Lecture 4- Attention bottlenecks

In human information processing, what are serial bottlenecks?

The filter points at which it is no longer possible to process incoming perceptual information (e.g. song, light smell) from our senses  simultaneously

When we're no longer able to process a lot of information at once

• At which point does this occur?

What do early selection theories say about information processing?

Theories of attention proposing that the selection of information occurs early in information processing

What do late selection theories say about information processing?

Theories of attention proposing that the selection of information occurs late in information processing

Load theory (Lavie, 2005, 2010) distinguishes between…?

Perceptual and cognitive load (perceptual demands of the task vs. burden placed on the cognitive system by the current task)

• High perceptual load is associated with low distractibility whereas high cognitive load is associated with high distractibility

According to load theory, brain activation associated with distractors should be less when…?

Individuals are performing a task in involving high perceptual load

• Found with visual tasks and distractors and with auditory tasks and distractors

Load theory- what did Burnham et al. (2014) find in terms of distraction effects?

Distraction effects on a visual search task were greater when participants performed another task placing high demands on the cognitive system (greater when cognitive or working memory load is high)

Load theory assumes the effects of perceptual and cognitive load are independent. However, Linnell and Caparos (2011) found…?

Perceptual and cognitive processes interacted

• Perceptual load only influenced attention as predicted when cognitive load was low

• Effects of perceptual load are not automatic but depend on cognitive resources being available

Give 2 limitations of load theory

1. The terms 'perceptual load' and 'cognitive load' are vague, making it hard to test the theory

2. The prediction that high cognitive load is associated with high distractibility has been disproved when the task and distracting stimuli are easily distinguishable

What is the cocktail party phenomenon (Cherry, 1953)?

How can we follow just one conversation when several people are talking at the same time?

McDermott identified two problems listeners face when attending to one voice among many. What are they?

1. Sound segregation; deciding which sounds belong together

2. Directing attention to the sound source of interest and ignoring the others

Various top-down factors based on listeners' knowledge and/or expectations are involved in early auditory processing. Give an example

Listeners are more accurate at identifying what one speaker is saying (in the context of other speakers) if they have previously heard that speakers voice in isolation- McDermott (2009)

Cocktail party phenomenon- what did Shamma et al. (2011) find in terms of distinctive features?

If listeners can identify one distinctive feature of the target voice, they can distinguish its other sound features via temporal coherence

Cocktail party phenomenon- what did Golumbic et al. (2013) find in terms of visual information?

Processing of the attended message was enhanced when listeners saw a video of the speaker talking

• Individuals can also use visual information to assist them in understanding what a given speaker is saying

Describe the procedure of the dichotic listening task

Participants wear a set of headphones

• Then hear two messages simultaneously (one in each ear)

• One is relevant, one is not

 Participants must 'shadow' (repeating back the message from one ear) the relevant channel and ignore the irrelevant information from the other

After the experiment, ask questions about the channel they should have blocked

• Seeing what they noticed in that message

Give a limitation of the dichotic listening task

However, listeners don't normally engage in shadowing so the task is artificial and it increases listeners' processing demands

Dichotic listening task- what do participants fail to notice in the unattended message?

Semantic features like..

• Message played backwards

• Several words repetitions

• Message played in a foreign language

Dichotic listening task- what do participants notice in the unattended message?

Physical features like…

• Male or female speaker

• Speech changing from male to female speaker

• Whether it was a human voice or just noise

What does the dichotic listening task tell us about the selection of information?

Semantic features (about the meaning) were not noticed but some physical features were

• Seems like the selection of information was mostly based on physical features and not semantic features

What does Broadbent’s (1958) filter theory say about information selection?

Sensory information comes through the system until it reaches a bottleneck

• Information can be selected based on a physical selection criterion (e.g. ear or pitch)

• The person filters out the information based on physical characteristics/features

The information we choose to select continues to be processed, but for the other information it's completed blocked

Filter theory- what were the methods in Moray’s (1959) study?

