Lecture 4/5: Perception Flashcards
Extroceptive sensations
- Any form of sensation that results from stimuli located outside the body detected by sensory organs
- Sensations that occur in our sensory organs. Your eyes, ear, nose, taste budss.
- Your sensory organs will absorb energy from a physical stimuli in the environment (light, vibrations, chemical compounds). The sensory organs have receptors that absorb this energy and then convert it.
- Sensations from the exterior world that we take in
Interoceptive sensations
Sensations from inside our body (source of what you are sensing comes from inside your body).
Examples of this:
* Proprioception: Sense of where our limbs are in space (good sense of where your limbs are in space)
* Nociception: Sense of pain due to body damage (sensation of pain)
* Equilibrioception: Sense of balance
* Dancers have increased interoceptive accuracy (Christensen et al., 2017). They are able to sense their heart rate (estimate their heart rate at different moments and then saw how far they were from actual heart rate).
* They could estimate heart rate more accurately than non-dancers
* This was unrelated to fitness levels or counting ability
Synaesthesia
Senses can mix: Synaesthesia
* A neurological condition in which one sense automatically triggers the experience of another sense. They perceive two senses, just by the trigerring of one.
Grapheme-color synesthesia
* A person sees colors with certain letters or numbers. They see letters and numbers with colours (ex: 5 are always in green and 2s are always in red.
* Most popular one
- Genetic component (abou 40% of synaesthesed have a family member with it)
- One hypothesis is that it is due to cross wiring (cross-talk) between processing areas in the brain. For example, the area of the brain that is involved in processing language stimuli and colour sit next to each other in the brain. It could be that for some reason there is some overlap there.
- Weird example: individual sensed time with a hoolahoop around her body (time and bodily)
- see something and automatically hear music.
Synesthesia is very popular in artists
- Artists are eight times more likely to have synesthesia than nonartists. Ex: pharell williams, stevie wonder, billie eillish (every person i know has their own color and shape and number in my head).
- One explanation is that maybe it is something that can be beneficial.
- The cross-talk between sensory areas in the brain increases the ability to think creatively and in metaphors
Sensations and Perceptions
- How do sensations merge to become something that you perceive? For all modalaties, we have this general process to get from sensation to perception.
- Stimuli is registered by a sensory receptor.
- Transduction of physical energy into a neural impulse (code).
- This code is translated into the brain through subcortical regions and then the cortex to generate behaviour and perception.
- Perception is when the brain is translating those impulses into something meaninful.
Sensation or Perception?
Touch
Smell
Bright color
Pleasant sounds
Body position
Pain
Touch - S
Smell - S
Bright color - P
Pleasant sounds - P
Body position - S
Pain - P
Vision
- Most well researched and most dominant form of perception in our world.
McGurk Effect
McGurk Effect: When you hear what you see
* A multisensory illusion such that there is a change in auditory perception from visual perception
* A voice articulating a consonant (/ba/) paired with a face articulating another one (/fa/) leads you to “hear” what you
“see”
* This shows us that there is an integration of sensory information. Shows that we integrate information across our senses. Other processes can affect what we can perceive.
* Question to think about: Can you experience sensory modalities separately?
* This also illustrates the dominance of visual input. The Mcgurk effect shows the power and importance of visual input.
The visual system
- Early visual processing (rhelm ofsensation)
→ supported by our Eyes and the optic nerve - Late visual processing (perception)
→ supported by The visual cortex (occipital lobe)
Early visual processing
1) Vision begins with out eyes. Light waves enter the eye, the light waves are then focused and inverted by the cornea and are projected onto the retina
* The retina, a thin layer of tissue at the back of the eye, forms an inverted image.
* Image is inverted because the front part of the eye is curved so it bends light.
* Later processes turn this image around
Retina has 3 types of receptive neurons in it: photoreceptors, bipolar cells and ganglion cellss.
2) Photoreceptors in the retina convert light to electrical activity.
Two types of photoreceptors:
* Rods: processing low light levels for night vision (light sensitive)
* Cones: processing high light levels for detailed color vision
3) The electrical signal is sent to bipolar cells and then to the ganglion cells.
