Perception Flashcards
Neisser‘s definition of cognition puts the first cognitive process as the conversion from sensation to a mental image. It is therefore no surprise that many early psychologists focused on perception, mainly visual perception.
Although the information processing approach assumes a strict linear processing pathway, it has become evident that there is continuous back-and-forth transmission of information.
What are those called?
Bottom-up and top-down processes
How do we perceive the world?
The specific receptor cells of the respective sensory organ converts specific stimuli energy into a sequence of neural action potentials. These neural signals are then transmitted to the sensory brain region (it takes about 30-40 ms) that are specialised in further processing of these signals. It is the brain where perception occurs.
What did Neisser state in 1967 about fundamental errors and erroneous assumptions of psychologists (which are still being made today) ?
1) Psychologists incorrectly assume that a person’s visual experience directly mirrors the stimulus pattern
(The rabbit-duck ambiguous drawing and other visual illusions demonstrate that the same image can be interpreted in different ways. Hence, our visual experience can not be a one-to-one reflection of the stimulus pattern that is sensed.)
2) There is the assumption that the visual experience begins when the pattern is presented and ends when it is taken away (The afterimage-effect - the effect of seeing an image on a blank screen after staring at an image for a certain duration - demonstrated that there is sensation in the absence of a stimulus.)
3) The experience is assumed to be a passive copy of the stimulus that in turn is mirrored by the verbal report. In other words, you can accurately describe what you are perceiving. This would mean that cognition is not alternating anything about the sensory information other than verbalising it. (This would mean that psychophysics is fundamentally flawed and it still matters today. When you conduct an experiment pressing buttons and measuring response times, the time it took you to press the button includes the time it took for you to perceive the stimulus during that trial and all other processes that happened after that, such as memory, attention dwell time, linguistic programming and so on.)
—> Contemporary psychology is not dealing with the world as it truly is, but instead is completely focused on how the mind represents and interacts with the world, using neuroscience as a link to the natural sciences.
How can we explain the afterimage effect?
To explain the afterimage effect we need to look closely at how neurons behave. Neurons are connected to each other into a network. Some of these connections are positive or excitatory and other connections may inhibit nearby neurons. For example, a neuron that responds to light at a particular location in the image, may inhibit a neuron that responds to darkness within that same location. When the stimulus is taken away, the initial neuron receive less stimulation and thereby causes a weaker inhibition to the other neuron. This neuron will feel less inhibition and now increases in firing rate, which is then interpreted by the brain as if a new stimulus is presented. The afterimage is always the negative of the original image.
Sensory neurons do not signal the occurrence of a particular stimulus, such as light energy or the particular chemical. What do they do instead?
Instead, they signal the presence or absence of a certain stimulus energy.
The sense organs contain sensory networks that modify the signal before it reaches the brain, where perception occurs. What is the function of these sensory networks?
These sensory networks are passive abd have the function to enhance differences for better perceptual discrimination.
—> signals that the brain receives are exaggerations of the sensations.
(This pre-perceptual processing produces some interesting illusions)
The neurons in the retina are interconnected to form a neuronal network with many interesting properties.
What happens before stimulus information reaches the ganglion cell?
Before stimulus information reaches the ganglion cell, the horizontal and bipolar cells form a convergent subnetwork of inhibitory and exicatory connections.
—> each ganglion cell therefore receives input from about 100 rods, which forms the circular receptive field of the ganglion cell. Some will excite the ganglion cell and some will inhibit it.
How is the network organised (neuronal network in the retina) ?
The network is organised in such a way that the ganglion cell is excited when light is presented in the center of the receptive field, but inhibited when light is presented in the donut-shaped area surrounding the center.
When they say tje ganglion cell has an ON-center-OFF-surround receptive field. The opposite is also possible with the ganglion cells having an OFF-center-ON-surround receptive field. What is an interesting consequence of it?
An interesting consequence of the receptive fields is that if light is presented to the entire receptive field, the on-and-off pathways cancel each other.
—> This means that the retina is only sending information about differences between center and surround.
Apart from light/dark, we also distinguish colours. We had red/green receptive fields, but also yellow/blue.
This may seem odd since we do not have cones coding for yellow.
How is sensitivity to yellow created?
Sensitivity to yellow is created by a sensory network in the retina. It involves adding the output from the red and the green cones and substracting the activity in the blue channel.
It works like that:
Both red and green cones become active when yellow light is present, but less so when blue is present.
The blue cone has virtually no sensitivity to yellow light; it has an opposing colour sensitivity compare to the sum of the red and green cones.
—> Therefore, despite not having a cone that is yellow-sensitive, a subnetwork in our retina is able to create that sensitivity.
What is perception?
Perception is the process by which the cognitive system CONSTRUCTS an internal representation of the external world.
This means that perception is an active process.
