1
Q
what is perception?
A
The ability to recognize and interpret information from all the senses
–> construction of the mind
2
Q
what is mental imagery?
A
We can imagine an object without seeing/hearing/touching
–> automatic but extremely complex process
3
Q
is perception an active or passive process?
A
Active!!
- Requires energy and active cognition
- Mind’s detective work
4
Q
How does perception work?
A
- We sense a stimulus
- Then form a coherent and stable mental representation of it
5
Q
What is the retina?
A
light sensitive area that creates images
6
Q
what happens with the retina?
A
it has a blindspot due to the optic nerve
–> the mind, specifically perception, fix this missing info with context clues
7
Q
what are visual ilusions an example of?
A
our brains filling in information from blind spots
8
Q
the dress example
A
- 60% view as black-and-blue // 30% view as white-and-gold // 10% can switch
- ultimately due to our mind’s different assumptions about lighting conditions
9
Q
bottom-up information
A
Sensory input, the detection of sensory signals, the beginning of perception
10
Q
what is transduction?
A
the conversion of sensory input to electrochemical signals used in the neuronst
11
Q
top-down information
A
knowledge and expectations that influence and enhance our interpretation of sensory output
12
Q
do we use bottom-up or top-down processing?
A
we use both! they run parallel and influence each other
13
Q
feedforward flow
A
signals sent forward in the brain via different pathways to increasingly sophisticated areas of visual processing
14
Q
what is the issue with initial feedforward processing?
A
the info may be coarse and consistent with multiple interpretationsh
15
Q
how can we resolve feedforward ambiguity?
A
comparing this info with feedback (reentrant) or reciprocal info
16
Q
what are unconscious inferences
A
educated guesses about the most likely interpretation based on visual cues
–> cotext
17
Q
what is predictive coding
A
Expectations are constantly compared with and validated against incoming bottom-up signals
–> our brain acts as a constant generator of expectancies to later interpret info
18
Q
what guides the mind to the most likely interpretations?
A
predictions and expectancies
19
Q
what does cognitively penetrable mean?
A
whether our cognitive processes like beliefs, motivations and knowledge can change our perceptual experience
20
Q
is perception cognitively penetrable?
A
there’s a debate surrounding this, some research says that yes (people who are tired overestimate the steepness of slope) but others insist that these are methodological errors or we’re confusing perception with cognitive judgement after the fact
21
Q
what is the part of the visual field near the nose on both of your eyes called? whats the other side called?
A
nasal half
temporal half
22
Q
where does visual info from both eyes go?
A
each hemisphere, which covers the opposite visual force
23
Q
what is the primary visual cortex
A
- first big processing stage
- specialized for the most rudimentary visual processes like orientation, contrast, motion directio and spatial frequency
24
Q
where does visual info go after V1
A
V2, 3, 4 etc to serve higher-order functions, sent through a network of interconnected pathways
25
what is object segmentation?
Visually assigning the elements of a scene to separate objects and background
--> takes a couple of steps and can depend on bottom-up cues
26
what is figure ground organization?
A form of object segmentation where the boundaries of the objects are clearly visible, but it is unclear which side of the boundary belongs to the object or figure and which side belongs to the background → Face/vase
27
What 4 rules do we use for resolving figure ground organization ambiguities?
- Rule of enclosure
- Rule of symmetry
- Rule of convexity
- Rule of meaningfulness
28
Rule of enclosure
Whatever is enclosed we perceive as the object
29
Rule of symmetry
Whatever is more symmetrical we perceive as the object
30
Rule of convexity
Whatever is building or convex we perceive as the object
31
Rule of meaningfulness
Whatever is more meaningful we perceive as the object (shaped by experience)
32
what does occlusion mean?
blocked view, we make assumptions to understand the whole image
33
Amodal completion
filling in or inferring missing information from a occlusion
the object we see or act on does not correspond to the physical input received by any sensory modality
34
Boundary extension
where people tend to remember pictures as having extended beyond their edges, as if their minds fill in a bit of what the scene would have looked like had it not been cut off at its border
--> happnes in 1/20th of a sec
35
Amodal nudity
when our minds complete pictures in ways that don't actually correspond with reality
36
modal completion
Where your view of the object is not obstructed, and there is no objective boundary between the illusory object and the background, yet we still see it (circles square example)
37
how do we follow moving objects when they're blocked?
we extrapolate their trajectory
38
what are the issues of perceiving a 3D world?
our retina is 2D so its really hard
39
inverse projection problem
multiple 3D representations are possible from a single 2D image so there's ambiguity
40
monocular depth cues
also called pictorial cues
can be seen with just one eye
things like perspective, blur, texture gradient, shadows, size, clarity of detail etc
41
binocular disparity
A phenomenon in which the closer something is to you, the greater the difference between what your two eyes see.
