Lecture 3 Flashcards
1
Q
Limits to attention
A
- Change blindness = The Attention Test
- Attentional Blink
- Inattentional blindness
2
Q
Inattentional blindness
A
- We overestimate how much of the world we are actually aware of
- Very salient (noticeable) things can be missed
- Example = gorilla study
3
Q
Gorilla study (Simons and Chabris, 1999)
A
- Two video styles
1. Transparent = both teams and unexpected event filmed separately then superimposed
2. Opaque = both teams and event filmed simultaneously = people and objects can be occluded
-Two team conditions, one focusing on each colour team
- Two counting conditions
1. Easy = count number of passes
2. Hard = count aerial and bounce passes separately
Results
- Inattentional blindness can be induced easily on healthy ppts
- Occurs more frequently if display is transparent as gorilla doesn’t block players
- Depends on difficulty of task = the more the primary task occupies attention, the less likely the gorilla will be seen
4
Q
Cocktail party problem (Cherry, 1953)
A
- Questions how we focus in on 1 conversation at a time
- Sound segregation = using physical differences to decide which sounds belong together
- Focusing attention to a chosen auditory message
-Johnsrude et al 2013 = a familiar voice is easier to pay attention to and easier to ignore
5
Q
Dichotic listening task
A
- Attempt to study the cocktail problem
- Participant wears headphones and presented with 2 independent audio streams to different ears
- May be asked to monitor single stream for a particular target
- Shadowing = participants asked to repeat out loud the target stream
- Concluded about physical differences
- Had no awareness of content in the ignored stream
6
Q
Broadbent’s Theory = attention as early selection
A
- Parallel input into sensory register
- Inputs filtered on basis of physical characteristics (prevents overload of limited capacity mechanism)
- Only a single input can reach awarenesses
- Accounts for Cherry’s basic findings = unattended stimuli only undergo minimal processing before filtering
- Accounts for findings of dichotic listening task = filter selects input based on most prominent physical characteristics
BUT
- At least some parts of unattended steam are processed semantically e.g. hearing name in separate conversation
- Stimuli that people don’t report experiencing can still change behaviour
- Words that were associated with electric shock elicited a physiological reaction when presented in ignored stream –> showed words processed and identified even when not aware
7
Q
Deutsch and Deutsch = attention as late selection
A
- Process everything cognitively
- Attention selects one item for behaviour
- Bottleneck = limited amount of attentional resources that can be used at one time, happens early during processing
8
Q
Treisman’s attenuation theory
A
- Leaky filter = filter serves to reduce the effect of irrelevant input but still processed
- Input goes through hierarchy of processing from physical features to meaning
- Ignored input stops being processed when capacity limit reached
- Bottleneck can occur at different times
- It explains why partially processed stimuli can sometimes reach awareness
9
Q
Cognitive load
A
- Used amount of WM resources
- Increases when unnecessary demands are imposed on the learner
- When ppts complete 2 simultaneous tasks (1 in central field of view, 1 in periphery) = possible to have high performance in both
- But simple tasks can be very difficult whilst complex tasks can be performed well in
10
Q
What is the field a distinction between? (When does attention selection happen?)
A
- Initially field considered distinction between early and late
- In reality it is flexible and influenced by top-down and bottom-up processes
11
Q
Covert attention: The Posner Cueing Paradigm
A
-Posner showed how attention can be directed to different parts in space = spatial attention
Exogenous study:
- Asked participants to respond as quickly as they can whenever star appeared within one of two peripherally presented rectangles
- Before onset of target one of rectangles briefly flashes
- Found when target appeared within flashing rectangle, participants were quicker to detect the target
- Represents exogenous attention in space
- Exogenous attention= bottom up signals draw attention into the rectangles
- Shorter reaction times is evident at stimulus onset asynchrony (amount of time between flashing rectangle and star)
Endogenous study:
- Star and arrow arrow (left or right) appeared in centre of screen
- Found ppts quicker to detect target on valid trials
- Endogenous attention = based on internal cue
- Choosing to pay attention to particular part of space (direction of arrow) makes you react faster
12
Q
Object-based attention
A
- Attentional selection works on the basis of whole objects
- Activity in brain areas used as an estimate to the degree to which participants attend to a certain thing
Study:
-Viewed hybrid image of a face and a place (one was moving)
-3 conditions:
–> Told to look for repetitions of same face
–> Same house
–> Same motion direction
–When participants looked for repetition:
Motion direction = middle temporal visual area
Face = fusiform face area
Place processing = parahippocampal place area
13
Q
Central capacity theory - Kahneman 1973
A
- A single central capacity (e.g. central executive) that supplies resources throughout system
- Strictly limited but flexible resources
14
Q
How to distribute our limited capacity
A
- Johnson and Zatorre = area in left DLPFC more active during divided attention = this brain region involved in distributing limited capacity to relevant cognitive function
- Johnson, Strafella and Zatorre = TMS of DLPFC disrupts divided attention = provides causal link between DLPFC and attention distribution
15
Q
Visual search
A
- Feature search = target does not share any features with other distracting objects
- Conjunction search = target does share features with distractors e.g. Where’s Wally
16
Q
Feature integration theory - Anne Treisman 1988, 1992)
A
- Explains why visual search is harder under conjunction conditions than feature search
- Features of objects are processed separately and separable from object itself
- Separate features are encoded by perceptual system in parallel even when unconscious
- Pop-out = unique feature that can be processed unconsciously
17
Q
Stages of visual search
A
- Rapid initial parallel process: to identify features, attention-independent
- Then a slower, serial process: to form objects from combining features
18
Q
What are illusory conjunctions
A
- Illusory conjunctions is when proper attention is not paid so features are reported correctly but not binding together
- Ppts are not just guessing and answers can occur with high confidence
19
Q
Evidence against FIT
A
- FIT argues that an object is only an object if attended to but negative priming tasks show semantic processing of unattended stimuli
- Negative priming tasks = hybrid objects in two different colours, attention only paid to objects in one colour
- Find that ignoring object on one trail makes us slower to name same object on next trial
- In semantically related objects, NP occurs in semantic categories
20
Q
Guided Search Theory/ Dual Path Model - Wolfe, 1998
A
- Search more or less efficient depending on whether serial/ parallel processes are deployed
- Every item in display produces an activation map = each item has certain level of activation
- Goal directed search will have higher activation
- Conjunction search is more efficient because parts of the display are lowly activated because items don’t share any features with target so can be ignored
- Selective pathway helps us search in a goal directed way BUT has a bottleneck
- Non-selective pathways allows us to gain a gist of what the scene represents, pre-activates locations in the brain
21
Q
Deficits in feature binding: Balint’s Syndrome
A
Treisman (1999): Patient RM
- Attentional deficit characterised by deficits in binding features of several different objects
- 2 strokes damaging large areas of bilateral occipital-parietal cortex
- Unable to focus attention on more than one object at a time = simultanagnosia
- Made illusory conjunction errors even when seeing objects for 10s
22
Q
Parietal Cortex and Feature Binding
A
Corbetta et al. (1995)
- During conjunction search, temporo-parietal regions increased activation
- Walsh et al. (1995): TMS to parietal lobe disrupts conjunction search but not feature search of healthy ppts
- Esterman et al. (2007): Stimulation of intraparietal sulcus reduces illusory conjunctions in healthy ppts
