Lecture 1

Question 1

What is attention?

Answer 1

Attention is a process of selection, as we have limited cognitive resources, thus we cannot attend to everything going on around us. Eyes scan the world every few seconds to build an image and once that’s done, attention is not attended to the unimportant details (e.g. colours).
● Selective attention selects certain things for attention and sacrifices others.
● Divided attention is multi­tasking; put much less attention into several tasks.
● Control of consciousness is related to attention ­ i.e. you have no attention when you are asleep.

Question 2

__________ selects certain things for attention and sacrifices others.

Answer 2

Selective attention

Question 3

__________ is multi­tasking; put much less attention into several tasks.

Answer 3

Divided attention

Question 4

_________ is related to attention

Answer 4

Control of consciousness

Question 5

Attention can be

Answer 5

1. automatic
2. voluntary
3. voluntary inhibited

Question 6

e.g. movement in peripheral vision causes eyes to move.

Answer 6

attention as automatic

Question 7

e.g. move the line of sight to display attention voluntarily.

Answer 7

attention as voluntary

Question 8

e.g. it may be rude to look somewhere.

Answer 8

attention as voluntary inhibited

Question 9

Explain the cocktail party effect and associated study.

Answer 9

● Cherry (1953) investigated the ‘cocktail party’ problem, where in a room of loud
chatter, people could selectively attend to one conversation (and claim to hear the name, if mentioned). Cherry found this was a result of physical differences i.e. the gender of the speaker, voice intensity and the location all contributed to how well someone could focus on one conversation. Cherry conducted a study, then, using dichotic listening tasks:
● He presented subjects, who were hearing headphones, two different messages in the same voice, one in each side of the headphone. He found listeners couldn’t separate the two messages on the basis of meaning alone.
● He then asked subjects to shadow (repeat out loud) what one of the voices was saying, and found that while subjects knew what the voice they were shadowing was saying, the unattended voice left little or no information. They’d only notice the other voice if the language changed or speech reversed. Physical changes were nearly always noticed. This study provided evidence that there was no processing done on the unattended voice. THUS: Cherry concluded selection is absolute, hearing becomes exclusive to one ear once selection occurred.

Question 10

What would ‘early’ locus of selection mean? What would ‘late’ locus of selection mean?

Answer 10

• A locus of selection was a point of when information is accepted or rejected. An early locus of selection means only attended stimuli is processed.
• A late locus of selection means attention operates only after stimuli is processed.

Question 11

What are the different theories of attention? How do they differ, what evidence is there to support those theories and potential criticisms?

Answer 11

Broadbent’s Filter Theory (1958)

Treisman’s attenuation theory (1960)

Question 12

Broadbent’s Filter Theory (1958)

Answer 12

Broadbent’s Filter Theory (1958): Broadbent used these findings for his theory; that two stimuli presented at the same time to a subject gain access in parallel (at the same time) to a sensory buffer. One of the inputs is then allowed through a filter, on the basis of its physical characteristics, while the other input remains in the buffer for later processing. This filter prevents overloading of the limited­capacity mechanism of the filter. This theory confirms Cherry’s findings, that unattended messages are minimally processed and do not go through the filter.

○ Thus, sound enters and some basic processing occurs but it is not understood semantically. This theory proposes an early locus of selection.

● But the following study provided evidence against Broadbent’s theory: Another study investigated memory tasks. 3 pairs of digits were presented to a subject, with one digit being presented to each ear at a time (so L, R, L, R etc.) however subjects did not recall the digits in that order, they recalled them in groups of 3 by each ear (LLL, RRR). Suggests that either information from the unattended ear is processed past the physical level (or that there is a memory buffer before the filter). What I’m unclear about is what does it mean for there to be an “unattended ear”? Can’t they just be switching between ears?