42
binocular depth cues
Visual cues that support depth perception but require 2 eyes
--> used to measure how close or far away an object is
43
4 common cues that we use to resolve the inverse projection problem
- object constancy
- size constancy
- motion parallax
- relative motion
44
object constancy
Although the same object looks different on ur retina depending on its orientation, people are still good at recognizing them
45
size constancy
The perceived size of an object is stable, despite changes in distance
--> evaluated using experience and then use the apparent object size to determine an object's distance
46
motion parallax
the retinal movement resulting from the observer moving through a stationary environment
--> the closer things are to ur retinas, the faster they move
47
relative motion
= If we’re moving but we don’t notice, stationary objects appear to move
48
3 color-light related ways that we use our knowledge to construct what we see
- Colour constancy
- Lightness constancy
- Color assimilation
49
Color constancy
Our visual system factors in differences in illumination when perceiving colour (a banana will look yellow in bright and in dim light)
--> the dress is an exception to this
50
Lightness constancy
We can factor in illumination conditions when perceiving the brightness of things
51
Color assimilation
Opposite of colour constancy. An object takes on or assimilates the colour of its context. We use the info around us to construct our realities
--> oranges in orange netting
52
what is object recognition
Our ability to match images to stored representations in the brain
--> very hard cognitively
53
early vision
= Detecting an object’s basic features (orientation, color, motion)
54
late vision
= Analyzing and combining these features to recognize objects, faces and senses
55
how do we make judgements on images?
- based on how familiar they are
- facial recognition, for example, is very easy when we know the person despite changes in angle, lighting etc
56
Agnosia
Inability to recognize objects and match them to correct categories and labels
57
Apperceptive agnosia
impairs early vision, ppl can’t perform even very simple visual feature tasks like discriminate in orientation or color
58
Associated agnosia
impairs late vision, people can copy and identify basic features but cannot name or categorize objects
59
what 2 basic approaches do we use to understand object recognition?
- View-based approach
- Structural rdescription
60
View-based approach
We match images to representations (templates) that are like 2D pictures
--> works great for simply and clearly defined objects
--> can be done by machines
--> but extremely hard when the image is ambigious
61
What is the multiple trace memory model?
States that we store multiple representations corresponding to multiple views or templates of the same object so that we can quickly match an incoming object to its representation
62
What is a template?
a representation that fully describes the shape of an object
63
structural descriptions
Based on models that represent objects as sets of 3d parts standing in spatial relationships to each other
--> Better for ambiguity but much slowed
64
Biederman's recognition by components
you go through the list of possible components that are combined to form a recognizable object
65
are the two approaches to understanding object recognition exclusive?
not exactly
--> can be used to predict different things
--> the line between them is hazy
66
Acc to Biederman, how do we recognize objects?
- there's an alphabet of about 36 or less basic shapes called geons
- for any given 3D object, theres a 'structural description' that acts as a recipe saying how the geons should be arranged
67
What is holistic perception
when we process a whole object at once, including the relations of the individual parts to each other
68
do we always process holistically?
nope, depends on how much experience we have w the object
69
what is the FFA
fusiform face area
--> faces are always processed holistically in this area
70
is the FFA only for faces?
nope, people who are extremely knowledgeable in certain areas (birder, for ex) show activation in the FFA when looking at their aoks (birds)
--> BUT this isn't applicable for all types of expertise, reading is excempt
71
Is vision a learning problem?
for us, we are either born w it or we learn quickly
--> but for AI, its a real problem
72
Deep learning
a form of AI that uses brainlike algorithm that analyze images in multiple steps to process categorize and label natural images
73
First stage of AI learning
process raw image
74
More layers of AI learning
extract more abstract features to recognize
75
Output of layer learning
categorization label
76
Do 'what' and 'where' compute in the same place
no, they have separate pathways and mechanisms but are very integrated
77
What is active when asked to focus on objects?
Ventral cortex
78
What is active when performing a location task on similar types of stimuli?
Dorsal/parietal cortex
79
Who proposed the perception and action pathway?
Goodale and Milner
80
# what is the perception pathway used for
Perception pathway
for locating where something is
--> people with agnosia struggle here
81
action pathway
uses perceptual info to guide ongoing actions
82
What is mental imagery important for
- Aids memory
- Importnat for mental rotation, comparing and matching rotated images
83
Aphantasia
Inability to engage in mental imagery
84
How does the memory strategy called loci work
To remember a list of objects, imagine u walk into a room and find them all
85
There's a debate around how the mind performs mental imagery. What re the two sides called?
Depictive vs Propositional
86
Depictive
the brain represents mental images like it represents real images coming through the eyes
87
What study supports depictive explanation?
Kosslyin
--> people had to look at maps and then w/o looking determine the closeness of certain areas
--> required mental scanning
--> the time to travel this is proportional to the distance of the objects
--> suggesting that imagery is like perception
88
What is mental scanning
The process of mentally moving from one point in an image to the other
89
Propositional
Mental images are held in a post-perceptual abstract way. More like a linguistic description
--> describes imagery as an epuphenomenon
90
What is an epiphenomenon
something that occurs together with a process of interest but is not central to its function
--> in relation to descriptive, insists that all K shows is that people simulate perception, not that they're actually engaged in it
91
What is the main prediction to come out of the descriptive vs propositional debate?
The same machinery used for everyday perception is also used in mental imagery
92
What does topographic mean? Which cortex is topographic?
topographic = items adjacent in visual space are represented by neurons that are close to each other in the cortex
--> ocipital cortex
93
Where are spatial deficits more common?
In individuals with spatial neglect
94
What does spatial neglect mean?
People that cannot visually attend to objects on one side of their visual field
95
What have we found about mental imagery in relation to spatial neglect?
People with left-neglect are only able to describe aspects of the scene on their right-hand size
But when asked to mentally stand in the opposite side of the piazza, they were able to report what was previously unreportable
96
Dissociations between 'what' and 'where' in mental imagery appear to mimic..
the what/where dissociations found in pereptual abilities
97
Is Kosslyin's view popular?
Yes
--> has received a lot of backing through imaging
--> but nuanced debate continues
98
does mental imagery elicit stronger or weaker emotions than verbal thinking?
stronger
