Cognitive Psychology Class Notes > Attention



What is Attention?

  • ability to focus on a task
  • ability to concentrate
  • refers to the allocation of processing resources (Anderson, 1995) (assumes limited resources)

Different Aspects of Attention

  • selective attention
  • divided attention
  • automaticity

Selective Attention

  • difficult to attend to more that thing at the same time
  • trying to attend to one task over another requires selective attention

Selective Attention (Visual)

  • classic Stroop task (1935): (show stimuli)
  • slower to name color when word says a different color than to name the color of an colored square
  • why does this happen?
    • reading is an automatic process
    • color naming is a controlled process
    • automatic process of reading interferes with our ability to selectively attend to ink color
  • variations on the Stroop task (lab activity)

read colored words vs. name colors of colored words

Selective Attention (Auditory)

dichotic listening (Cherry, 1953; Moray, 1959)

[in text] in a typical dichotic listening task, Ss hear two messages simultaneously...one message in one ear and one message in the other;

as they are listening, they are asked to 'shadow' one of the messages (i.e. repeat back the words from one message only); most Ss can do this

very little about the unattended message is processed:

can tell whether it was a human voice or a noise

can tell whether the voice was male or female

this information is limited

cannot tell what language was spoken

cannot report any of the words spoken, even if the same word was repeated over and over again

Treisman (1960) shadowing study

what does this tell us?

theoretically, difficult to attend to two things at once; can tell us what can draw attention...such as hearing your name in the unattended ear but when your attention gets drawn to the unattended ear, you lose information from tha attended ear (cocktail party phenomenon.)

practically, don't think you can do more that one at a time without one or both tasks suffering (e.g. reading a book for class and watching Melrose Place

Theoretical Interpretations of Selective Attention

Bottleneck theories or filter theories (Broadbent, 1958)

bottleneck is a mechanism that limits the amount of information to be attended to

what gets through? what is selected and when?

early selection

Broadbent (1958) proposed that physical characteristics of messages are used to select one message for further processing and all others are lost

attenuation

Treisman (1964) proposed that physical characteristics are used to select one message for full processing and other messages are given partial processing

late selection

Deutsch & Deutsch (1963) proposed that all messages get through, but that only one response can be made (late selection)

Treisman & Geffen (1967) tests between attenuation and late selection -- guess who wins?!

dichotic listening + detect target words in either channel (tap upon detection)

detection much worse in unattended channel, supporting attenuation...if late selection, detection should be no problem since all info is getting through

Divided Attention and Dual Task Performance

difficult to attend to more that thing at the same time

trying to attend to two stimuli at once and making multiple responses rather than making one response to multiple stimuli (interference)

Theoretical Interpretations of Divided Attention

Capacity Theories

  • limited amount of resources available to conduct tasks (Kahneman, 1973)
  • multiple resources, only one cognitive process can occur at a time (Pashler)

Automaticity

automatic processing:

  • does not require attention
  • driving a car & listening to the radio
  • reading (as in the Stroop task)

controlled processing:

  • requires attention

Feature Integration Theory (Treisman & Gelade, 1980, Treisman, 1992. 1993)

theory of attention and perceptual processing

a) sometimes process all parts of a scene in parallel (at the same time)

b) sometimes process parts of the scene serially (one at a time)

respectively, testing two types of processing:

a) processing that involves divided attention (automatic registration of features in parallel)

b) processing that requires focused attention (a more demanding kind of processing that is required when objects are more complex)

respectively, using two types of stimulus situations to test two different types of processing:

a) isolated features T I Y

b) combination features T I Z

Treisman and Gelade (1980) hypothesized that if isolated features involved divided attention and targets could be identified in parallel with fillers, and combination features involved focused attention and targets could be identified serially with the fillers, then:

isolated features RTs < conjunction features RTs

they found with T I Y = 800 ms and T I Z = 1200 ms

what does this tell us?

theoretically, provides support for feature integration theory -- people must focus attention on a stimulus before they can synthesize its features into a pattern

practically, this is not just visual...auditory too...you must focus your attention on complex incoming information in order synthesize it into a meaningful pattern