Visual Object Recognition Flashcards
What two levels does regcognition occur?
Perceptual (allows us to recognise objects across different objects angles/lighting, when certain parts are occurred) and semantic recognition (allows us to know function of the object)
What is perceptual agnostia (lissauer 1890
Patients can assemble individual attributes of objects. Perceptual recognition is impaired, shown in copying performance
What’s apperceptive agnostia?
Could use hair informstion to pick up objects and not recognise them
Sensory perception is intact
Tactile recognition is broadly intact, semantic processing is ok
General intellect is intsft
Patients can reach accurately but not make perceptual judgements (angle to post it)
Can grasp as normal when ask them to grab things but can’t distinguish them from their background unless asked for them.
What is assoicatice agnosia (lissauer,1890)
Patients can properly form object structure but unable to access stored knowledge about this. Can copy picture correctly but no idea what it is-semantic memory impaired
What is optic aphasia (lhermitte and beauvois 1973?
Patients can apprehend object structure (perceptual recognition) and show semantic knowledge but can’t name what they see. Looks like recognition failure
What is object constancy?
Perceptual recognition allows us to know object has a constant shape despite the fact it changes in appearance and viewing position changes
How do we know we’re looking at the same object from different angles?
Template matching- rotate/manipulate the shape until it matches, but problem for similar objects
Critical features: look for features unique to the object. Works well for distinctive but hard when insured or defined by spatial arrangements
Structural descriptions: encode features and their relative positions to an internal frame of reference that can be seen across a viewpoint. Pick out constant aspects of an object
What is marr and nishiharas 1978 theory of object recognition?
A property object doesn’t tend to change accord the main axis. They way in which objects relate to each other doesn’t change across the main axis
Each object makes uo a generalised cone. Objects can be distinguished by the arrangement and length of them . Once axed is identified it is possible to identify it
What is the geon theory by biederman 1987
In addition to axis info, there are other invariant properties. These properties consist of wedges cylinders, collectively described as geons whenever objects intersect they form concavity, this allow them to be distinguished from each other
By removing the twins objects are v hard to identify
What is the face inversion effect?
Disrupts familar configured cues while leaving the identities of the features untouched, suggests faces are viewpoint dependant
What is important for face recognition
Configuration
Within category discrimination: need discrimination bettween highly similar examples of the same semantic category
The perception of curvature (can are face perception
Evidence for face and object represented by same/different system
Behavioral (RT/accuracy)
Brain imaging
Neuropsychological (RT/accuracy in brain damaged patients
What is behavioral support for different systems for object and face recognition?
Inversion effects are less apparent in objects than faces
What is brain imaging support for face and object recognition are different
The face selective fusiform gurus only lights yo when faces are viewed
What is neuropsychological support human and object recognition are different?
Pure prosooagnosia: patients can recognize objects but poor at recognizing faces
Wikininson et al 2009 patient RC had a unilateral right hemisphere stroke, able to detect objects and photographs but unable to recognize people by facial appearance . Issues innfusiform gurus
Patient CK had bilateral brain damage, visual acuity intact, can recognize faces, cartoons… but can’t identify them-associative agonosia
