LEC 8: Obj Recognition Flashcards

1
Q

what’s the end of the retino-geniculate-striate pathway?

A

primary visual cortex - that when higher processing begins!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

magnocellular cells

A

info from RODS but some cones
position and motion /large and fast things/colorblind

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

parvocellular cells

A

primarily CONES
detail and texture, small cells,
more layers in LGN than m-cells
small/slow/colorful things

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

what are the 2 retinoganglion cells

A

p and m cells

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

how are you able to see left and right sides of space w one eye?

A

each eye synapses on both LGNs depending on L and R visual field,
allowing one eye to present both hemispheres w info on L and R visual field
R visual field for each eye maps on L hemisphere
L visual field for each eye maps onto R hemisphere

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

where is primary visual cortex located?

A

occipital lobe

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

indirect pathway

A

stops btwn LGN and MT

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

direct pathway

A

LGN straight to MT
mostly associated w magnocellular pathway (fast to detect motion)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

dorsal pathway

A

the where pathway
- where
- top

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

ventral pathway

A

the what pathway
- what
- bottom

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

obj cue + damage implication

A

monkeys associate shape of obj w treat (ex: raisins under cube evry time)
damage to temporal lobe causes impairment here

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

landmark cue + damage implication

A

monkey finds raisin near wherever certain landmark is
damage to parietal lobe causes impairment here

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

dual stream hypothesis

A

as visual info exits occiptal lobe, it follows 2 streams (dorsal and ventral)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

receptive fields become ______ further along the ventral stream

A

larger

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

ventral pathway steps

A

retina - LGN - V1- (V2-V3-V4) - PIT - CIT - AIT

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

visual agnosia

A

modality specific
gnosis - knowledge
A - without

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

apperceptive agnosia

A

cannot copy, cannot recognize objs by shape
can sometimes recognize by other features (texture, color)
some patients have deficit only when noise in pic or non-standard viewpt

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

associative agnosia

A

can copy but unaware of what obj is
difficult to read cannot transfer visual inputs into words
can be category specific (ex: prosopagnosia)
normal sensory functions but striped of meaning

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

apperceptive damage

A

occipital lobe

20
Q

associative damage

A

occipitotemporal cortex damage on both hemisphere

21
Q

prosopagnosia

A

inability to recognize faces (face blindness)
type of associative agnosia

22
Q

population encoding

A

uses hierarchal feature coding up to a pt

23
Q

prbs w population encoding

A

if whole population of neurons required to identify single perception,
how do you link that percept to stored knowledge (there isn’t one cell to connect, there are many)

24
Q

form-cue invariance

A

our ability to recognize objects regardless of their dimensionality (diff shapes but same thing)

25
Q

perceptual constancy

A

perceiving objs as unchanging even as illumination and retinal images change

26
Q

lateral occipital complex (LOC)

A

may represent abstract shape representations to support recognition despite invariance

27
Q

fusiform face area (FFA)

A

area in temporal lobe that contains many neurons that respond selectively to faces
possibly just due to experience (expertise)

28
Q

greeble experiment

A

concluded that there was some sort of expertise in facial recognition; visual neurons can be shaped by experience

29
Q

parahippocampal place (PP)

A

an area in the temporal lobe that contains neurons that are selectively activated by pictures of indoor and outdoor scenes

30
Q

extrastriate body area (EBA)

A

responds specifically to pictures of bodies and parts of bodies

31
Q

word form area (VWFA)

A

responsive to written words in ventral stream of left hemisphere

32
Q

feature-based encoding

A

encoding based on individual parts

33
Q

configural encoding

A

encoding based on whole unit or figure

34
Q

conjunctive enoding

A

assumes that features are explicitly conjoined, or linked together, through hierarchical processing in which lower-level regions representing features send their output to higher-level regions representing the shapes that result from the joining of those features

35
Q

damage to right hemisphere on representation

A

struggle to see a unit as a whole

36
Q

sparse encoding

A

“grandmother hypothesis” (one cell for when you see your grandma)
using hierarchal feature combination (one cell or small set for one thing like apple etc)

37
Q

patients w left hem damage show struggle with

A

seeing features or smaller parts of unit

38
Q

prbs w sparse encoding

A
  1. what abt hwere you learn something new
  2. how does this work developmentally
  3. what happens if neuron at top of hierarchy dies
39
Q

is location important in primary visual cortex or the IT (inferior temporal)

A

primary visual cortex
IT relies on stimulus but not location (size and complexity, larger receptive field)

40
Q

form-cue invariance

A

the brain’s categorization is constant regardless of form of cue that represents obj

41
Q

lateral occipital complex (LOC)

A

may represent abstract shape reresentation to support recognition despite invariance

42
Q

inversion effect

A

phenomenon in which recgnition (particualrly of faces) is poorer when obj is turned upside down; thought to reflect disruption to which recognition of obj relies on configural relationships

43
Q

Kanwisher et al (1997) (FFA)

A

FMRI study
participatns look at car or face and BOLD signal measured in FFA
activity showed increase when shown face

44
Q

greeble experiment

A

concluded that there was some sort of expertise in facial recognition. Visual neurons can be shape by experience

45
Q

James & James 2013 FFA/ Pokemon

A

experts showed inc BOLD activity in FFA when seeing pokemon characters

46
Q

MT (medial temporal)

A

detects motion and diretion of obj in visual field, part of dorsal/parietal pathway