4) The signal exits through the optic nerve to the brain
Information compression (phenomena that occurs in early stages)
- Information is compressed quite a lot in your early visual system.
- You have about 125 Million of photoreceptors in each retina that converge onto 100 x fewer ganglion cells → optic nerve → brain
- Input from the eyes to the brain is compressed. Your vision in that process is compressed 100times, we are not having a 1 to 1 correspondance.
- You don’t ‘see’ everything that is out there in the world
Photoreceptors distribution in the retina
Photoreceptors aren’t equally distributed in the retina → cells in the retina are set up in a non uniform way
* Cones (pick up high level of light, detail) are most concentrated in the fovea, which is a small area on the the central part of the visual field
* Center of your visual field is most detailed. Information in the center of your visual field is the most accurate or detailed.
* Rods (don’t pick up a lot of light) are mostly outside of the fovea, in the periphery
* Periphery of your visual field is less detailed and less accurate (more rods then).
Perceptual filling-in
- Visual processing systems will fill in what your periphery should see/should be perceiving.
- You are filling in the outside of the image with details that you see in the center.
Blindspot
- Photoreceptors are at the back of the retina (farther from the ‘world’)
- Ganglion cells are at the front of the retina (closer to the ‘world’)
- Ganglion cells make up the optic nerve that take this signal to the brain and have move past the photoreceptor layer
- At this ‘exit location’, there are no photoreceptors so visual stimuli are not received
Axons of the ganglia make up the optic nerve. The optic nerve has to pass through the photoreceptors to exit. So there will be a spot with no photoreceptors. No photoreceptors = you cannnot take in light = don’t see anything.
Why do we not see our blind spot??
But we do not ‘see’ our blindspot!
* This is because of perceptual filling-in. Visual systems will take information around your blind spot and fill it in.
* Later visual processes in the brain provide the missing information by ‘interpolating’ visual information (e.g., colors) from surrounding areas
* This is also because the left and right visual fields can compensate for each other’s blindspot. Due to the way that information travels from the early to late visual sytem. The optic chiasm is what allows your left and right field to fill in the blind spot
Early to late visual processing
Perceptual filling in requires the late visual processing.
* For the information from the early system to be received by the late processing, it has to exit the eye through the optic nerve and then hit the thalamus (subcortical structure - deep within the brain) and tthen there is a transfer of information between sensory, the optic nerve and information to the cortex.
* Thalamus: A way-station between sensory inputs and the cortex
* The optic nerve of each eye transmits information to both hemispheres, leading to the principle of contralateral representation. Some of the axons from the right or left side will swtich sides. Each optic nerve does transmit a signal to each hemisphere.
Contralateral:
* Left visual field is perceived via the right hemisphere
* Right visual field is perceived via the left hemisphere
* This is what allows your left and right visual field to fill in the blinspot.
Late visual processing
*Late visual processing occurs in the occipital lobe.
* Primary visual cortex - area is responsible for the late processing of the visual system.
* Some of the most basic processing in the late visual system goees from very simple feature based processing all up to something that is more integraded and complex.
1) Primary Visual Cortex contains specialized regions that process particular visual attributes or features (functional specialization).
Differerent types of cells process different kinds of visual information. They will process different types of attributes:
* Edges
* Angles
* Color
* Light
- For each aspect of the signal that reflects these more simple features, they are going to be processed in different parts of the primary visual cortex. It does happen in parallel, which means it happen at the same time.
- Whatever you are seing is broken down and processed seperatly in the primary visual cortex.
2) Visual Association Areas interpret visual information and assigns meaning
- information from the primary visual cortex is projected to the visual association area. At this point, we start assigning meaning to what we see
- What and where pathways
- what pathways: visual object recognition
- where pathway: object location
Pathways to the visual association areas
What (ventral) pathway:
* Occipital to temporal lobes
* Shape, size, visual details
Where (dorsal) pathway
* Occipital to parietal lobes
* Location, space, movement information
- Neuroimaging studies show separation of what and where pathways. You see seperate activity along the what and where pathway.
