Weeks 5-7 - 'Where' Pathway (Rebecca Champion) Flashcards
What is a cell’s receptive field?
The area in which, when stimuated, elicits a change in the firing rate of the cell
How does a single cell recording work?
An electrode is inserted into an individual neuron and it measures the electrical activity
How do receptive fields change as we go further into the visual system?
They become more complex
What do V1 simple cells respond to?
Oriented bars and edges
What did Hubel and Weisal find about orientation sensitivity in cats using single cell recordings?
- In the V1 there are oriented bar detectors
What are the three components which can be used to explain the tilt after effect?
- Orientation tuned neurons respond best to preferred orientation but also respond to other similar orientations
- Perceived orientation determined by distribution of responses across the cell
- Adaptation, the cell’s response decreases following prolonged activity
What is an explanation of the tilt after-effect?
- Before adaptation, vertical lines look vertical as the vertically aligned respond strongly and it becomes weaker the furhter away from this preferred neuron we go
- At the start of the adaptation, the tilted line looks tilted, this is because neurons further along will be have this tilt as their preferred stimulus
- During adaptation, tilted line continues to look tilted, but cells’ response decreases
- After adaptation, vertical line looks tilted due to asymmetrical response distribution (because some cells are adapted and others are not)
How does the size of the adaptation stimulus alter the size of the after-effect?
- When there is a large tilt of the adaptation stimulus, this means the neurons responsive are further away from those that have preferred vertical orientation
- This means there is less after effect as the neurons preferring vertical orientation aren’t being adapted
- The more similar the adaptation stimuli to the test stimuli, the larger the after effect
- Except when there is the same stimuli as the test stimuli as the adaption is symmetrical
What shape is produced by the plotting of adapt orientation and after-effect size?
Sine wave relationship
What does the tilt after-effect provide insight into?
- Provides evidence for orientation tuned cells in the human visual system
- useful because we cannot do single cell recordings in humans
What type of experiments are the tilt after-effect and size after-effect studies?
Psychophysical experiments
What is the size after-effect?
- Provide with adaptation stimuli which is fatter bars with thinner bars underneath
- Then provide the test stimuli which is equally sized bars on top and underneath
- People perceive the bottom bars of the test stimuli to be fatter and the top bars to be thinner
What is an explanation for the size after-effect?
- Before adaptation size is perceived vertically, certain neurons are tuned to different sizes of bars
- We then adapt to the bars presented in the adaptation stimulus so the cells’ response decreases
- After adaptation lines look thinner/fatter due to the asymmetrical response distribution.
- Cells that weren’t adapted in the adaptation stimuli will respond more strongly to the test stimuli giving the illusion of a fatter/thinner line
What does the size after-effect provide evidence for?
Size-tuned cells in the human visual system
What do after-effects suggest about size and orientation in the visual scence?
- Suggests that size and orientation are fundamental features of parts of the visual scene and the brain has cells tuned to these features
What is spatial frequency?
Number of bars per unit distance (usually cycles per degrees)
What is an example of spatial frequency?
- A one degree sine wave is compared to different sizd bars
- If the bar is fatter, only one bar will fit into one cycle of the wave
- If the bar is thinner, 2 bars will be able to fit into one cycle of the wave
What sized bars are considered low spatial frequency?
Fat bars
What sized bars are considered high spatial frequency?
Thin bars
What type of frequencies do natural images contain?
- Contain information at many spatial frequencies
- The extraction of high spatial frequency contains the fine details of the image
- Extraction of low spatial frequency contains the shading and the coarse details of the image
What is contrast?
The difference in luminance between the lighter areas and darker areas on either side of a boundary
What is the relationship between contrast sensitivity and spatial frequency?
- When contrast is low, we lose the ability to see high and low spatial frequencies
- We can only see intermediate spatial frequencies
What is the spatial contrast sensitivity function provide evidence for?
- We have greater sensitivity to intermediate spatial frequencies and can perceive these at lower contrasts
- We have lower sensitivity to high and low spatial frequencies. They need higher contrast to be perceived
Does spatial frequency tell us about retinal size or real size?
- Retinal size
- Does not indicate real size in the world since the projected size depends on distance
What is size constancy?
We perceive an object’s real size in the world regardless of distance
What is orientation constancy?
We perceive an object’s orientation in the world regardless of the orientation on the retina
Why is Depth Perception important?
3D perception is vital for interacting with the world and recognisinig objects
What is the inverse problem when it comes to 3D perception?