A dichotic listening task;

• What happens when we introduce the name of the participant in the non-shadowed ear (irrelevant channel)?

Filter theory- what were the findings in Moray’s (1959) study?

33% of the participants detect their name (semantic characteristics)

 Seems to challenge Broadbent's filter theory

• According to this theory this wouldn't be possible

Filter theory- what were the methods in Gray and Wedderburn’s (1960) study?

A dichotic listening task

• Message 1; string of numbers

• Message 2; meaningful sentence

 Alternated the ear to which words in a meaningful sentence was played

Filter theory- what were the findings in Gray and Wedderburn’s (1960) study?

When asked to report what they listened to, participants would have no problems in reporting the meaningful sentence correctly

• According to Broadbent, participants shouldn't have been able to do this at all

What does Gray and Wedderburn’s (1960) study suggest?

• Results suggest that participants can alternate between channels (select information) based on the semantic properties of the stimuli

• The finding proposed a problem to the early selection theory of auditory attention (filter theory)

Filter theory- what were the methods in Treisman’s (1960) study?

A dichotic listening task

• Meaningful message in one ear and meaningless in the other

• The meaningful message is switched to the other ear (non-shadowed ear)

Filter theory- what were the findings in Treisman’s (1960) study?

Some participants switch ears and keep shadowing meaningful message, breaking the instruction

• Some others keep attending the message in the first ear

What does Treisman’s (1960) study suggest?

• Results suggest that selection of information can be flexible

• Participants can sometimes select information based on physical characteristics (e.g. a particular ear) and sometimes based on the semantic characteristics (e.g. meaning of the message)

• Therefore another theory is needed to account for this flexible type of information selection

What does Treisman’s (1964) attenuation theory say about information selection?

Another type of early selection theory

 Sensory information comes though the system until it reaches an attenuator

• Information is attenuated (weakened), not filtered out

• Information can be selected based on a semantic selection criterion (e.g. meaning of the message)

 Why we can detect our name or some semantic information in some cases

Treisman also argued top-down processes (e.g. expectations) are important. Why did they think this?

Listeners performing the shadowing task sometimes say a word from the unattended input (mostly occurs when the word on the unattended channel is highly probable in the context of the attended message)

What does Deutsch and Deutsch’s (1963) late-selection theory say about information selection?

All information is processed completely (physical properties and meaning) without attenuation

 The bottleneck (capacity limitation) is in the response system and not the perceptual system

• Can only respond to a certain amount of information at once

Most important/relevant stimulus determines the response

Summarise what each theory says about unattended input

• Broadbent; little/no processing of unattended auditory messages

• Treisman; flexibility in the processing of unattended messages

• Deutsch and Deutsch; reasonably thorough processing of such messages

What did Aydelott et al. (2015) find in terms of attended target words?

Performance on attended target words was enhanced when unattended words related in meaning were presented shortly before (meaning was processed)

• Support for attenuation theory?

What did Li et al. (2011) find in terms of words with a special significance?

Often more processing of unattended words that have a special significance for the listener

• Unattended weight related words (e.g. fat) were processed more thoroughly by women dissatisfied with their weight

What did Coch et al. (2005) find in terms of auditory inputs and ERPs?

Asked listeners to attend to one of two auditory inputs and to detect targets presented on either input

• ERPs as a measure of processing activity

• ERPs 100ms after target presentation were greater when the target was presented on the attended message

• More processing of the attended targets

What was the research question in Wood and Cowan’s (1995) study?

Which of the three theories is correct?

• Filter theory, attenuation theory, or late-selection theory?

What were the methods in Wood and Cowan’s (1995) study?

Dichotic listening task;

• Introduce participants' name in the irrelevant channel

• Monitor the percent of shadowing errors

Wood and Cowan- what would happen if filter theory was correct?

• Participants should only detect their name if their attention has wandered in the irrelevant channel (not doing the task properly)

• Would expect participants to be unable to process semantic information

• Prediction; more shadowing errors before the presentation of the name

Wood and Cowan- what would happen if attenuation theory was correct?

