18&20. size & shape constancy Flashcards
Law of size constancy?
We tend to perceive an object to be a fixed size regardless of the distance to the object (i.e., regardless of the visual angle)
Holway & Boring experiment – what principle were they trying to test?
Law of size constancy –> can we account for distance when we perceive size?
Holway & Boring experiment – how was it set up?
- subjects sat at the corner of 2 hallways
- saw a test circle that changed in distance (10-120ft) and size, but was always 1 degree of visual angle & a comparison circle (10ft away)
Holway & Boring experiment – what did participants do?
- had to make the comparison circle the same size as the test circle
Holway & Boring experiment – results?
There is a linear relationship between the distance to test circle and the size of comparison circle
IE –> participants judged distance right and perceived size accurately
Holway & Boring experiment – (pt 2) what did they change? Why?
- got rid of all depth cues
- turned of lights –> gets rid of most cues
- look through peep hole
–> no binocular cues, no motion parallax - test if depth cues are the cause of size constancy
Holway & Boring experiment – (pt 2) –> results/ conclusion?
- “Law” of visual angle
- judgements of size (without any depth cues) are based only on visual angle
- NO size constancy –> graph is flat line
- size constancy can’t happen without depth cues!!
Why are the moon and sun perceived as approx. the same size? What “law” explains this?
- there is almost NO good depth information that far away in the sky/space
- Law of visual angle
–> both take up approx. the same visual angle, so we perceive them as the same size
Size-distance invariance def? What is it?
- Relationship!!
- the perceived size of an object DEPENDS ON its perceived distance, and vice versa.
Emmert’s Law def? What “object” does it involve? What depth property does it relate to?
Object: retinal after images!
Property: size-distance invariance
- perceived size of afterimage changes (proportionally) based on distance of surface its “projected” on
- if you look at a wall far away, afterimage is big
- look at a desk up close, after image is small
- shows that we take depth into account when perceiving object size
How does depth perception affect size constancy? (negatively)
- if we misperceive depth, we misperceive size
- we judge size based on depth
- causes “failures” of size constancy, or illusions
6 examples of “failures” of size constancy caused by misperceived depth? (illusions) (1 is sorta repeated)
- Ames room (2 versions)
- Moon illusion
- Ponzo illusion
- Ponzo variant with fMRI
- Muller - Lyer illusion
- Outside rear-view mirror illusion
Ames room –> what happens? why do we misperceive depth? What are the results?
- objects look like they’re all the same distance, leading to misperception of relative size
- depth cues eliminated through peep hole
- all “rectangular” shapes are actually trapezoidal, so its angled back but you perceive it as perpendicular to you
- you think objects on the left side are much smaller than objects on the right side
Ames room –> difference between version 1 and version 2?
Version 1:
- DIFF. retinal size
- same apparent distance
- kid looks small on one side, huge on other
Version 2:
- SAME retinal size
- same apparent distance
- mom looks same size as kids
Moon illusion –> what happens? best explanation for why?
- moon appears much bigger at horizon than peak
- perceive distance at horizon to be much bigger
- very difficult to study and prove, so we still don’t know for sure
Ponzo illusion –> what is it? Why does it happen?
- 2 of the same object (taxi, RR X sign, moon, etc) are placed on a railroad track, one “above” the other
- due to depth cues like linear perspective, texture gradient, and relative height, you think the “top” object is farther away
- top object looks far, perceive it as big
- lower obj. looks close, perceive as small
Ponzo variant with fMRI –> what is it? What did they conclude?
- Uses depth cues to make 2 checkered circles (with equal retinal images) seem diff. sizes
- they showed that the brain activity is V1 was greater for the circle that was perceived as farther and bigger
conclusion:
- we actually do perceive it as bigger, and devote more brain activity towards it
Muller-Lyer illusion –> what is it?
- 2 lines (red and green) are exactly the same retinal image size
- add Vs facing away from the red line and facing towards the green line
- makes the red seem bigger than green
Muller-Lyer illusion –> best explanation for why it happens?
- the red line with Vs facing away from it looks like the far corner of a room
- this corner would be the farthest away from you
- the green line with Vs facing inwards looks like the edge of a box protruding towards you
- this corner would be closest to you
- you perceive the line of the “far away wall” as longer than the “close up box”
Outside mirror illusion –> what is it? What problem would it cause?
- the outside mirror is convex, so it makes objects seem smaller than they really are
- it assumes that you use familiar size to judge distance to objects you see
- if you see a small object in the mirror, you’d think you’re farther away than you really are, since the mirror is making it seem smaller than it really is
Shape constancy def?
- we perceive the shape of objects correctly, regardless of the retinal image shape, which changes due to changing orientation
- tendency to perceive an object’s shape as constant despite changes in retinal image due to changing orientation.
Shape-slant invariance def?
- relationship between perceived shape and perceived slant
- perceived shape of object DEPENDS ON its perceived slant, and vice versa
- shape of an object can be determined by COMBINING info about the shape of ret. image and slant of its surface
**If no slant information –> no correct shape information
Anamorphic projection meaning? Why do we do it?
- an image that is seen “correctly” only from a specific viewpoint
- implying a failure of shape constancy
–> if shape constancy was working, you wouldn’t need to change the shape, or you could see it correctly from any view point - ex. road signs –> you only read it at an extreme slant
When does shape constancy fail?
- extreme slants
- anamorphic projections –> road signs
- if you perceive slant wrong
- Ames room (don’t perceive slant of back wall)
- examples of street art that you see 3D
Tabletop illusion –> what is it? why does it happen? success/failure? What illusion is it sorta similar to?
- shape constancy is working, but causes an illusion!
- the retinal image of 2 table tops are exactly the same, but they appear very different
- we take slant into account, and perceive the length and width of the tables very differently
- while the retinal projections are the same, it doesn’t really matter, because IRL, the tables would be different sizes
–> same as 3 car illusion that kept getting “bigger” but was the same