Any retinal image is consistent with infinitely many possible confgurations of the world
What is Binocular disparity/stereo vision?
- Types of information that arises because we have two eyes
- Our eyes are in different positions in our head so get a different view of the world
- The brain can put together the two retinal images and analyse the differencs and this can tell us about the 3D structure of the world
What is the motion parallax?
- Motion cue due to the self in motion
- Different objects move on the retina at different rates
- Images that are closer to us move faster than those furhter away
What is Kinetic Depth (KDE)?
- Motion cue due to an object in motion relative to ourselves
- Different parts of the surface of an object will move differently within the retinal image
- The surface of the object closest to us will move faster than the parts that are further
What are 6 pictorial cues for depth?
- Texture
- Elevation
- Relative Size
- Perspective
- Shading
- Occlusion
What are the two types of oculomotor cues?
- Convergence
- Accomodation
What is convergence as an oculomotor cue?
- The muscles in the eye need to converge more or less depending on how far away the object is
- If it is closer, more convergence is needed
What is accommodation as an oculomotor cue?
- The lens changes shape in order to focus the light on the retinal image
- The lens is fatter for closer images/objects
What are the two reasons as to why depth perception is so hard even with many cues?
- Many cues are ambiguous- 2D image is compatible with infinite 3D worlds
- With multiple cues available- how do we combine the information to perceive a single unified world?
Which type of cue provides the most ambiguity when it comes to depth perception?
Pictorial cues
How do we overcome ambiguity in depth perception?
- By using prior knowledge, or prior assumptions, to interpret the image
- These assumptions are gained through our knowledge and experience of the physical properties of the world
- This is a type of top-down processing and supports the constructavist approach to vision
What is an example using train tracks as to how ambiguity about depth perception is overcome?
- When looking at an image of a train track, there are infinite amount of possibilities about what we are actually observing
- For example either parallel lines or converging lines
- To interpret this image we can make the assumption that lines in the world tend to be parallel or perpendicular to each other
What is an example using shading as to how ambiguity in depth perception is overcome?
- An image of a circle with shading at the bottom could have an infinite amount of possibilities about what the retinal image actually is
- For example, A light is coming from above and the image is convex or the light is coming from below and the object is concave
- To interpret this retinal image, we use the assumption that light normally comes from above (so would be the convex option)
What is an example using elevation as to how we overcome ambiguity in depth perception?
- A retinal image of multiple cylinders are spread across the surface, some are higher than others
- The retinal image could have been created by objects sitting on a surface or floating in the air etc.
- To interpret the image, we assume that objects rest on a ground plane
What is a ground plane?
- Any surface that is parallel to the ground such as a table.
- We experience gravity, so objects tend to get drawn towards the ground.
What happens when our assumptions about depth perception are wrong?
Assumptions will be valid in most situations, but in some cases they will not be valid, which will lead to perceptual errors or illusions.
What is Ames room as an erroneous assumption of depth perception?
- When we view Ames room, we assume line are parralel, however, this is not the case
- The perspective assumption of parallel lines is invalid which causes an erroneous percept
What 3 problems does integration of multiple cues help to overcome?
- Reliability
- Ambiguity
- Conflict
Why is multi-cue perception important?
- Real-world scenes have multiple cues present
- Cues must be integrated to acheive a single unified percept
What are three types of multi-cue integration?
- Compromise
- Dominance
- Interaction
What is compromising as a type of multi-cue integration?
- When two sources of depth information are conflicting, the brain will try to find a compromise between the two
- Does take into account how reliable each cue is
What is dominance as a type of multi-cue integration?
- When two cues define very different shapes or depths, the brain may choose to ignore one in preference for the other
- Large cue-conflicts tend to arise when one cues’s assumptions are invalid
- Usually the cue with the valid assumptions will dominate
What is Interaction as a type of multi-cue integration?
- Occurs when information is incomplete or ambiguous
- Uses other cues in order to disambiguate the ambiguous ones
What is a study conducted by Young et al into multi-cue integration and compromising?
