PSYC 2650 Lecture Notes - Headphones, Searchlight, Daniel Kahneman

26 views16 pages
Published on 4 Oct 2012
Department
Course
Professor
Page:
of 16
CHAPTER 4
Paying Attention
Pgs for Midterm 97-106; 110-111; 116-123; 126-129
-The stimulus you're attending to only one of many available to you
- this fact invites two crucial observations.
1) seems clear you could choose to pay attention to any of the just mentioned, if you did, would be virtually
oblivious to other things on list.
-Indeed, until you read previous paragraph, probably oblivious to other stimuli mentioned!
-How do you do this:' How dose one manage to avoid distractors and focus attention in just the manner one wishes,
selecting only one input of many:'
-2) seems to be one thing cannot do: you cannot pay attention to all things at once.
-If start musing about the weekend, likely to lose track of what's on the page;
-if start planning term paper, won't finish reading assignment.
-Of course possible in ome circumstances to divide attention, deal with two different inputs at once.
ex- hum melody while reading words;
-most people walk and chew gum at same time; and so on.
-But where are the limits:' When can you do two (or more) things at the same time, and when can't you:'
Selective Listening 98 Paying Attention
-early studies of attention employed task called shadowing.
-participants hear tape recording of someone speaking must echo speech back, verbatim , while listening to it.
-Shadowing initially challenging, but becomes relatively easy after minute of practice.
-most experiments, message to be shadowed, the attended channel, presented through stereo headphones, so participants
hear attended channel through, say, right earphone.
- different message-the unattended channel- presented in left earphone, participants instructed simply to ignore
message.
-overall setup referred to as dichotic listening.
-Under circumstances, participants easily follow one message, shadowing performance generally near perfect.
-same time, however, hear remarkably little from unattended channel.
- If ask them, after minute of shadowing, to report unattended message was about, cannot (Cherry, 1953).
- cannot even tell if unattended channel contained coherent message or just random words.
-one study, participants shadowed coherent prose in attended channel, while unattended
channel heard text in Czech, read with English pronunciation.
-individual sounds (vowels, consonants) resembled English, but message itself (for
an English speaker) gibberish.
-After minute of shadowing, only 4 of 30 participants detected peculiar
character of unattended message
-More recent studies documented similar pattern w/ visual inputs.
-Participants watched TV screen that showed team of players in white
pg 99
-shirts passing ball back and forth;
- participants had to signal each time ball changed hands.
-Interwoven w/ players ( visible on same TV screen) another team, wearing black shirts, also passing ball back
and forth;
-participants instructed to ignore these.
- easily did selective task, so intent on white team they didn't see other, rather salient, events
that appeared on screen, right in front of their eyes.
ex-, they failed to notice another player wearing gorilla costume walked through middle of game,
pausing briefly to thump his chest before exiting!
-people not altogether oblivious to unattended channel:
-In selective listening experiments, easily accurately report whether unattended channel contained human
speech, musical instruments, or silence.
-If unattended channel contains human speech, can report whether speaker male or female, high or low
voice, speaking loudly or softly.
-seems, physical attributes of unattended channel heard, even though participants seem oblivious
to unattended channel's semantic content.
-Some Unattended Inputs Are Detected
-some results do not fit pattern.
-Some bits of unattended input seem to "leak" through get noticed.
- one study, people asked to shadow one passage while ignoring second passage.
-Embedded within unattended channel, though, series of names, including participant's own name.
-Overall, heard very little of unattended message, keeping with other studies mentioned.
-Nonetheless, about third did hear their own name
Pg 100
- commonly say, name seemed to "catch" attention.
-other contents also catch attention, if suitably primed for them:
-Mention movie just seen, favorite restaurant, often noticed in unattended channel.
- generally, words with some personal importance also]noticed
-results often referred under banner of cocktail party effect.
- at party, engaged in conversation.
-Many other conversations are taking place in room, somehow able to "tune them out:'
-aware other people talking, but don't have a clue what they're saying.
-All you hear is single conversation you're attending, plus buzz of background noise.
- imagine someone few steps away from you mentions name of a close friend of yours.
- attention immediately caught, find yourself listening to other conversation (momentarily) oblivious to
conversation had been engaged in.
- experience, easily observed outside laboratory, precisely parallel to pattern of experimental
data.
-How can we explain both general insensitivity to unattended channel also cases which unattended channel "leaks
through"
Perceiving and the Limits on Cognitive Capacity
-very start, two broad options how we might think about attention.
- 1) focuses on what we do with unattended input.
- Specifically, proposal : we somehow block processing of inputs not interested in, much as sentry
blocks path of unwanted guests but simply stands back , allows legitimate guests to pass.
