Title: Dynamic attention and predictive tracking
1Dynamic attention and predictive tracking
Lomonosov Moscow State University Cognitive
Seminar, 6/10/2004
- Todd S. Horowitz
- Visual Attention Laboratory
- Brigham Womens Hospital
- Harvard Medical School
2lab photo
David Fencsik
Randy Birnkrant
Jeremy Wolfe
Linda Tran (not pictured)
3Multi-element visual tracking task (MVT)
- Devised by Pylyshyn Storm (1988)
- Method for studying attention to dynamic objects
4Multi-element visual tracking task (MVT)
- Present several (8-10) identical objects
- Cue a subset (4-5) as targets
- All objects move independently for several
seconds - Observers asked to indicate which objects were
cued
5Demo
demo
mvt4
6Interesting facts about MVT
- Can track 4-5 objects (Pylyshyn Storm, 1988)
- Tracking survives occlusion (Scholl Pylyshyn,
1999) - Involves parietal cortex (Culham, et al, 1998)
- Clues to objecthood - Scholl
7Accounts of MVT performance
- FINSTs (Pylyshyn, 1989)
- Virtual polygons (Yantis, 1992)
- Object files (Kahneman Treisman, 1984)
- Object-based attention
8These are all (partially) wrong
- FINSTs (Pylyshyn, 1989)
- Virtual polygons (Yantis, 1992)
- Object files (Kahneman Treisman, 1984)
- Object-based attention
9Common assumptions
- Low level (1st order) motion system updates
higher-level representation - FINST
- Object file
- Virtual polygon
- Continuous computation in the present
10Overview
- MVT and attention
- Tracking across the gap
- Tracking trajectories
11MVT and attention
- Clearly a limited-capacity resource
- Attentional priority to tracked items (Sears
Pylyshyn) - Hypothesis MVT is mutually exclusive with other
attentional tasks
George Alvarez, Helga Arsenio, Jennifer DiMase,
Jeremy Wolfe
12MVT and attention
- Clearly a limited-capacity resource
- Attentional priority to tracked items (Sears
Pylyshyn) - Hypothesis MVT is mutually exclusive with visual
search
13MVT and attention
- Clearly a limited-capacity resource
- Attentional priority to tracked items (Sears
Pylyshyn) - Hypothesis MVT is mutually exclusive with visual
search - Method Attentional Operating Characteristic (AOC)
14AOC Theory
15General methods - normalization
- Single task 100
- Chance 0
- Dual task performance scaled to distance between
single task performance and chance
16General methods - staircases
- Up step (following error) 2 x down step
- Asymptote 66.7 accuracy
- Staircase runs until 20 reversals
- Asymptote computed on last 10 reversals
17General methods - tracking
- 10 disks
- 5 disks cued
- Speed 9/s
18AOC Theory
19AOC reality
- Tasks can interfere at multiple levels
- Interference can occur even when resource of
interest (here visual attention) is not shared - How independent are two attention-demanding
tasks which do not share visual attention
resources?
20Gold standard tracking vs. tone detection
21Gold standard method
- Tracking
- Duration 6 s
- Tone duration
- 10 600 Hz tones
- Onset t 1 s
- ITI 400 ms
- Distractor duration 200 ms
- Task target tone longer or shorter?
- Target duration staircased (?31 ms)
- Dual task priority varied
N 10
22Gold standard AOC
23Tracking search method
- Tracking
- Duration 5 s
- Search
- 2AFC E vs. N
- Distractors rest of alphabet
- Set size 5
- Duration staircased (mean 156 ms)
- Onset 2 s
N 9
24Tracking search method
25Tracking search AOC
26Tracking search AOC
27Does tracked status matter?
L
L
L
28method
- Tracking
- Duration 3 s
- Search
- 2AFC left- or right-pointing T
- Distractors rotated Ls
- Set size 5
- Duration staircased (mean 218 ms)
- Onset 1 s
N 9
29search inside tracked set
L
L
T
L
L
30search outside tracked set
L
L
L
T
L
L
31(No Transcript)
32inside vs. outside AOC
33Does spatial separation matter?
P
E
H
F
V
34method
- Tracking
- Duration 5 s
- Search
- 2AFC E vs. N
- Distractors rest of alphabet
- Set size 5
- Duration 200 ms
- Onset 2 s
N 9
35spatial separation AOC
36search v track summary
37MVT and search
- Clearly not mutually exclusive
- Not pure independence
- Close to gold standard
- MVT and search use independent resources?
