Psych 351A: Midterm 2 Review (Attention)

Attention

Cognitive mechanisms by which information is prioritized for processing

What are the 2 sources of attention

1. Exogenous

1. Endogenous

Exogenous attention

Reflexive, automatic, bottom-up

Endogenous attention

Voluntary, intentional, top-down

What are the 2 targets of attention

1. External

2. Internal

External target of attention

Sensory info, includes locations in space, points in time and modality-specific input

Internal target of attention

Memories, rules, goals, etc.

What are the 3 types of attention

1. Overt/covert

2. Momentary/sustained

3. Selective/divided

Selective attention

Ability to focus on one stimuli and ignore all others, some info is filtered and some is promoted for further processing

T or F: We do not attend to a large portion of info in the environment

T, we tune out what is not needed

Flanker task

Participants are to focus on detecting target and ignoring flankers. Respond left to A/B and right for C/D (e.g. XAX)

What is found regarding neutral, compatible and incompatible trials in the Flanker Task

Compatible - Leads to quicker RT (e.g. BAB)

Incompatible - Leads to slower RT (e.g. CAC)

Neutral - Leads to moderate RT (e.g. XAX)

Flanker compatibility effect

RT is quicker when additional information is congruent with what is being attended too

(The difference between the incompatible and compatible trials)

What causes the flanker compatibility effect

The leftover attentional capacity that is not occupied by the simple task makes us easily distracted

What was Lavie’s (2005) X’s and O’s experiment

One responds left for X and right for N over two conditions on congruent and incongruent trials:

Low-load - O’s as distractors

High-load - Variety of letters as distractors

What did the X’s and O’s experiment find regarding the flanker effect

That there is a big difference on the low-load condition, but a very small difference on the high-load condition, lends merit to the attention capacity claim

Cognitive load

How much a person’s cognitive resources are used to accomplish a task, can be a low-load or high-load

What was Forster and Lavie’s (2008) Smiling Dog experiment

Same as X’s and O’s experiment except sometimes a smiling dog would appear, found that they reaction time was longer with distractor but only in the low-load condition

What are the two types of determinants of attentional focus

1. Bottom-up Determinants

2. Top-down Determinants

Stimulus salience

Bottom-up determinant, areas that stand out capture attention (particularity colour, motion, contrast and orientation)

Salience map

Calculated salinity of specific areas in a scene summarized

Scene schema

Top-down determinant, knowledge about what is contained in typical scenes can help guide fixations from one area to another

Vo and Henderson’s (2009) unexpected scene photo test

Asked participants to view a kitchen scene with either a pot or printer on the stove that was either floating or resting.

Found that objects that don’t fit the scheme attract more attention (bigger difference for compatible object and non-compatible)

Task demands

Top-down determinant, goals and steps of a task guide attentional focus causing eye movements to precede motor actions by a fraction of a second

How is task demands tested

Showing individuals a typical scene and tracking their eyes, when plotted circles are locations of things viewed and lines are the movement eyes make (like connect the dots), bigger circle = longer time spent looking

Land and Hayhoe (2001) PB&J experiment

Tracked eye movements of participants while they were making a peanut butter and jam sandwich, or making a cup of tea and tried to anticipate hand actions

What did the PB&J experiment find

Found that whole body movement occurred prior to visual fixation, which in turn preceded manipulation of object

(Body movement → visual fixation → manipulation of object)

What did Yarbus (1967) find regarding eye movements

That eye movements and amount of attention is based on what one is attempting to understand about a scene

Covert attention

Shifting’s one attention without physically moving their eyes

What was the Posner (1978) Arrow experiment and what were the findings

Would show one a screen that had a cue arrow pointing left, right, or neither. Then they would show a square that was either on the right or left. 80% of the time the cue was valid.

Found that people responded much quicker if they were shown a valid cue, and slower with an invalid cue

What was the source of and target of attention in the Posner task

Source - endogenous

Target - external

Posner’s spotlight theory of attention

Argues that one’s attention works much like a spotlight shining on what one is attending to 

How is slower RT time on invalid trials explained by Posner’s spotlight theory

Because the individual must take time to covertly shift attention by disengaging, shifting and reengaging attention

Spatial cueing experiment

Same as Posner’s task but used a box as a cue instead of an arrow

Demonstrates an exogenously sourced task and external focus of attention

Inhibition of return

When there is a long delay between cueing and target appearing, one will actually perform slower if the cue is valid then when it is invalid

Why does inhibition of return occur according to Posner and Cohen

Because they argue that one would shift their attention to the cue, but because nothing happened they would be inhibited from looking over there again when the target was presented due to the assumption that it is a “false alarm“

Garavan Task

One is shown either a triangle or a square and asked to keep count of how many of each they had seen, one is able to choose when to advance to the right shape

What did the Garavan task find regarding RT of participants

Those who saw the same shape after an advance would respond much quicker than those who saw a different shape after advance (response meaning advance to next shape)

What is the source and target of attention in the Garavan task

Source - endogenous

Target - internal

What was the question Egly et al (1994) tried to answer with their Rectangle Spotlight Experiment 

If visual attention truly behaved like a spotlight

Methods of the Egly et al (1994) Rectangle Spotlight Experiment

Had two identical rectangles on either side of the centre of cross, one was cued on one end of one of the rectangles and then a target appeared in one of the 4 locations (each end of respective rectangles), reaction time was measured

What were the findings of the Egly et al (1994) Rectangle Spotlight Experiment

That people responded quicker not only if cued, but also if the cue was on the same rectangle but invalid, demonstrated that attention focuses on objects as a whole and not on a specific part of an object

