Attention - PowerPoint PPT Presentation

About This Presentation
Title:

Attention

Description:

Only information that passed the filter received further ... (Macaque monkey) Signal. Analysis. Receptive Field. Attention Effects in. Single Neuron Responses ... – PowerPoint PPT presentation

Number of Views:65
Avg rating:3.0/5.0
Slides: 46
Provided by: awh
Category:

less

Transcript and Presenter's Notes

Title: Attention


1
Attention
  • Part 2

2
Early Selection Model (Broadbent, 1958)
FILTER
input
detection
recognition
Only information that passed the filter received
further analysis (e.g. meaning).
3
Late Selection Theory (Deutsch Deutsch, Norman)
FILTER
input
detection
recognition
4
Early Attenuation Model (Treisman)
FILTER
input
detection
recognition
5
Lab Feature Search
Parallel processing of simple visual features
(e.g., color).
6
Typical Results for Feature Search
Reaction Time (msec)
6 10 20
30
of items in display
7
Conjunction Search
When targets are defined by
  • Combination of features (e.g., red AND
    horizontal)
  • Spatial arrangements of features (e.g. black
    above white)

8
Lab Conjunction Search
Find the blue square
9
Lab Conjunction Search
10
Treismans Results for Conjunction Search
Reaction Time (msec)
2 4 6 10 20
30
of items in display
11
Lab Voluntary Cueing
Valid Trials 70
Invalid Trials 15
12
Voluntary Cueing
  • Same result for short and long cue-to-target
    delays (short green, long blue)

13
Lab Automatic Cueing
Cued Trials ??

14
Automatic Cueing
  • For short cue-to-target delay (green), same
    result as for voluntary cueing (validly cued
    faster than invalidly cued)
  • For long cue-target delays, the reverse pattern
    (inhibition of return)

15
Neurological Deficits in Visuo-spatial attention
  • Hemi-spatial Neglect
  • lesion in right temporo-parietal junction
  • Inability to
  • attend to the left side of visual space, and thus
    to
  • be aware of visual stimulus in the left visual
    field
  • Represent spatial relations.

16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
Line-bisection task
20
(No Transcript)
21
(No Transcript)
22
LVF
RVF
Right
Left
23
LVF
RVF
Right hemisphere
Left hemisphere
24
To study the neural substrate of visuo-Spatial
Attention, we need
  • A patient group
  • Hemispatial neglect
  • A simple method
  • Spatial Cueing
  • A cognitive theory
  • Posners three stage model

25
  • Disengage
  • stop attending to what is currently being
    attended
  • Move
  • refocus spotlight on new location
  • Engage
  • begin attending to new stimulus

26
Spatial Cueing
Cued Trials

27
Which part of the brain is the source of
attention? Where does attention have its effects?
(Relay Station)
(Detects Edges)
(Color, Form)
V4
Parietal Cortex
(Location, How to reach or act upon)
(Shape, Object Recognition)
Inferotemporal Cortex
28
Memory/Attention Task (fMRI / ERP)
time
29
Regions of Interest
LVF
RVF
30
(No Transcript)
31
Single-Unit Recording
Receptive Field
(Macaque monkey)
spike single neurons action potential
Signal Analysis
Stim 1
Stim 2
32
Attention Effects in Single Neuron Responses
Attended bar
Frequency of spikes
Unattended bar
100 msec
(Robert Desimone, NIH)
33
No Attention Effects Here
(Relay Station)
(Detects Edges)
(Color, Form)
V4
Attention Effects Here
Parietal Cortex
(Location, How to reach or act upon)
(Shape, Object Recognition)
Inferotemporal Cortex
34
Conclusions from Neuroscientific Evidence
Early visual processing IS affected by selective
attention. This is a challenge for a pure late
selection model. BUT, it does not mean that
late selection is not occurring.
35
Automatic vs. Voluntary Priming
Warning signal
Test signal
AA
neutral
(Posner Snyder, 1975)
36
S
KK
70
Often misleading
15
PP
P
Automatic Priming
Automatic Priming
Voluntary Priming
37
High validity
Low validity (often misled)
P -gt P
P -gtG
faster
P -gtG
P -gtP
Difference between experimental and neutral
conditions
15 primed
70 misled
70 primed
15 misled
Automatic Priming Benefit without a cost
Voluntary Priming Benefit with cost
slower
38
Automatic vs Voluntary priming (part 2) Neely
(1977)
  • If you see a body part as a Prime, expect a
    building part as a target. For example,
  • Body -gt door
  • some pairs were semantically related, but
    unexpected
  • body -gt heart

39
Priming Results
  • Blue Expected pair
  • Body -gt door
  • voluntary priming
  • Evolves with time
  • Green Related but unexpected
  • Body -gt heart
  • automatic priming, followed by a cost from
    voluntary priming

Cost facilitation (ms)
Cue-target delay (ms)
40
Question
  • Predict pattern of performance when
  • the delay between cue and target is very short,
  • the cue-target delay is longer
  • For automatic priming
  • For voluntary priming

41
3 meanings of the word Attention
1. Selectivity only aware of a subset of
stimuli--selective attention.
2. Capacity Limitations limited ability to
handle different tasks or stimuli at
once--divided attention.
3. Sustained mental effort limited ability to
engage in protracted thought, especially on the
same subject--vigilance.
42
Attentional Blink
P
S
B
O
N
K
G






You will see a stream of letters rapidly
presented in the center Group 1 memorize any
vowels Group 2 memorize any vowels and red
letters
43
S
Target 1
B
Encoding into Working Memory
N
Target 2
O
44
Attentional Blink Early or Late Selection?
  • Instead of letters, use words.
  • An initial word establishes context (e.g., milk)
  • Target 2 is a word that is semantically related
    to the context word or not (e.g., sugar, shoes)
  • When subjects fail to report T2, look at their
    brain waves (ERPs) to assess whether the meaning
    of that target has been processed or not
  • N 400 (ERP marker of semantic processing)

45
Spotlight metaphor
- metaphors are not right or wrong, they are
useful or not
Write a Comment
User Comments (0)
About PowerShow.com