38Two explanations
- Separate attention mechanisms
- Time sharing
39Predictions of time sharing hypothesis
- Should be able to leave tracking task for
significant periods with no loss of performance - Should be able to do something in that interval
40Track across the gap method
41Track across the gap method
- Track 4 of 8 disks
- Speed 6/s
- Blank interval onset 1, 2, or 3 s
- Trajectory variability 0, 15, 30, or 45
every 20 ms - Blank interval duration staircased (dv)
- N 11
42track across the gap asymptotes
43Predictions of time sharing hypothesis
- Should be able to leave tracking task for
significant periods with no loss of performance
(see also Yin Thornton, 1999) - confirmed - Should be able to do something (e.g. search) in
that interval
44search during gap method
- AOC method
- Tracking task same as before
- Search task in blank interval
- Target rotated T
- Distractors rotated Ls
- Set size 8
- 4AFC Report orientation of T
- Duration of search task staircased (326 ms)
45search during gap AOC
46(No Transcript)
47Predictions of time sharing hypothesis
- Should be able to leave tracking task for
significant periods of time with no loss of
performance (see also Yin Thornton, 1999) -
confirmed - Should be able to do something (e.g. search) in
that interval - confirmed
48Summary
- MVT and visual search can be performed
independently in the same trial - May support independent visual attention
mechanisms - May support time-sharing
49Summary
- Tracking across the gap data support time sharing
- Tracking across the gap data raise new questions
50What is the mechanism?
- Not a continuous computation in the present
- Not first order motion mechanisms
- Not apparent motion
Randall Birnkrant, Jennifer DiMase, Sarah
Klieger, Linda Tran, Jeremy Wolfe
51None of these theories fit
- FINSTs (Pylyshyn, 1989)
- Virtual polygons (Yantis, 1992)
- Object files (Kahneman Treisman, 1984)
52What is the mechanism?
- Some sort of amodal perception? (e.g. tracking
behind occluders, Scholl Pylyshyn, 1999) - but there are no occlusion cues!
53(No Transcript)
54(No Transcript)
55Scholl Pylyshyn, 1999
56Maybe the gap is just an impoverished occlusion
stimulus
- No occlusion/disocclusion cues
- Synchronous disappearance
57Predictions of impoverished occlusion hypothesis
- Occlusion cues will improve performance
- Asynchronous disappearance will improve
performance
58Method
- Track for 5 s
- Speed 12/s
- Track 4 of 10 disks
- Independent variables (blocked)
- Gap duration107 ms, 307 ms, 507 ms
- Occlusion cues absent, present
- Disappearances synchronous, asynchronous
- N 15
59synchronous disappearance
60synchronous disappearance occlusion
61Occlusion/Disocclusion
62asynchronous disappearance
63asynchronous disappearance occlusion
64comparing cue types
65Occlusion hypothesis fails
- Occlusion cues dont help
- Asynchronous disappearance doesnt help
66Method
- Track for 5 s
- Speed 12/s
- Synchronous condition only
- Independent variables (blocked)
- Gap duration107 ms, 307 ms, 507 ms
- Occlusion cues absent, present
- Track 4, 5, or 6 of 10 disks
- N 11
67comparing cue types
68Occlusion hypothesis fails
- Occlusion cues dont help
- Occlusion cues can actually harm performance
- Asynchronous disappearance doesnt help
69What is the mechanism?
- Not a continuous computation in the present
- Not first order motion mechanisms
- Not apparent motion
- Not amodal perception (occlusion)
70How do we reacquire targets?
- remember last location (backward)
- store trajectory (forward)
David Fencsik, Sarah Klieger, Jeremy Wolfe
71location-matching account
Memorized pre-gap target location.
Nearest to memorized location identified as
target.
First Post-Gap Frame
72trajectory-matching account
Memorized pre-gap target trajectory.
On target trajectory identified as target.
First Post-Gap Frame
73Shifting post-gap location
74shifting post-gap location predictions
75Shifting post-gap location methods
- track for 5 s
- speed 8/s
- track 5 of 10 disks
- gap duration 300 ms
- post-gap location condition blocked
- stimuli continue to move after gap
76shifting post-gap location
77Location vs. trajectory-matching
- support for location-matching
- see also Keane Pylyshyn 2003 2004
- but advantage for -1 is suspicious
78Location vs. trajectory-matching
79shift stop methods
- track for 4-6 s
- speed 9/s
- track 2 or 5 of 10 disks
- gap duration 300 ms
- post-gap location condition blocked
- stimuli stop after gap
80moving vs. static after gap
81moving vs. static after gap
822 vs. 5 targets
83Location vs. trajectory-matching
- support for location-matching
- However...
- conditions are blocked
- observers might see their task not as tracking
across the gap, but learning which condition
theyre in - might not tell us about normal target recovery
84Location vs. trajectory-matching
- can subjects use trajectory information?
- always have items move during gap
- vary whether trajectory information is available
or not
85moving condition
86static condition
87manipulate pre-gap information methods
- track for 4 s
- speed 9/s
- track 1 to 4 of 10 disks
- gap duration 300 ms
88manipulate pre-gap information
89manipulate pre-gap information
90Location vs. trajectory-matching
- observers can use trajectory information
- unlimited (or at least gt 4) capacity for
locations - smaller (1 or 2) capacity for trajectories
91Conclusions
- Flexible attention system allows rapid switching
between MVT and other attention-demanding tasks - Some representation allows recovery of tracked
targets after 300-400 ms gaps - This representation includes location and
trajectory information
92Speculation
- MVT reveals two mechanisms, rather than just one
- Frequently (but perhaps not continuously) updated
location store - Attention to trajectories