Divided attention

Sharing of one’s attention among multiple stimuli/tasks

What are the two models of divided attention

1. Broadbent’s filter model

2. Treisman’s Attenuation Model

Broadbent’s filter model

Made up of three parts, an all or none system which claims attention filters messages before information is analyzed for meaning

What are the 3 parts of Broadbent’s filter model

1. Sensory memory

2. Filter

3. Detector

What is the role of sensory memory in Broadbent’s filter model

Holds all incoming information for a fraction of a second before transferring all info to the next stage

What is the role of the filter in Broadbent’s filter model

Identifies attended message based on physical characteristics in which only the attended message is passed on to the next stage

What is the role of the detector in Broadbent’s filter model

Processes information to determine higher-level characteristics of the message and then passes the info to STM (which can then hold info for 10-15s and choose to transfer it to LTM)

Triesman’s attenuation model

System that proposes that multiple qualities of perceptions can be used to select attended vs unattended message, has 2 components

What are the 2 components of Triesman’s attenuation model

1. Attenuator

2. Dictionary Unit

Attenuator component of Triesman’s model

Works like a leaky filter, analyses incoming messages in terms of physical characteristics, language, and meaning

Attended message is let through at full strength, unattended is let through at weaker strength

Dictionary unit component of Triesman’s model

Contains words each of which have thresholds for being activated, the more common/important the word the lower threshold and vice versa

What was the Shneider and Shiffrin (1977) 4 frame experiment

Individual is asked view a set of 1-4 target characters and then shown test frames that contained masks, distractors and sometimes a target, asked participants if target was a part of one of the frames.

Had two types of conditions (consistent and varied mapping)

Consistent mapping condition (Schneider and Shiffrin experiment)

Target stimulus and distractors stayed the same throughout trials

Could be carried out through automatic processing which improved as trials commenced

Varied mapping condition (Schneider and Shiffrin experiment)

Distractors may become targets and targets may become distractors over trials

Participants never achieved automatic processing, only controlled processing

What were the findings of the 4 frame (divided attention) experiment

Individuals did far better on the consistent condition even if the frame duration was 100ms slower for the varied conditon

Pashler (1992) Auditory and Visual Task

Had participants complete an auditory-vocal task and a visual-manual task in order, had varying levels of stimulus onset asynchrony and viewed reaction times

What was the auditory-vocal task in Pashler’s experiment

Stimulus was a low, medium or high tone that one had to respond to with one, two or three audibly

What was the visual-manual task in Pashler’s experiment

Stimulus was either (O - -), (- O -), (- - O) and one had to respond with index, middle or ring figure

Stimulus onset asynchrony (SOA)

The time between when the first and second stimuli were presented

What were the findings of Pasher’s experiment

That there was a response selection bottleneck

If one was engaged in selecting the response for one task, then they could not proceed to select a response on a different task, regardless of how much practice they had

Occurred when SOA was too short

What was Schumacher et al. (2001) Rebuttal to Pashler

Argued that the bottleneck was actually a form of strategic control that participants used to ensure that task 1 was completed prior to task 2

How did Schumacher et al. (2001) prove their argument was correct experimentally

Did the same task as Pasher but instead asked participants to complete each task as quickly as possible, this made it for some participants they could respond very quickly to both tasks (some participants still showed data similar to Pasher’s findings)

What were the two major assumptions that caused Pasher’s experiment findings to be incorrect

1. Assumed that the task instructions did not effect results

2. Assumed that all participants performed the task in the same way

What is the visual search experiment

Participants search for a particular target (may be present/absent) among distractors

What are the two types of trials in the visual search experiment

Feature - target differs based on a single feature (e.g. red/blue)

Conjunction - target differs based on a combination of features (e.g. red & square vs blue & square or red & circle)

What are the two types of strategies that can be assumed to be used for the visual search and what are the expected findings

1. Parallel - assumed the results would be consistent on RT no matter the number of distractors (flat line)

2. Serial - positive slope, RT would increase with distractors

Serial exhaustive

One looks through all the items even if target is found, slope would be positive, if it was present then the slope would be positive but much shallower and vice versa

Serial self-terminating

One stops after finding the target, slope would be positive and look similar no matter if the target was absent or present

What were our coglab findings and were they consistent with the hypothetical results

Yes, exactly the same, feature stayed the same at around 750ms RT, and conjuction present/absent started at 850ms and peaked around 1250ms and 2250 ms respectively

How many ms did it take to check each item in a serial exhaustive task

20ms, 1200ms to test all items

T or F: Treisman’s findings were consistent with our coglab

T, consistent for both feature and conjuntion search

Feature integration theory

Focused attention is required to bind different features into consciously experienced wholes

Processing changes depending if information is a feature search or conjunction search

What does feature integration theory say regarding detection through feature difference

It is rapid, parallel and target pops-out

What does feature integration theory say regarding detection using conjunction search

It is slow, serial and target needs to be found methodically (does not pop-out)

What are the two stages that FIT says are required to bind different features into consciously experienced wholes

1. Preattentive stage

2. Focused attention stage

Preattentive stage

Analyzes into features, is automatic, takes no effort and one is not aware of process

Focused attention stage

Combines features together, attention is needed

Triesman and Schmidt shapes and numbers (feature detection) experiment

Shapes and 2 single digit numbers on each side were flashed and then masked, participant does one of two tasks:

Task 1 - first report numbers, then objects

Task 2 - first report objects, ignore numbers

Task 1 findings

Participants reported illusionary conjunctions, occurred because features are free floating since attention focused on numbers

Illusionary conjunctions

Combinations of features from different stimuli

Task 2 findings

Participants showed fewer illusionary conjunctions due to focusing attention on features of objects

Psychological refractory period

Delay in response time to a second stimulus that occurs because the brain must first process the first stimulus