Neuropsychological case of dorsal
‘where’ pathway
Ventral damage with intact dorsal stream
* Impaired performance on visual object recognition or matching tasks (cannot match the cat to the image visually).
* Can usually recognize other obkects with other sensory modalities (touch, smell)
Dorsal damage with intact ventral stream
* Accurate performance on object recognition or matching tasks
* Impaired performance on visual guided action (picking up an object appropriately). Cannot do any task that requires space or motion. Cannot locate an object in space. Bad grasping.
- Some suggest this may mean that dorsal and ventral pathway represent “perception” and “action”
Lessons from the visual system
- Visual stimuli is altered at many stages of the
processing pipeline (e.g., inversion, compression,
within the primary visual cortex). You do not perceive stimuli directly - In the cortex, visual input is broken down, processed
separately and then combined to form a perception of
an entity - The reality we perceive is a construction of the brain
Bottom-up vs Top-down processing
- Bottom-up processing: the influence of information from the external environment on perception
- Information from the sensory organs (eyes) to the visual cortex. Vision: input from eyes to brain (represents bottom up processing). What we see is influencing what we are processing. “Signal is on the ground”
- Top-down processing: the influence of knowledge (expectations, context and goals) on perception
- Information from final stages of higher areas of the brain (prefrontal cortex or higher visual processing areas) that is sent back to the visual cortex
- Information from your mind is influencing the processing. Information processed in these higher regions of the brain will send a signal back down to the visual cortex.
Ambiguity in what we perceive
Constructivist Theory of Perception: Top down processes can influence data.
* Governed by top-down processes
* We use what we know, and current context to predict how to perceive sensory data. Sensory input can be interpreted in a variety of ways. There is ambiguity in sensory data due to top down processing. Your braiin navigates your world and predicts things.
* your brain will use what it knows to guide perception and so forth. It is a predictive organ
Pain perception is subjective
- Perception of pain is partly determined by expectation
- Rate pain of the shock in Phase 2 in the low and high cue trials. Sometimes the participants were given the shock with a queue (a shape) “when you see a shape, the shock that is coming next will be very painful” (high pain trial). Low pain trials “the shock coming will not be so bad”
- The shock levels were the same across conditions → regardless of the high or low pain trial, they gave the same shock (objectively no difference in shock)
- Pain ratings were higher in the high-cue than the low-cue. Expectation of pain changed their perception of it.
The Ponzo illusion
Visual illusions show what assumptions you bring about what the world should look.
- you are bringing your assumptions about depth perception. or assumption about how light comes from above.
Perception is predicted by what
Perception is predicted by knowledge
* We use assumptions about what we expect to see to guide perception
* Knowledge, heuristics and schemas that reflect assumptions about how the world works, affects perception
* The specific illusions we are susceptible to illustrate some of these assumptions
Contexts affects visual perception
- Changes in visual perception based on the surrounding
information (the context).- top down processing
- context you are in can shift what you see
- how eople ask you something or what questions you see affects your perception.
Three examples
* Ames Room
* Letter in Context
* Color in Context
Ames room
- A functional illusion when expectations guide perception
- These are expectations of ‘observation’
- Room is constructed as trapezoid (walls are slandered)
- Person will assume they are looking into a rectangle room not a trapezoid room.
Context constructs perception: The letters
in context effect
- The ability to read words in sentences even when the letters in the middle of some of the words are mixed up
- This is because you ‘expect’ to see real words in a sentence
- We perceive and read words as words even if the letters are mixed up.
- When we see sentences, we are in context that assumes the words are correct.
You can stlil raed this senetnece even thuogh lettres in the wrods are jubmled.
Context construct perception: the color in context effect
The context a color appears in can influence how you see that color.
* Color perception depends on both:
* the wavelengths of light that fall on our retina
* Our past experiences of how objects look under different contexts of illumination
Context of a colour can shift your perception of it. Dark colour background makes inner square look like a different colour than the little square in little background.
The early visual system
- Light waves enter the eye through the cornea, focused onto the retina as an inverted image
- Photoreceptors in the retina convert light to electrical activity. Retina contains rods and cones (cones are more concentrated in the fovea). Our vision is high resolution (detailed) in the center of the eye.
