task 4 Flashcards
(39 cards)
perceptual organization
process where elects in a person’s visual fields become perceptually grouped or segregated to create a perception.
grouping
elements are put together to create a perception
segregation
elements are separated
Biederman’s recognition -by-components (RBC) theory
objects are made up of geometric components called aeons (3D shapes).
we recognize objects based on their arrangement
limitations: not all objects fit perfectly into neons (e.g. cloud) and struggles to distinguish between objects within the same category that share similar elements
Holistic processing
perception pf a stimulus is based on the overall structure, not on the individual components. Eg. face recognition: faces are recognized based on the whole, not on the mouth or nose, etc.
gestalts principle of perceptual organization
explains grouping-> we make suumputions based on physical contours and knowledge. They rejected structuralism (=belief that complex perceptions are build up from individual perts/ sensations) because of:
*Apparent movement
*Illusory contours
Apparent movement
(Wertheimer) -optic illusion of movement created by 2 same objects presented rapidly one after other, separated by period of darkness
Illusory contours
contours that are perceived even though not physically present ex: 3 pac men form an illusory triangle
Herman grid illusion
grid that with black squares with white borders, corner look black even though they are white
gestalt principles
- Good continuation- w points are likely to be grouped together if when connected, they result in a straight or smooth curve line-> our perception prefers smooth and continuous paths.
*Pragnanz (good figure or simplicity)- every stimulus pattern is seen in the simplest and most organized way
E.g. Olympic rings are perceived as 5 circles, not 9 shapes - Similarity- similar things appear to be gourd together: in color, size, shape, etc
*Proximity- things that are bear each other appear to be gored together - Common fate - things that move in the same direction appear to be grouped together
- Common region - elements that are in the same space appear to be grouped together.
- Uniform connectedness - when elements shared something, they are perceived as belonging together.
inference projection problem
there is a challenge to determine 3D structures of the world form 2 D retinal images because:
*ambiguity- many different 3D objects can create the same 2 D image
*occlusion - some objects block parts of another object, so it’s hard to perceived full structure and we nee to make inference based on the information we have
*viewpoint invariance- human can recognize objects regardless of the angle which they view it, but that doesn’t work with the computers
Helmholtz’s unconscious inference theory
we make inferences and choose the most likely explanation for what we see based on past experiences
Bayesian inference- explains unconscious inference mathematically by combining prior beliefs with new evidence to make conclusion
our perceptions are the result of the unconscious assumptions, or inferences, that we make about the environment
border ownership
the border that separates the figure from the ground belongs to the figure
neurons in v2 recognize when an object is sitting on the background based on the contrast of the edges defining the border
why do we get gist( general impression of a scene)
*degree of naturalness- natural secenes have lectured zones and unique contours, whereas man-made scened are made up straight lines
*degree of openness- open scenes have visible horizon line and a few objects
*degree of roughness- smooth scenes contain few small element, whereas rough scenes contain more complex elements
*degree of expansion- converge of parallel lines has a high degrees of expansion.
color: some scenes have characteristic colors.
*global image features - can be perceived hollistically
physical regularities: regularly occur in the physical properties of the environment.
oblique effect
we notice vertical and horizontal lines easier and quicker than angled lines because they are more frequent
light-from-above assumption
we assume that light comes from above because the sun is above
semantic regularities
environmental properties associated with activities and meanings (when we imagine a scene, we see ourselves in specific setting based on scene schemas)
predictive coding theory
our brain makes predictions about the world based on higher levels of the visual system -> so based on prior experiences
new visual input is sent up to the brain (visual system) and is compared to preedictions flowing down from higher levels.
*if visual input ( what we see) ashes the higher-level prediction (what we expect), nothing happens
- if they are different, a prediction error signal is sent to the higher levels to modify and update the existing prediction
lateral occipital cortex (LOC)
area in the ventral visual system pathways that contains neurons that respond to objects but not to texture or specific parts
extra striate body area (EBA)
contains neurons that respond strongly to bodies
visual word from area ( VWFA)
contains neurons that respond strongly to written words
grandmother cells
cell that seems to be selectively responsive to one specific object (for example: your grandmother)
feed-forward process
carries out object recognition one neural step after the other, with need for feedback (explain how fast we recognize things)
reverse-hierarchy theory
fast, feed-forward process give you crude informations about objects/ scenes on high-level are activity. You become aware of the details when activity flows back to low level areas.