It also means that it is not an accurate representation of reality.
Instead, it is an interpretation of the sensory input (image).
This paves the way for different interpretations of the same sensory input and therefore different percepts.
What is ambiguous stimuli and what do they demonstrate?
Ambiguous stimuli - stimuli that can be interpreted in multiple ways- demonstrate that the perceptual system is sensitive to expectations.
This is known as top-down bias.
The sensitivity of our perceptual system to expectations is known as top-down bias (contrast with bottom-up processing).
Name examples for top-down bias?
Top- down bias occurs due to the information provided by the sensory neurons.
Top-down influence can be contextual, such as when perception of an image is influenced by images in the immediate vicinity of the ambiguous stimulus (e.g. duck-rabbit stimulus).
Top-down influence can also be conceptual, such as when after seeing one Interpretation, you find it hard to switch and see the other interpretation (e.g. young/old person stimulus).
How do we aquire visual information?
When looking at a picture, we do not receive continuous information. Our eyes move in jerked movements called saccades and they last about 25-100 ms. The pauses between saccades are called fixations and it is during the first 50ms of a fixation that we aquire visual information.
During a saccade, visual processes are suppressed.
In other words, we are blind when we make eye movements.
What is an example of bottom-up processing?
Our eyes are attracted to salient pieces of information in the environment, such as a flashing traffic light.
How do we know that saccades are also influenced by top-down processes?
Yarbus (1967) presented pps with a picture entitled ‘The unexpected visitor’. He then recorded 3 minutes’ worth of eye movement trajectories under different instructions. When asked to give the ages of the people in the picture, the fixations were clustered around people’s faces. When asked to remember the clothes worn by people, the eyes were scanning the entire bodies of the people. The trajectories differ among instructions, indicating a top-down influence on eye movements.
A bolt of lightening last about 50 ms, but still we perceive it as lasting for much longer. How can this be?
This is due to visual persistence and forms a visual or iconic memory.
During this lingeringof the icon, further information can be aquired. The icon can be ‘seen’ in neurons firing long after the flash has disappeared.
What procedure to investigate the iconic memory was developed by Sperling?
He used a whole report and a partial report procedure. In the whole report procedure, pps would see 9 letters flashed on the screen for 50 milliseconds, after which the screen goes blank. The ppt’s task is to recall all letters, but they recall only 4-5 out of 9 letters. In the partial report procedure, pps stared at the screen with letters and after intervals varying up to one second, a tone was sounded that indicated the row of letters the pps should report. If the tone sounded with 300 milliseconds, pps were able to report the letters in the indicated row.
Sperling observed that all letters in the display were initially in the icon and that this icon disappears through decay and interference within 250 ms.
What are the three main approaches ir theories to pattern recognition?
1) Gestalt approach
2) Template theory
3) Feature theory
What is the gestalt approach?
The position of gestalt researchers, such as Wertheimer, Köhler, and Koffka, is that the atoms of behaviour, as investigated through behaviourism, is too strict and does not allow explaining behaviour at a higher level. Instead, they argued that psychological phenomena should be broken down until you reach the smallest component that is still a structured whole.
It is like breaking down a sentence into words instead of letters.
They promoted the Law of Prägnanz or Good Gestalt. In vision, of several
possible interpretations of a scene, the interpretation that will occur is the one that is the simplest and most stable interpretation.
There are several Gestalt principles that have been demonstrated through visual illusions, such as?
- Law of figure-ground:
Part of an image is treated as figure and the other part as the background.
Selecting which part of the image is the figure amd which the background can be influenced by existing knowledge.
—> That is, there is top-down influence that stabilises the percept. - Law of closure:
This is the tendency to close gaps in an image. - Law of proximity:
Elements that are near one another tend to be grouped together (when no social cue is present).
When social cues are present , the grouping immediately changes. Social cues lead to underestimating the distance between interacting people. - Law of similarity:
Elements that are visually similar tend to be grouped together. - Law of Good continuation:
Where an edge is occluded, assume it continues in a regular fashion.
(This particular law can be used to make some stunning visual illusions and magic tricks).
Although the Gestalt principles were visual, the Gestalt principles apply to other modalities as well. They nicely describe the organising principles that guide perceptual processes, but they do not explain why these principles exist. They also do not provide a model of perceptual processing, as they are merely descriptive. Finally, none of these principles address depth perception.
What is the template theory (pattern recognition) ?
The template theory proposes that we recognise patterns because we have a template in our mind.
Every time we see a new exemplar we compare it to our templates. However, our perception needs to be invariant whilst the retinal image varies.
The occlusion problem states that we rarely see all of an object, as it is partially occluded by other objects. Yet, we still perceive an occluded object as what it is. We can even infer an object if it is not physically present, such as the illusory triangle.
The template theory has problems accounting for multiple perceptions of the same ambiguous image.