- Electrical activity is sent to the bipolar cells then ganglion cells
- This signal is sent to the brain via the optic nerve
Late Visual system
- Optic nerve will send info to both hemispheres
- The primary visual cortex (PVC) is the first stop in cortex for processing visual input
- Demonstrates functional specialization
- Features of visual input are processed in different regions - Damage leads to conscious vision loss
- Demonstrates we can have perception without awareness
Test of blindsight (experiment)
- Over trials, turn a light on or off in the blinded visual field
- Ask patients to guess if the light was on or off (forced choice responding task)
- Patients performed above chance on the forced-choice responding task for lights in the blinded area
Blindsight and visual imagery (experiment)
Trial 1: Ask patients to look at images of houses and faces.
Trial 2: Ask patients to imagine houses and faces
Results:
* people with blindsight didn’t show activity in areas that are related to explicit visual perception.
* Brains of patients looked more similar when they were imagining vs looking at images.
* Activity of patient with blindsight perceiving faces and houses was reduced when compared to control participants
* Activity of patient with blindsight when imagining faces and houses was similar to control participants
* Blindsight leads to deficits in consciously processing incoming visual information but not imagery
* There is something different between perceiving and imagining perceptions. Or, the blinsighted individuals have a selective deficit when they are processing information that has to travel from external world up visual processing pipeline to the higher association areas. For imagery, you do not need that pipeline. It comes from inside.
Damage to the “where” pathway
Damage to the what and where pathways affects conscious perception
* The dorsal ‘where’ pathway goes to the parietal lobe
* spatial information
* depth perception
* estimating movement and direction of objects
* problems with motion processing (akinetopias)
Akinetopsia
- Damage to the dorsal where pathway
- Visual motion blindness: cannot see motion. Instead, perceives motion as a series of stationary objects
- Movement is just really a bunch of static snapshots for these individuals.
Optic Ataxia
- Damage to the dorsal where pathway (this is why people have suggested that the where pathway is important for action)
- Inability to reach for objects with the ability to name objects. Can still name the object, just can’t reach out and grab it.
- Problems reaching for a cup of coffee … can recognize coffee
- Problems pouring milk … can recognize milk - There might be action specificity in this pathway
- Selective damage leads to problems with certain types of movement. Damage to different regions in the brain within the where pathway will lead to specific action impairements (person only has problems for a certain type of movement: eye movement)
Visual agnosia
- Damage to the ventral what pathway
- Difficulties recognizing everyday objects. Cannot identify visual objects with visual sense but can with other senses.
- Often from damage to the Lateral Occipital Cortex
- Difficulties can be selective to visual categories (faces)
- Functional specialization within the ventral pathway
- Can get damage to different areas along the pathway that will lead to different visual agnosia
Prosopagnosia
- Fusiform face area (FFA) damage leads to a selective deficit in recognizing faces, keeping intact the ability to visually
recognize other objects - Rely on non facial characteristicss to recognize the person (clothing, hairstyle).
What was the theory of people that did not believe in the functional specialization for faces?
- Is the FFA special for faces or just discrimination?
- Participants learn to discriminate between “Greebles”. The ones that had learned to discriminate them had more activity in the FFA.
- fMRI data as participants viewed greebles and other objects
- Greebles activated FFA more than other objects (cats, household objects)
- FFA is very important for face processing. There is some controversity - maybe it is just important for discriminating things that are similar.
Apperceptive visual agnosia
- A failure in recognizing objects due to problems with perceiving the elements of the objects as a whole. Can still detect specific features but cannot group them as a whole.
- Single visual feature perception (e.g., color, motion) are relatively intact. Can still reach/grasp for obkjects - motion perception is intact.
- Problems with perception and discrimination of objects
- Impairment is in grouping visual features to form perceptions that can interpreted as meaningful
- Cannot copy image but can draw from memory.
Constructivist Theory of Perception
A Constructivist Theory of Perception
* A top-down theory of perception
* Perception is influenced by stored knowledge and context
- Mental models
- mental models of how things work and we activate these mental models when we perceive.
