Chapter 8 Attention and Scene Perception
Visual acuity is limited in the periphery. Hence you cant read more than 2 books at a time. even
if the letters are large.
^ general problem that the retinal array contains more information than that can be processed
Why can’t we process everything at once?
We don’t have the brains for it. recall recognizing a single object requires a sizeable
chunk of the brain and its processing power.
Processing everything all at once requires a brain that will not fit in the human head.
If itisn’t possible to process everything then what should be processesd?
--- we pay attention to some stimuli and not others.
Attention- is not a single thing and it doesnot have a single locus in the brain.but attention is
the name given to family of mechanisms that restrict processing in various ways
Can be overt or covert
Overt attention- refers to directing a sense organ at a stimulus.
covert attention -refers to discreet attention while stimulus organ is pointed somewhere
Divided attention – paying attention simultaneously to two things
Sustained attention – paying attention for prolonged period of time
In this chapter, mostly concerned with selective attention.
Selective attention - the form of attention involved when processing is restricted to a
subset of possible stimuli. The ability to pick one or few things to focus on.
But these terms are not mutually exclusive.
Operational mechanisms operate in all of the senses not just visual attention
We can also use attentional mechanisms to give one sense priority over another
Selection in Space
What does it mean to “pay attention”?
Michael Posner – created the classic situation. The subject must hit the response key as fast as
possible when the probe appears.
Measure of interest is the reaction time (RT)- amount of time that elapses between the
point when probe appears and the point when the subject hits the response key.
What if the subject is given a cue? Cue – is a stimulus that might indicate where(OR
what) a subsequent stimulus will be.
Valid cue – when the probe appears in the cued location.
- RT decreases when valid cue.
- RT is faster because the participant is paying attention to the current
No cue condition – RT is longe(=slower) compared to valid cue
Invalid cue –occurs when the probe id misleading. Appears in the area opposite of the
cued location. - RTs are slower here than in the control condition because the subject has
been fooled into attending to the wrong location.
The experimenter doesnot tell the participants whether the trial will be valid or invalid
Peripheral cue vs symbolic cue
How long does it take for a cue to redirect our attention?
Depends on the nature of the cue.
Time 2: when the cue appears
Time 3: when the probe appears
Stimulus onset asynchrony: the time between the onset of one stimulus and the onset of
When SOA is o millisec, no time for it to be used to direct attention and there is no diff
between the effects of a valid or invalid cue.
When SOA is 150 ms, the magnitude of the cueing effect from a valid PERIPHERAL cue increases
Symbolic cues take longer to work because we need to do ome work to interpret the arrow.
BUT some symbolic cue work like fast, peripheral cues.
The spotlight of attention
Does attention actually move from one point to another
Attention could be deployed from one point to the next in several ways.
1. It might move in the manner analogous to the movement of the eyes.when we shift our
gaze, our point of gaze sweeps across the intervening space—attention might sweep
across space in similar manner like a spotlight beam.
2. It might expand from the fixation point, growing to fill the region from the fixation spot
to the cued location.---zoom lens model
3. When attention is removed from fixation point, it might just melt away etc.
Visual search experiments- looking for a target in a display containing distracting elements.
- Provide a closer approximation of the actions of attention than posner
- Target item, distractor item
- Visual searches are ubiquitous in everyday life but vary in their difficulty and
- It is harder to find a target as the number of items increases.
- One standard measure ; how much time is added for each item added to the
- As the tasks become harder, the slope relating RT to set size grows steeper.
- Saying yes is faster than saying no because even in the hardest task, it is
easier to spot the target item than to spot the absence of a target item. - Improving search tasks : efficiency –directing attention to the target as soon
as the display appears. Diff types of test tasks differ in in their efficiency.
Feature Searches are efficient
Feature search – search for a target defined by a single attribute
If feature is salient, doesn’t matter how many distracters they are. The target seems to pop out
of the display.
-we can process the color or orientation of all the items at once parallel
- when we measure the RT, it doesn’t change with the set size
There are about dozen or so diff basic attributes that support parallel visual search.
Many searches are inefficient
When the target and basic distracters in a visual search task contain the same basic features,
search is inefficient.
Serial self-terminating search
- A search from item to item ending when a target is found or until all the
items are checked.
T-among-Ls search interesting is that they are inefficient even though items are easy to see and
In Real-World Searches, Basic features Guide Visual Search
Guided search- search in which attention can be restricted to a subset of possible items on the
basis of info about the target items basic features.
Vs conjuction searches, no single feature defines the target. Instead, the target is defiend by
the conjuction or co-occurrence of two or more features
Two-feature conjuction searches lie between the very efficient feature searches and inefficient
serial searches.but conjuctions can be made more complex to get closer approximations.
The Binding Problem in Visual Search
Binding problem : we might be able to analyze a collection of basic features in a preattentive to
the object.. the challenge of tying different attributes of visual stimuli which are handled by diff
brain circuits to the appropriate object so we percieve a unified object.
Feature integration theory
- A limited set basic features can be processed in parallel preattentively but
that other properties require attention.
- Includes guided search
- We can use some preattentive feature information to guide the choice of
what should be attended and bound.
If attention is needed to bind features correctly, what happens if we don’t have time to
complete the job?
-illusory conjuction Attending in Time : RSVP and Attentional Blink
RSVP – rapid serial visual presentation
- An experimental procedure in which stimuli appear in a stream at one
- Rate of presentation
- But the task need to be limited to simple characters
Attentional blink :
- The difficulty in perceiving and responding to the second of the two target
stimuli amid a rapid stream of distracting stimuli if the observer has
responded to the first target stimulus with in 200 to 500msec before the 2
stimulus was presented.
- Its as if our ability to visually attend to the characters in RSVP is temporarily
- Fish analogy provided by Chun.
- Performance is good if T2 appears immediately after T1 scooping two
fishes at once!
- Metaphor reveals two processes at work; fast process can identify each
object as it appears and disappears. And a slower process is required if the
observer will actually respond to a particular item.
- Example of two different group who perform differe