first. The neural response is more efficient when it’s recently activated
What is fNIRS?
Functional near infra-red spectroscopy; method for measuring bloodflow in infants head with an EEG like cap, but an fMRI concept (oxygenated vs. not oxygenated blood)
What is multistable perception?
Observer experiences competition in perception. E.g. in binocular rivalry task
What are the grandmother cells hypothesis and the ensemble coding hypothesis?
Grandmother cell: states that there is one neuron that fires for for example recognizing your grandmother
Ensemble theory: a set of cells respond to your grandmother. This theory accounts for recognizing novel objects and seeing similarities to familiar things in them
What is the difference between a shallow feed forward network and deep feed forward network?
The shallow feed forward network is an artificial network that has no hidden layers and has a clear structure.
The deep feed forward network has multiple hidden layers and reflects for example the visual system
What is the difference between decoding and encoding brain signals?
Decoding: predicting the stimulus based on brain activity
Encoding: predicting brain activity based on the stimulus
What is the difference between top-down and bottum-up effects?
Top-down: you are searching something, controlled by prefrontal regions
Bottum-up: a stimulus gets your attention
What is the function and location of the fusiform face area (FFA)?
Recognizing faces independent of facial expressions. It also responds to other familiar things, but it has a strong bias for faces
What is the difference between the activity in de lateral face areas and activity in anterior face area?
Lateral: responds mostly to shape of the face
Anterior: responds mostly to appearance of the face
What is the PPA and what is its function?
Parahippocampal place area is responsive when judging spatial properties or relations, such as landscapes
What are the EBA and FBA?
Extrastriate body area and fusiform body area are both responsible for making judgments about the body
What is the OFA?
Occipital face area responds, just like FFA, to faces
What is apperceptive visual agnosia?
The components of an object are perceived, but they can’t be put together to a whole. It’s related to object constancy: you can only recognize it when it includes the important features (seeing stack of lego, but not the building)
What is integrative visual agnosia?
Patients don’t see objects holistically. They perceive parts of an object (they see a door, but not a house)
What is associative visual agnosia?
Perception without recognition (perception of house and can draw a picture of it, but can’t describe)
What are forms of specific agnosia?
A form of agnosia can be worse for certain categories. E.g. selective agnosia for fruit and vegetables
When do you use the unusual object test and how does it work?
Use it to examine if you have agnosia. E.g. identification of an object shown from different angles.
If you have damage in posterior regions right hemisphere –> poor performance
If you have left-hemisphere or anterior lesions –> perform fine
What are the two theories of object knowledge?
1) Sensory/functional hypothesis
2) Domain-specific hypothesis
What is the sensory/functional hypothesis?
Conceptual knowledge is organized around representations of sensory properties (such as shape and color) and motor properties and these representations depend on specific neural systems
E.g. a hammer’s visual representation is associated with shape and motor representations of actions
What is the domain-specific hypothesis?
Conceptual knowledge is organized by categories that are evolutionary relevant to survival and reproduction
Dedicated neural systems evolved because they enhanced survival by more efficiently processing categories of objects
What is prosopagnosia?
Failure to recognize face, often due to lesions in ventral pathway in FFA or OFA. It’s often bilateral, but otherwise right side
Congenital prosopagnosia: brain activity in FFA remains the same, but structural connections of anterior temporal cortex with the rest of face processing is disturbed
What is the relationship between face recognition and autism?
Fewer neurons in fusiform gyrus and less activity, so they are poorer in recognizing faces
What are the two types of recognition?
Holistic analysis = seen as a whole, such as face perception
Analysis by parts = parts are combined to a whole, such as reading