-sort of proposal central for early theories of attention, suggested: we erect filter shields us from
potential distractors.
-Desired information (the attended channel) not filtered out , so goes on to receive further
processing
-Evidence suggests can shut out distracters in way, but "shutting out" seems to work on distractor-by-distractor basis.
-other words, can inhibit response to this dis tractor , do same for that distractor, these efforts can be
quite successful.
-However, same efforts little value if some new distractor comes along;
-that case, need to develop new skill aimed specifically at blocking new intruder
-further evidence indicates this is only small part of the story.
- not only do we block processing of dis tractors, but also able to promote the processing of desired stimuli.
pg 101
Inattentional Blindness
- Chapter 3, perception involves considerable amount of activity-as one organizes and interpret incoming stimulus
information.
- might think this activity require some initiative and resources from you, evidence suggests it does.
- experiment, participants told they would see large" +" shapes on computer screen, presented for 200
milliseconds, followed by pattern mask.
- If horizontal bar of " +" longer than vertical, participants supposed to press one button; if vertical
longer, press a different button.
-complication, participants weren't allowed to look directly at " + :'
-Instead, fixated (pointed eyes at) mark in center of computer screen-a fixation target-
- " +" shapes shown just off to one side.
-first three trials, events proceeded just as participants expected, task relatively easy.
- Trial 3, ex, made correct response 78% of time.
-Trial 4, slightly different:
-While target" +" on screen, fixation target disappeared replaced by one of three shapes-a triangle, a
rectangle, or a cross.
-entire configuration (" +" target and new shape) replaced by pattern mask.
-Immediately after trial, asked: anything different on this trial? anything present, changed, that wasn't on previous
trials?
- 89% of participants reported no change;
- apparently failed to see anything other than (attended) "+:'
- researchers told them (correctly) during the previous trial fixation target momentarily disappeared , replaced
by a shape.
- participants asked what shape had been ,explicitly given choices of triangle, rectangle, cross.
-responses to question essentially random.
-Even when probed , seemed not to have seen shape directly in front of their eyes
-What's going on in study?
-researchers proposed participants did see target shapes but, moment later, couldn't remember what just seen
(Wolfe, 1999).
-Mack and Rock, offer stronger claim: They note participants not expecting any shapes, not any way prepared
for shapes.
-claim, participants literally failed to see shapes, despite staring straight at them.
-failure to see, caused by inattention, dubbed inattentional blindness
-Which account is correct? Did participants fail to see input? did they see it but, just few milliseconds later, forget
what they'd seen?
-purposes of theory, distinction is crucial, but for moment emphasize two proposals have in common:
-By either account, normal ability to see what's around us, and make use of what we see, dramatically
diminished in absence of attention.
pg 102
-Conscious Perception, Unconscious Perception
Mack and Rock, documented inattentional blindness phenomenon, draw strong claim from these data:
-There is no perception, they claim, without attention.
-However, claim needs important refinement, more precisely, Mack and Rock : no conscious
perception without attention
- why refinement needed, consider following experiment
-Participants shown series of images on computer screen;
-each image contained two horizontal lines surrounded by pattern of black and white dots
-task : decide which of two lines was longer, top or bottom.
- first three trials background dots on screen arranged randomly.
-Trial 4, however, with no warning pattern of dots shown like one in Figure 4.2B, creating stimulus
configuration that reproduces standard geometric illusion, the Muller-Lyer illusion.
-Focusing attention on two horizontal lines, didn't perceive pattern.
-When asked immediately after trial whether noticed pattern in dots, none reported seeing the pattern.
- told directly there had been a pattern ,asked to choose it (our of four choices available);
-90% selected one of incorrect patterns.
-experiment reproduces finding of inattentional blindness.
Pg 103
-Nonetheless, participants influenced by dot pattern.
-In standard Muller-Lyer display, "fins" make top horizontal line in Figure 4.2c appear longer than bottom
horizontal line, though both lines exactly same length.
- dot pattern in Moore and Egeth displays did same,
-95% of participants reported top line in Figure 4.2B longer than bottom one, even though, in display,
both same length.

Document Summary

The stimulus you"re attending to only one of many available to you. This fact invites two crucial observations: seems clear you could choose to pay attention to any of the just mentioned, if you did, would be virtually oblivious to other things on list. Indeed, until you read previous paragraph, probably oblivious to other stimuli mentioned! How do you do this:" how dose one manage to avoid distractors and focus attention in just the manner one wishes, selecting only one input of many:" 2) seems to be one thing cannot do: you cannot pay attention to all things at once. If start musing about the weekend, likely to lose track of what"s on the page; If start planning term paper, won"t finish reading assignment. Of course possible in ome circumstances to divide attention, deal with two different inputs at once.