Motion Perception

Question 1

Motion perception occurs

Answer 1

in the brain

Question 2

How can we influence motion perception through actions?

Answer 2

This can be explained by the perceptual cycle. We can directly influence our motion percepts through our actions (perception <–> actions)

Question 3

Autokinetic Motion (definition)

Answer 3

If you look at a point light source in a dark room, it appears to jiggle. This motion percept is an artifact of eye movement

Question 4

Corollary Discharge

Answer 4

There is a neural circuit responsible for cancelling motion signals generated by eye movement; If the comparator receives both an image motion signal and corollary discharge signal at the same time, the motion percept is cancelled

Question 5

Behavioral Support for CDT

Answer 5

Perceived motion in afterimages
Motion from eye displacement
Motion during object tracking
Motion percepts following eye paralysis

Question 6

How can the perceived motion in afterimages be explained using CDT

Answer 6

The comparator is receiving a CDS, but it is not receiving an IMS because nothing is changing on the photoreceptor array; motion is perceived because the CDS is not cancelled by an IMS

Question 7

How can we explain motion perception following eye paralysis?

Answer 7

When the eye muscles are paralyzed (no IMS), the motor command is still being generated so a CDS is sent to the comparator and motion will be perceived

Question 8

Image Motion (definition)

Answer 8

percept resulting from things in the world changing their retinal position over time in the absence of observer motion

Question 9

Reichardt Detectors

Answer 9

A basic mechanism of image motion detection; a given circuit codes motion only in a single direction. Motion is perceived if the signal from two receptors reach a comparator cell at the same time

Question 10

Where do RDs exist in the brain

Answer 10

More peripherally for lower animals (e.g. flies, rabbits, cats)
Centrally (mostly cortex) for humans

Question 11

Kinetic Depth Effect

Answer 11

It is hard to recover the 3D structure of an object from its 2D projection on the retina
Rotational object motion helps you recover the objects 3D shape

Question 12

Self-Motion (definition)

Answer 12

motion percepts that result from moving our head or body

Question 13

Optic Flow (definition)

Answer 13

motion regularities in the visual field resulting from observer motion
flow field - the array of motion signals generated by observer movement through an environment

Question 14

Induced Motion (definition)

Answer 14

Illusory motion percept where if you surround a smaller object with a bigger object, then move the bigger object, we will often see the smaller object as moving. Caused by figure/ground assignment (biased to interpret big things as stationary and little things as moving)

Question 15

Motion Aftereffects (definition)

Answer 15

Illusory motion percept where if you stare at a moving pattern, then look at a stationary pattern, the stationary pattern will appear to move in the direction opposite that of the moving pattern. Caused by selective neural fatigue in the motion system

Question 16

Selective neural fatigue (definition)

Answer 16

cells coding for one motion direction “get tired” so the opposite motion direction dominates the percept

Question 17

Apparent Motion (definition)

Answer 17

illusory motion percept arising from the rapid presentation of still stimulus. The type of motion perceived on TV, movies, and computers.

Question 18

Short-range AM

Answer 18

small stimulus displacement
motion seen in movies and on TV
believed to be implemented by RDs

Question 19

Long-range AM

Answer 19

larger spatial separations between stimuli and longer temporal gaps
Influenced by many higher-level factors; likely implemented by chaining together groups of RDs

Question 20

Stimulus onset Asynchrony (SOA)

Answer 20

the time between the onset of the two stimuli (more likely to perceive AM with SOAs in the 30-300 ms range)

Question 21

Three types of depth cues

Answer 21

Oculomotor cues (based on position of our eyes and tension of eye muscles)
Monocular cues
Binocular Cues

Question 22

Types of Oculomotor Cues

Answer 22

Vergence - inward or outward movements of the eye as we converge to look at near objects and diverge to look at far objects
Accommodation - change in the shape of the lens to focus on objects at varying distances

Question 23

Types of Monocular Cues

Answer 23

Pictorial cues - sources of information about depth in a 2D picture
Motion-based cues - source of depth information created by movement

Question 24

Pictorial Cues

Answer 24

Occlusion
Relative height/size
Perspective convergence
Familiar size
Atmospheric Perspective
Texture gradient
Shadows

Question 25

Motion-based cues

Answer 25

Motion Parallax (fear objects move faster than far ones when we move) Accretion and Deletion (the object that covers/uncovers another is closer)

Question 26

Types of Binocular Depth Perception

Answer 26

Binocular Disparity - each eye has a slightly different view on the world (the cue for stereoscopic vision)

Question 27

Challenges of Object Perception

Answer 27

Stimuli are variable and ambiguous Inverse Projection Problem Occlusion Viewpoint Invarience/Change in Illumination

Question 28

The Inverse Projection Problem

Answer 28

Scientists could not explain how can we recover 3D information about an object when we are beginning with 2D information, and this 2D information could be resulting from an infinite number of objects

Question 29

Object occlusion Problem/Blurred Object Problem

Answer 29

Occlusions remove parts and features of objects, making feature-matching difficult, yet we can still recognize the objects. The same is true for blurred images.