• The name should activate the appropriate lexical unit in memory only weakly (low signal initially but slow increase in understanding)

• Detect a low signal, takes some time as the information is attenuated- participants then think they're talking about them so switch to the other ear to hear what's being said

• Prediction; more shadowing errors after the presentation of the name

Wood and Cowan- what would happen if late-selection theory was correct?

• Participants would detect their name routinely (all information is processed fully, immediately go to the wrong ear)

• Would hear their name in the irrelevant channel very clearly and loudly

• Prediction; more shadowing errors during the presentation of the name

What were the findings in Wood and Cowan’s (1995) study? Which theory seems to be correct?

• 34.6% of the participants recall hearing their name in the channel to be ignored

• More shadowing error after the presentation of the name

• These results support attenuation theory

What does Posner (1980) say about visual attention?

Visual attention is like a spotlight

• Illuminates a fairly small area, little can be seen outside is beam and it can be redirected to focus on any given object

What do Eriksen and St. James (1986) say about visual attention?

Attention is more flexible

• More like a zoom lens- can increase or decrease the area of focal attention

• E.g. car drivers narrowing attention after spotting a potential hazard

Attention as a zoom lens- what were the methods in Muller et al.’s (2003) study?

• Observers saw 4 squares in a semi circle and cued to attend to one, two or all four

• 4 objects presented (one in each square)

• Observers decided whether a target (e.g. a white circle) was among them

Attention as a zoom lens- what were the findings in Muller et al.’s (2003) study?

Performance was best with the smallest attended region and worst with the largest (as predicted by the zoom lens theory)

Attention as a zoom lens- what were the methods in Goodhew et al.’s (2016) study?

Focused on temporal perception- was a disc presented continuously or were there two presentations separated by a brief interval?

Attention as a zoom lens- what were the findings in Goodhew et al.’s (2016) study?

• Spotlight size had no effect on temporal acuity, which is inconsistent with the theory

• Spatial resolution is poor in peripheral vision, whereas temporal resolution is good

• A small attentional spotlight is more beneficial for spatial than temporal acuity

What do Awh and Pashler (2000) say about visual attention?

Attention is more like multiple spotlights

• Sometimes exhibit split attention- directed to two or more non-adjacent regions in space

Attention as multiple spotlights- what were the methods in Awh and Pashler’s (2000) study?

• Identifying two digits presented to two cued locations a little way apart

• On some trials a digit was presented between the two cued locations

Attention as multiple spotlights- what were the findings in Awh and Pashler’s (2000) study?

• According to zoom-lens theory, the area of maximal attention should include the two cued locations and the space in between

• However, found that detection of the digits presented in the middle was poor

• Support for the split attention approach

What is a limitation of all 3 visual attention theories?

Metaphors (e.g. attention is a zoom lens) are used to describe experimental findings but they fail to specify the underlying mechanisms (Di Lollo, 2018)

Describe the procedure of Neisser and Becklen’s (1975) visual shadowing task

• Participants watched superimposed videotapes (hands clapping and people passing a ball to each other)

• Instructed to pay attention to one of the two films and watch for odd events (e.g. shaking hands)

Describe the results of Neisser and Becklen’s (1975) visual shadowing task

• When asked to monitor both films for odd events, ppts experienced great difficulty and missed many of the critical events

• We can also filter visual information

• Seems like there's a limited amount of visual information we can process at once (similar to auditory information)

What is intentional blindness?