(Hint:dotted clyinders)
METHOD
- Participants viewed computer generated cylinders which were defined by two depth cues:
- Texture (cylinder was covered in dots)
- Motion (cylinder would rock backwards and forward)
- The experimentor could change the two cues in order for the cylinder to seem flatter or more stretched
- The experimentor then put the two cues into coonflict- either texture made the cylinder look flatter while the motion made it look stretched and vice versa
- Participants were asked to make the judgement of whether the cylinder looked flattened, circular or stretched
- The experimentor then changes the reliability of the different cues where texture would become more reliable because they made the motion more jerky or the motion was more reliable because the dots were distorted
RESULTS
- Initially when the cues were of the same reliability participants perceived shape is a compromise between the two cues
- When the texture was made less regular, the perceived shape is biased towards the motion cue
- When motion is made less smooth, perceived shape is biased towards the texture cue
CONCLUSION
- When two cues are conflicting, the brain will try to average them
- Final percept of shape will be biased towards the most reliable cue
How is Ames room an example of dominance of multi-cue integration?
- The invalid cue is dominating- creating the illusion
- We assume that the lines in the world are all parallel
- Also the relative size cue, the bigger person should be closer and the smaller further from us (correct cue)
- However, we for the invalid assumption of perspective as the lines are not parallel
What is Landy et al’s two stage model of cue integration?
- The interaction stage occurs first, where we disambiguate cues from the environment
- Then the compromise stage
How do we detect motion in the V1?
There are specialised cells that respond to lines and edges moving in a particular direction
What is a summary of what motion after-effects are?
- When viewing a stationary object ‘up’ and ‘down’ detectors fire equally
- Prolonged viewing of downward motion causes reduced firing of ‘down’ detectors (adaptation)
- Viewing a stationary object post-adaptation resilts in greater ‘up’ activation than ‘down’, hence we perceive upwards motion
What is local motion?
Motion of individual elements of the retinal image
What is global motion?
We can group the motion of many individual (local) elements to percieve a complex pattern of global motion
What are motion coherence studies?
- Participants are presented with a pattern of random dots moving in random directions (noise dots)
- The researcher will select a proportion of the random dots to move coherently in one direction (signal dots)
- It is then measured by what proportion of signal dots to random dots are needed to perceive coherent motion
What is the usual motion coherence threshold for humans?
10% of dots being signal dots (5% when highly practiced)
What does the coherent threshold suggest about the presence of global and local motion detectors?
- We have local motion detectors in the early stages of the visual processing system which have small receptive fields
- The local motion detectors outputs are grouped together by global motion detectors which have large receptive fields
What is the primary brain area for processing motion?
Area MT (Middle temporal)
How have single cell recordings shown that the area MT is important for motion processing?
- Inserted individual electrodes into the neurons in the MT and record
- Nearly all the cells in the area MT respond to motion
- And they have a preferred direction
How has artifical stimulation shown that the area MT is important for motion processing (Salzman et al)?
- We can artificially electrically stimulate neurons to fire and measure behaviour from this
- Salzman identified cells in monkey area MT that all had the same preferred direction (down)
- Presented coherence patterns of random motion and stimulated these ‘down’ preferred neurons
- Articifical stimulation of the cells led to motion judgements being biased towards preferred direction even when the dots were moving randomly
How has brain imaging shown that the area MT is important for motion processing (Tootell et al)?
METHOD
- Conducted an fMRI study of motion after-effects
- Set up two stimuli, one was an adapting stimulus of expanding circles
- The other was an adapting stimulus of expanding and contracting circles
- Participants took parts in one of the conditions and adapted to the stimuli
- They then were presented with a test stimulus
RESULTS
- When the paricipant adapted to the expanding only condition they experienced after effects
- When the participant adapted to the expanding and contracting condition, they did not experience any after-effects
- Looking at the fMRI, MT is active upon both adaption conditions and remains active during the period of after-effect for expanding stimuli but not for expanding and contracting stimuli because there are no after effects
- Shows that the area MT is linked to perception of motion and actual motion on the retina
How have lesions shown that the area MT is important for motion processing (Newsome and Pare)?
- Monkeys were trained to do the motion coherent task and then their threshold was measured before and after a small lesion to the area MT
- In undamaged MT’s the coherence threshold was 5%
- In damaged MT’s the coherence threshold was 80% showing that they could only detect coherence when 80% of the dots were signal dots
What is optic flow?
- Type of global motion
- Patterns of retinal motion produced when we move
- Different movements cause characteristic patterns on the retina
What type of movement creates the optic flow of expansion?
Forward translation (movement)
What is the focus of expansion (FOE)?
The stationary part at the centre of an expansion optic flow
What type of movement creates the optic flow of contraction?
Created by backwards translation (movement)
What type of movement creates the optic flow of horizontal?