Question 13

Treisman’s attenuation theory (1960)

Answer 13

Treisman (1960) found within the shadowing task that some subjects would state words from the unattended ear, and this would occur when such words were related to the attended words. Thus, Treisman argued that the filter reduces the analysis of the unattended words, however the filter is more flexible than Broadbent claimed. She also claimed that stimulus analysis followed a hierarchy and the level of analysis reached depending on the processing capacity ­ words that were expected were more likely to be processed. Thus, her theory was ‘leaky filter’.

Question 14

Deutsch and Deutsch (1963)

Answer 14

Deutsch and Deutsch (1963) argued that all stimuli are processed, with the most important determining the response. This theory claims processing and resources are almost limitless, which doesn’t make sense. Treisman and Riley (1969) conducted a study to test this:
○ Subjects were asked to shadow 1 of 2 auditory messages, but to stop and tap when they heard a target stimulus in either message. According to Treisman’s theory, there should be more taps from the attended message, but according to Deutsch and Deutsch, there should have been no difference as both messages are claimed to be fully processed. Results showed that more target words were detected in the shadowed (attended) message.

Question 15

Johnston and Heinz (1978)

Answer 15

Johnston and Heinz (1978) claimed that selection occurs early, and uncertainty leads to processing irrelevant stimuli. As many stimuli are complex, they require early selection and processing.

Question 16

Broadbent’s mechanical switch model:

Answer 16

Broadbent’s mechanical switch model: Broadbent returned, and argued that there is an immediate memory buffer that briefly holds relatively unprocessed information, and that if we switch our attention rapidly to the information in the appropriate buffer, we would be able to process unattended stimuli thoroughly, and this would be one
explanation of how unattended stimuli breakthrough and are processed. This theory was the ‘slippage’ theory. Slippage is when attention shifts to an unattended channel, in contrast “leaky” filters allow some content through even when that channel isn’t being focused on.

Question 17

Lavie’s (1995) perceptual load theory:

Answer 17

Lavie’s (1995) perceptual load theory: Unattended stimuli, naturally, gets less processing than attended stimuli, however there remains still some amount of processing of the unattended stimuli – evidence coming from brain imaging. Lavie (2005) developed a theory with two assumptions:
a. Susceptibility to distraction is greater when the task involves low perceptual load than high perceptual load. Perceptual load depends on things such as the number of task stimuli needed to be processed. High perceptual load leaves no spare capacity for perception of task­irrelevant stimuli.
b. Susceptibility to distraction is greater when there is a high load on executive cognitive control functions i.e. working memory, than when there is a low load. Cognitive control is needed for actively maintaining the distinction between targets and distractors.
A lot of evidence supports this theory.

Question 18

Working memory and cocktail party effect.

Answer 18

Working memory and cocktail party effect.
● Moray conducted and repeated the shadowing experiments, and found only 10% of subjects heard instructions from the unattended ear. However, when the subject’s name preceded the unattended ear instruction, 33% of subjects heard.
● It was later proposed that people with low working memory are likely to hear their own name because they’re more distractible, have less control over their attention and more likely to respond to distractors. There seem to be two distinct theories here, that working memory is related to control over attention or that they don’t need to scan the environment as they “know what’s going on”.

Question 19

Normand, Bouquet & Croizet:

Answer 19

● More pressure when test was presented as an assessment of intellectual ability than
training, pressure increased distraction by task related features and reduced distraction by irrelevant features

● Task was based on eye­tracking and fixation points. Spatial cue was a circle or a cross. Most of the time, the target would appear at the cue, but sometimes it would appear elsewhere.

Question 20

What did Jiang, Costello, Fang and He (2006) experiment of displaying naked images demonstrate about attention between left and right eyes?

Answer 20

What did Jiang, Costello, Fang and He (2006) experiment of displaying naked images demonstrate about attention between left and right eyes?
○ They found that attention cannot be focused on both eyes, and many subjects won’t even see the naked images thus they cannot consciously report anything ­ but did they detect attention? Men had their attention drawn to images of naked women on the left side but no one was able to discriminate which eye the image came up on. This shows that consciousness and attention overlap, but they are indeed different.