Refers to the phenomenon in which we are unaware (unable to detect) what is happening in our direct field of view if we are not paying attention to it

Describe the procedure of Mack and Rock’s (1998) intentional blindness study

• 5000 ppts perform a perceptual task in which they judged whether the horizontal or vertical bar of a cross was longer (boring task)

• After a number of trials, there would be a surprise trial in which an additional stimulus such as a rectangle would appear on the screen along with the cross

• Stimulus only shown for 200ms

Describe the results of Mack and Rock’s (1998) intentional blindness study

Ppts were more likely to detect their surprise stimuli if it was their name (consistent with the cocktail party phenomenon)

• Can process some semantic information from the non-relevant task

Describe the procedure of Simons and Chabris’ (1999) intentional blindness study

• Ppts are asked to watch a video with two teams (black and white)

• Must either count the number of passes from the black or white team

• In the middle of the game, a person in a black gorilla suit walks through the room

Describe the results of Simons and Chabris’ (1999) intentional blindness study

Almost all participants fail to notice the gorilla when tracking the team in white (92% fail)

How do we process information in the visual field?

Information about the left side of the visual field goes to the right brain (and vice versa)

• Process visual information contralaterally in the brain

What is visual neglect?

Absence of awareness of stimuli presented to the opposite side of the brain damage (contra-lesional side)

• Most neglect patients have damage in the right hemisphere (lack of awareness of stimuli in the left visual field)

• The right hemisphere is also dominant in spatial attention in healthy individuals

What kinds of tasks help us detect this damage (neglect)?

Sacks (1985)

• The task was to put slashes through all circles (right hemisphere damage)

• What happens to the circles in the left part of the visual field?

What is space-based or egocentric neglect?

Lack awareness of stimuli on the left side of the visual field

What is object-based or allocentric neglect?

Lack of awareness of the left side of objects

What is seen with damage to the right parietal lobe?

The right parietal lobe is more important in the spatial allocation of attention (directing our attention)

• Important for global features

• Able to reproduce specific features of the pictures but not global features

What is seen with damage to the left parietal lobe?

The left parietal lobe is important for specific features

• Able to reproduce global features of the pictures but not specific features

What is extinction in neglect patients?

Involves a failure to detect a stimulus presented to the side opposite the brain damage when a second stimulus is presented to the same side as the brain damage

What did Molenberghs et al. (2012) find in terms of what brain areas are damaged in neglect patients?

Meta-analysis

• Found the main areas damaged are in the right hemisphere

• Nearly all the areas are within the ventral attention network, suggesting brain networks are damaged rather than simply specific brain areas

However heterogenous condition- brain areas damaged vary considerably across patients

What is typically seen in neglect patients in terms of conscious awareness and processing?

Neglect patients generally report no conscious awareness of stimuli presented to the left visual field

• Doesn't mean they're not processed

Conscious awareness and processing- what did Vuilleumier et al. (2002b) find?

• Presented extinction patients with two pictures at the same time, one to each visual field

• Showed very little memory for left-field stimuli, however demonstrated a facilitation effect for left-field pictures when identifying degraded pictures

• Suggests they had been processed to some extent

Conscious awareness and processing- what did Sarri et al. (2010) find?

• Found extinction patients had no awareness of left field stimuli

• However, these stimuli were associated with activation in early visual processing areas, indicating they received some processing

What did Corbetta and Shulman (2011) find in terms of networks damaged in neglect patients?

The bottom-up ventral attention network is damaged in patients with neglect

• Also assumed that damage to the ventral network impairs the functioning of the goal-directed dorsal attention network (even though it's not damaged itself)

• This is due to the interaction between the two networks

Research into neglect has produced several important findings. Give 3 examples

1. Neglect and extinction patients can process unattended visual stimuli in the absence of conscious awareness of those stimuli

2. Most neglect patients have damage to the ventral attention network leading to impaired functioning of the undamaged dorsal attention network

3. Extinction occurs because of biased competition for attention and reduced attentional capacity

Give 4 limitations of research into neglect

1. It is hard to produce theoretical accounts applicable to all neglect or extinction patients because the precise symptoms and regions of brain damage vary considerably across patients

2. Neglect patients vary in their precise processing deficits but this has been de-emphasised in most theories

3. The precise relationship between neglect and extinction remains unclear

4. The dorsal and ventral networks generally interact but the extent of their interactions remains to be determined