When we make a rotation of the body, head or eyes at a constant speed across the retina
What type of movement creates the optic flow of horizontal?
Created by lateral (sideways) translation where closer objects move faster on the retinal than further objects (motion parallax)
What type of movement creates the optic flow of roll?
Created by eye, head or body roll such as when we look around with our heads or walk around (swaying motion)
What type of movement creates the optic flow of complex motions?
The accumulation of multiple optic flow input on the retina throughout normal movement in the world
What is a study by Smith et al into the fMRI imaging of optic flow in the MT and MST?
METHOD
- Compared two sub regions, the MT and the MST and their response to 5 types of motion:
1. Complex
2. Expansion
3. Roll
4. Translation
5. random
- They were interested in the response of the MT versus the MST to optic flow components compared to random motion (5)
RESULTS
- The difference in response to optic flow compared to random motion is greater in the MST than the MT
- Suggests that the MST is more specialised for processing optic flow motion that the MT
How is optic flow useful for determining where we are heading (Gibson)?
- The focus of expansion is particularly importnant as it tells us where we are heading to
- If we need to change where we are heading, we need to change the focus of expansion until it matches up to where we need to go
What is evidence to suggest that optic flow is not only used to determine where we are heading (Land and Lee)?
- Measured where people are actually looking when they are driving
- It showed that we don’t look at the FOE, we look at other parts of the scene
How is optic flow used for postural stability?
- Optic flow provides information on how our posture is changing
- For example, if there is pattern of expansion, this suggests that we are swaying forwards and we can try and counteract this
What is the swinging room experiment as evidence for optic flow being used for postural stability (Lee and Aronson)?
METHOD
- A room is made where the floor is fixed to the ground and the walls and ceiling can move independently to the floor
- This stimulates the optic flow that would be experienced by a person swaying
- Toddlers of 13-16 months are put into the room and the walls and ceiling are swayed
RESULTS
- When the room swings towards the toddler, it creates a pattern of expansion and the toddler compensates by swinging backwards
- When the room swings away, the toddler sways forwards
- 26% swayed, 23% staggered and 33% fell over
CONCLUSION
- Optic flow is an important source of information for balance and can override other sources of balance information such as vestibular and proprioceptive
What is a theory by Warren and Rushton into using optic flow to determine the movement of an object during self-motion?
- Suggests that retinal motion is due to 2 sources:
1. Self motion
2. Object motion - We have to determine the individual movement of the two components from the total optic flow which is the retinal motion from both sources
- We use the flow-parsing hypothesis to work out object motion from our self-motion
What is the flow-parsing hypothesis?
- Retinal motion due to self motion is subtracted from global motion
- Remaining motion is attributed to object motion
What is a study by Warren and Rushton into flow parsing?
METHOD
- Presented a pattern of optic flow (expanding pattern) and a horizontally moving probe
- The participants are asked to judge the motion of the probe
RESULTS
- The optic flow field influenced the perceived trajectory of the moving probe even (even when expansion was contained to one part of the screen)
CONCLUSION
- This provides strong evidence for global processing of optic flow
What is biological motion?
- The motion of another person’s body creates a complex pattern of movement
- Local motion signals (limb movement) need to be integrated to recover the global pattern of motion
- We appear to be particularly adept at perceiving biological motion
What is point-light walker stilmuli?
Where small lights are placed on a person’s joints
The visual motion is recorded, so only the lights are visable when doing characteristic tasks
What kind of information does the point-light walker stimuli provide (4)?
- Biological motion
- Gender
- Identity
- Affect (emotional state)
What is an fMRI study into biological motion processing by Grossman and Blake?
METHOD
- Participants were presented with either one of two stimuli:
- Biological motion
or
- Scrambled biological motion
- These had the same global motion but different local motion
- Using an fMRI scan, the brain was recorded to see which areas were most responsive to the biological motion
RESULTS
- Area STS (superior temporal sulcus) was the most active for biological motion compared to scrambled
What is a Trans-magnetic stimulation experiment conducted by Grossman et al into biological motion processing?
METHOD
- Participants had to determine whether they were viewing biological motion or scrambled motion whilst a TMS current interfered with the STS and the MT
- Noise dots were added to the stimuli to make the task more difficult
RESULTS
- TMS to STS caused significant decrease in the ability to distinguish biological motion from scrambled
- TMS to MT had no effect on biological motion perception
CONCLUSION
- Biological motion is a special type of complex global motion processed in a specialised area of the brain (STS)