Question 30

Viewpoint Invariance/Change in Illumination Problem

Answer 30

We can recognize objects regardless of the viewpoint and can recognize when changes in patterns are due to lighting or is a part of the object itself

Question 31

Structuralist Approach

Answer 31

Perceptions are created by combining elements called "sensations" (aka all information from low-level inputs) Failed to explain apparent motion and illusory contours

Question 32

Apparent motion

Answer 32

We sometimes perceive motion even when nothing in the visual field is actually moving. So how can this perception result from receptor information if there is no information (movement) causing the receptors to fire

Question 33

Illusory Contours

Answer 33

Your visual system creates illusory contours between vertices of a pattern, but these contours do not actually exist. If the contour does not actually exist, the receptors for this orientation should not be firing, yet we can still perceive a shape.

Question 34

Gestalt Psychology Approach

Answer 34

Arose after the failure of the Structuralist Approach Perception is not built up from sensations but it is a result of perceptual organization "The whole differs from the sum of its parts"

Question 35

Pragnanz

Answer 35

every object is perceived as simple as possible

Question 36

Feature Similarity

Answer 36

Similar things are grouped together

Question 37

Good continuation

Answer 37

points connected by straight or smooth curves belong together

Question 38

Proximity

Answer 38

things that are near each other are grouped together

Question 39

Common fate

Answer 39

things that are moving in the same direction are grouped together (e.g. a group of birds flying the same direction are assumed to be together)

Question 40

Meaningfulness

Answer 40

things are grouped to form meaningful patterns

Question 41

Synchrony

Answer 41

elements occurring at the same time period are perceived as belonging together

Question 42

Perceptual Figure-Ground Segregation

Answer 42

separating the objects in an environment (the figure) from the background The figure is usually seen in front of the ground and takes ownership of the contour distinguishing it from the background

Question 43

Factors determining which is the figure

Answer 43

patterns in the lower visual field symmetrical or small patterns vertically oriented patterns meaningful patterns

Question 44

Neural Evidence for F-G Segregation

Answer 44

neural firing was seen when presenting a bar in the receptive field that was also located on the figure; no figuring was seen when the figure is moved outside of the receptive field

Question 45

Two basic approaches to object recognition

Answer 45

Structural Description Models Image Description Models

Question 46

Structural Description Model

Answer 46

Explains object recognition as objects represented in terms of 3D volumetric features, the unique combination of which defines a given object (you extract 3D features from the world and compare them to a 3D model in your memory)

Question 47

RBC Theory

Answer 47

3D volumetric features called geons can be combined to form all 3D objects - the more geons used, the more detailed and specific the object representation

Question 48

Principle of componential recovery

Answer 48

The more geons you can retrieve from an image the better the object recognition (an occluded objects with geons present can still be recognized, more geons present --> easier recovery)

Question 49

Object recognition will be viewpoint [variant/invariant] to the extent that its geons remain visible (and discriminable) over the changes in view

Answer 49

invariant

Question 50

Image Description Models

Answer 50

ability to recognize a 3D object comes from stored 2D viewpoints from different perspectives

Question 51

Predictions made by Image Description Models

Answer 51

For familiar objects, view invariance should occur because, you will likely have a similar 2D view in memory corresponding to that object For a novel object, view invariance should NOT occur because there would not be other viewpoints stored in memory

Question 52

Testing the Image Description Model

Answer 52

When subjects were trained on one view of a novel object. then were testing on many other views. Recognition was good only for the training views.

Question 53

Perceptual Intelligence

Answer 53

Object perception is more than collecting features and matching them to memory - we also use scene context and assumptions about the world

Question 54

World Knowledge/Assumptions definition

Answer 54

Using pre-existing knowledge and assumptions to perceive and recognize ambiguous patterns (e.g. assumption that light comes from above)

Question 55

Scene Context definition

Answer 55

Using information about the other objects appearing in a scene to recognize a pattern