Chapter 4 Flashcards
after the striate, information travels to the extrastriate cortex, which contains…
multiple sub-areas
after extrastriate, what are the two pathways?
ventral “what”
dorsal “where”
ventral “what” pathway
simply concerned with detecting identity - sometimes you need to recognize an object regardless of where you are seeing it
ex) can still recognize Dr. Dulas outside of the classroom
consists of:
V2
V4
Inferotemporal cortex
dorsal “where” pathway
simply concerned with detecting presence - sometimes you just need to avoid something, regardless of what it is
ex) if a car is coming towards you, don’t care what model it is just know to get out of the way
V2 receptive fields categorize boundary ownership, meaning
codes which part of a visual image is object and which is background
understands transparency/space
depicted by gray square images
inferotemporal cortex
high-level vision (objects)
part of the brain that knows what objects are
lesions to this region produces agnosia
agnosia
failure to recognize an object even though you are able to see it
can come in different forms, but all involve a breakdown in object recognition
receptive fields in the inferotemporal cortex are less about a part of _____ and more about particular types of _______. what are the two theories behind this?
space; stimuli
networks/ensembles of cells or individual cell coding/”grandmother cells”
grandmother cells
very hypothetical that specific cells code for a specific face
feed-forward process
generally occurs bottom-up, unidirectionally without higher levels feeding back info to lower levels
reverse hierarchy theory
allows for bottom-up, feed-forward processing that gives initial info about objects
BUT as additional processing at higher levels occurs, info flows back down the hierarchy to lower visual regions to alter their processing and refine details
top-down feedback
- higher levels communicate back to lower levels
mid-level vision
in-between
defining boundaries, grouping parts of an image
illusory contours and types
when you perceive a contrast even though the actual visual info doesn’t change between the two perceived parts
types:
contrast changes
continuation
emergence
gestalt processing
the whole is greater than the sum of its parts
our perception cannot be defined bu actual pieces of the visual world
4 rules of gestalt processing
closure
similarity
proximity
continuation
closure
assumption that shapes/objects are complete despite being obscured by other info
usually applying to negative white space
similarity
assumption that info that is of the same kind (shape, color, etc.) groups together
proximity
assumption that info that is relatively close together must also group together/potential emergent shape
continuation
assumption that lines/edges/contours keep going in their general direction even when encountering other objects
assumptions can _______
compete
2 types of competition in assumptions
parallelism and symmetry
common region and connectedness
heuristics
shortcut in the brain –> effective processing
fast and often effective but imperfect
2 types of camouflage
unintentional
dazzle
- can see but not identify edges, direction or distance
levels of “what” vision
low-level
- V1/striate
- line and edge detection
mid-level
- extrastriate (V2-V4)
- more complex “parts” of objects
high-level
- inferotemporal cortex
- specific types of objects (faces, scenes, tools, etc.)
3 inferotemporal cortex regions
fusiform face area
- highly sensitive to faces
parahippocampal place area
- highly sensitive to scenes
word form area
- attends to physical properties of words
V4
understands angles intrinsic to an object vs. those caused bu accidental occlusion
these complex shapes could potentially be the start of our visual system’s ability to detect object parts
theoretical
what gestalt principle explains our perception of the upside down white triangle (white negative space)?
a. continuity
b. closure
c. similarity
d. proximity
closure
what other gestalt principle explains our perception of the right side up triangle (lines)?
a. continuity
b. connectedness
c. similarity
d. proximity
continuity
t/f: heuristics are generally fast
true
t/f: heuristics are rarely accurate
false
t/f: heuristics can always be overcome by attention
falsee
in general, camouflage of any kind disrupts our ability to segment textures in our visual world, meaning …
hard to figure out when one texture goes into another
the visual system’s process of carving an image into regions of common texture properties
texture segmentation is shaped by _____ and _______
experience; evolution
perception can be thought of as the result of a “________” of principles that come to some sort of consensus to resolve ________
committee; ambiguity
mid-level committee “rules”
bring together that which should be together
split asunder that which should be separate
use what you know
avoid accidents
seek consensus to avoid ambiguity
global superiorty effect
properties of the larger figure override properties of the components
ex) cloud art
perception assumes viewpoints are ___ accidental
not
what heuristics are used to determine figure vs. ground
ground tends to be larger in size
ground tends to surround figure
things with high symmetry generally are figure
relatability
assuming two line segments that share the same slope/contour are part of a continuous whole
generally needs to involve only 1 bend
object recognition is not only fast, but _____
robust
2 potential models of object recognition
template models
- visual pattern that is stored in memory
- learned through experience
- assumes that view point matters
structural
- emphasizes the importance of object components and their spatial relations
- geons
- assumes that view point doesn’t matter
geons
building blocks of larger objects
3 levels of object recognition
global (broad)
basic (general)
specific
ex) global - furniture
basic - table
specific - kitchen table
2 types of agnosia
prosopagnosia
- facial recognition deficit
alexia
- visual word recognition deficit