- We generate a mental model of the world based on the sensory inputs and we will use these mental models to make predictions.
* We make unconscious inferences to interpret and to predict sensory data
Ambiguous Bistable figures
Ambiguity in the sensory data: the visual input
* Your brain will rely on certain cues to perceive these.
* You will see what you are more familiar with.
Principle of experience
Gestalts
- Image segmentation (figure-ground)
depends on sensory input, detect edges or shadows. You segment visual info into a foreground, background, edges, shadows (bottom up process) - Experience and knowledge also drives
figure-ground segmentation - Regions perceived as the figure are the
ones that are more familiar and more
easily named to the observer. We see what we are familiar with.
Familiarity effects on segmentation
People are more likely to state that the white part in the first row is the object whereas the balck part in the second is the object.
- This because the white part in the first row is shaped like familiar objects (cowboy bootss, butterfly, grapes)
Principles of proximity
Objects or features that are close to one another in a scene will be judged as belonging together. Things that are close together are part of the same package.
Principle of closed forms
We see a shape in terms of closed forms, and we like to see items that enclosed
as whole
Principle of good contour
We perceive objects as continuous where it is expected that it is continuous. You expect shapes that are sharp and smooth.
* We perceive objects as continuous in cases where it is expected tha they continue.
Principle of similarity
We organize objects or features of a scene based on similarity
* when all else is equal we organize objects based on similarity.
Context affects interpretation
Gave this image to children at two different time points (one in october and one during easter).
* Children was more likely to say it wass a suck (in october, ducks were present at the zoo)
* On easter sunday, rabbit wass more likely.
Direct models
J.J Gibson
An alternate account: Direct models
* We do not need to create these mental models, we do not need to infer. The cues in your environment will guide how your sensory information is processed.
*No mental model for sensory input to guide perception and action
* Perception involves using information directly from our environment directly, without transforming it in our minds
* **A passive bottom-up approach to perception **
* States that we do not need to use assumptionss or prior knowledge to know what we see.
Texture gradients give the appear of depth
- Variations in the surface of an object gives you cues about the distance.
- The density of a texture (gradient) provides information about distance
- Near objects are farther apart and Far objects are closer together
- Incremental changes in texture
can provide information about
your movement and distance - Closer together cobble stone = illusion that is further.
Topological breakages
- Discontinuity created by the intersection of two textures. This is what allows you to see it a two seperate objects.
- allows you to perceive edges
- bottom-up heavy processing
- Provides information about edges of object and aids in object identification
Direct models of perception
- We don’t need to create a mental model of the world
- What we perceive from the environment is to help guide actions. We pick up these cues from the environment automatically. They tell us how to act, what we are perceiving.
- Cues provides information on the potential function of an object and are perceived directly, immediately
* Buttons, levers, slots
*Gibson thought that action and perception were heavily linked. Something that comes naturally (reflect like). - You have a perception that allows you to know how to engage with these actions. Influence of knowledge here is very limited.
Prototype theory
- A prototype is the average representation of an object concept
- what is the most typical thing about that object.
- recognition is determined by a ‘good enough’ match (resemblance)
- allows for ‘flexible’ object identification
- you do not have this template matching, it is based more on resemblance than a 1 to 1 match. Good enough match = still recognizing.
Feature detection
- Visual input is broken down into individual parts (features). Visual input is first decomposed into features.
- each feature is processe separately and then put togeher to recognize the thing.
- the combination of features is used as a pattern for recognition to compare to a
prototype
- the combination of features is used as a pattern for recognition to compare to a
- each feature is processe separately and then put togeher to recognize the thing.
Example of feature detection: Recognition by components
- All objects are reducible to a set of features, geons, basic geometric shapes.
- Geons are 3D shapess that combine to form any object to help assist in object recognition.
- When you recognize something, you first break it down into a set of features (a geon) and then you compare the configuration of those geons in your visual perceptual input to the configuration of geons you have in memory and a match means recognition. . - Recognition involves:
- Mentally separating a visual object into geons
- Examining the arrangement
- Finding best match of arrangement to memory representations of geon combinations
