Sensation and Perception Flashcards
What is attention?
ability to preferentially process certain parts of a stimuli at the expense of other stimuli
-processed stimuli will be more clear than others
Why is attention needed?
- our perceptual system capacity is limited
- can’t process everything in visual scene simultaneously
- therefore, attention helps us from being overwhelmed with stimuli
Types of attention
Overt attention - directly looking at an object
Covert attention - looking at one object but focusing attention on another (e.g. looking one way in basketball but passing in another direction)
Monitoring Attention
-generally can tell one’s attention by tracking their eye movements
attention is when one fixates on an object
Saccades - very rapid eye movements between fixations
Fixation - rests between jumps where eyes look directly at one object
-fixations are directed by goals and expectations
Direction of Attraction
Initial Involuntary Process (attentional capture)
-initial attention fixates on noticeable/important parts (dependent on salience)
Salience - quality of being noticeable
-contrast captures our attention (colour contrast, shape contrast, motion contrast, size, orientation)
Theeuwes (1992) - asked participants to report the orientation of line inside the square. There was a circle of shapes, all green besides one red circle. Although participants were aware of goal, initial attention was was directed at red circle.
Subsequent Voluntary Process (guided by goals/expectations)
-after few fixations, we can direct our fixations depending on goals/expectations
-Unexpected objects can also cause fixations (related to expectations)
SEMANTICALLY consistent and inconsistent objects are objects that do/do not belong to scene
SYNTACTICALLY inconsistent objects are objects that belong but behaving unexpectedly
Effects of Attention
Attention increases rate of response
-Posner (1978)
3 groups: valid, invalid and neutral
valid - saw arrow above fixation point pointing in direction of where X will appear
invalid - arrow was pointing in opposite direction of X
neutral - no arrow
Participants had to indicate when X appeared
Results showed valid group responded fastest and invalid group was slowest (when fixated in a spot, participants observed faster)
Attention can influence appearance
-Carrasco (2004) demonstrated attention changes apparent contrast of an object
Shown grating sheet with two sides
Was asked to report which side had higher grating
Fixation point was given either a neutral cue (middle of sheet) or peripheral cue (on the left side of image)
In trials where grating was same contrast, those shown the peripheral cue reported left side to have higher grating
(attention can cause objects to appear bigger, faster or more richly coloured)
Attention can affect physiological responding
-neurons in brain respond more to attended stimuli rather than unattended stimuli
What is the Binding Problem?
How an object’s individual features are combined to form a coherent perception (harder with lots of object)
- different parts of a stimuli is processed by different parts of the brain (processed independently)
e. g. motion is processed by dorsal stream and form is processed by ventral stream
Feature Integration Theory (FIT)
- one solution for binding problem suggesting that binding problem is solved by attending to only one location at at time
- features that are present at a given location are processed so those features are bound together
- this avoids binding features from other objects
Illusory conjunctions = due to lack of attention, features of different objects will be incorrectly bound together
Tresiman and Schmidt (1982) - showed participants character string very briefly followed by a noise mask
Participants were asked to report the two numbers at the end of string, and colour of letters inbetween
Results show that participants tend to associate the wrong colours to wrong letters
What is Balint’s Syndrome?
Difficulty to focus attention on one object (prone to illusory conjunctions)
RM is a patient with parietal lobe damage where RM can’t focus on a single object (focuses on multiple objects at once)
When shown two letters with different colours, RM gave wrong colour combinations even when looking at them for up to 10 seconds (25% wrong)
-proves feature integration theory
Visual Search
perceptual tasks where requirements for binding takes place
e.g. if target contains same features of distractor, binding is required (red horizontal bar is target, but distractor includes green and vertical features too on top of red horizontal bar)
CONJUNCTION search = target differs from distractors only by its specific conjunction features (looking for object that is red AND horizontal)
-FIT predicts conjunction search is very slow, takes time to process each individual object
FEATURE search = focusing on features that the distractors do not have
type of visual search without solving binding problem
-FIT predicts binding does not need to occur as attention does not need to be applied to every object (faster)
Change Blindness
Changes that are obvious with attention but missed with no attention
(attention can determine what we remember)
-this occurs due to our limited capacity of perception (only few parts of scene can be remembered)
Simons and Levin (1998) - demonstrated change blindness in real life situation
Experimenter asked pedestrians on street for directions whilst people carrying a door walk in between them. People swapped behind door but continued to act as if asking for directions.
Results show approximately 50% did not notice person changed
-change blindness is only when changes are large and obvious, missed due to lack of attention
Problem of object and scene perception
Perception is harder than it seems due to:
-Stimulus on retina can be ambiguous (if shown different types of lines in 1D, it will all look the same, curved, zigzag, straight liens) (can also occur in 2D)
- Objects can be hidden or blurred (e.g part of glasses are hidden under a book)
- Objects look different at different angles/viewpoints/poses (machines cant detect when in unexpected angles/poses)
To demonstrate how difficult it is, even computers are worse at recognising objects compared to humans
-Computer object recognition systems use artificial neural networks
purposely created images to trick system: showed an image of a turtle but misidentified it as a rifle
Misclassifications of natural objects can happen if presented in unexpected orientations
Alcon (2019) - presented natural objects in different orientations to object recognition systems
School bus, motor scooter and fire truck were all recognisable unless in different orientation (viewed from above, object was rotated or sideways)
Solving Perceptual Problem - Structuralism
Arose from Edward Titchner studying Wilhelm Wundt
-distinguishes between sensations and perceptions
Perception = conscious awareness of objects and environment
Sensation = natural reflexes/behaviour in response to stimulation
Structuralism is the combination of sensations to form perception
Conscious awareness is made up of the elementary sensations (nothing is made up in conscious awareness, everything is already present and all from individual sensations
Solving Perceptual Problem - Geraltism
- Claims that conscious awareness if MORE than just the sum of elementary sensations
- conscious awareness can have characteristics not present in sensations
Evidence: APPARENT MOTION AND ILLUSORY CONTOURS
-observers see two stationary points flashing in succession. even though they are stationary, observer sees motion. shows that conscious awareness has a character (motion) thats not present in elementary sensations
(conscious perception of motion was created)
-illusory contours are seen in scenes where there is no physical contour
conscious awareness of illusory contour is created
Perceptual Organisation
Perceptual organisation grants humans ability to perceive objects and scenes
Perceptual organisation of images to form an object
There is Grouping and Segregation
With grouping and segregation, we can perceptually organise a scene into its individual objects to understand the scene
Grouping (perceptual organisation)
-Gestalt principle of grouping
process of which individual parts of an image is perceptually bounded together to form perceptual whole
6 principles (more principles increases likelihood of grouping to occur):
- good continuation = aligned (nearly aligned) contours are grouped together to form an object
- Pragnaz = ‘good figure’, grouping of images to make an image as simple as possible (e.g. WWF panda is made up of black splotches)
- similarity = similar objects group together
- Proximity = closer objects are more likely to be grouped together
- Common fate = objects moving in the same direction/way are grouped together
- Common region = objects in same area tend to be grouped together
- Uniformed connectedness = connected regions with similar characteristics tend to be grouped together (e.g. colour)
Segregation (perceptual organisation)
-Parts of a scene that is separated to form ‘wholes’ (perception of individual objects)
Figure-ground perception = objects are considered ‘figures’ and background is ‘ground’
Figural properties: objects are in front of image/scene, bottom of image, convex (curved) images, more recognizableimages
e.g. Rubin vase (where you see either faces or a vase; if silhouette was placed differently, you would see face/vase more easily)
Bottom of image - Vecera et al (2002)
Convex images - Peterson & Salvagio (2008)
when shown images with single border, observers tend to perceive border that is convex to be the ‘object’
Gist Perception
Gist perception occurs when an overall impression of a scene can be determined just by scene being rapidly flashing in front of us
-Potter (1976)
Observer was cued (1-2 words) of a particular scene than flash 16 random scenes. When asked whether one of the scenes matched the description, observers were close to 100% accuracy
Rapidly can get a gist percetion
-Fei-Fei (2007)
Investigated the minimum amount of time required to get a gist perception
Showed the scene then a mask and asked to describe the scene
27ms is minimum amount of time required to perceive gist (although not accurate)
Accurate gist can be achieved in just 250ms