-
Notifications
You must be signed in to change notification settings - Fork 0
/
TODO
42 lines (41 loc) · 1.52 KB
/
TODO
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
- Reach target:
- N trials of each type
- 8 types of trial 2x2x2: (hrf)x(visible)x(dof)
(t/f)x(t/f)x(1/2)
- 2s TR w/ 'm' noise from cts/term
* Questions:
* what would minimum N be?
* what would maximum N be?
* can we have less trials of the easy one?
* ask stats guy: 5 trials of each?
try 20 trials, tweak to get:
(2+2+6+6+8+16+16+32)*2*20/60
# ToDo:
# code cleanup
# start from left side/bottom left-hand corner
# ability to pause
# indication to go back to start circle
- Cts/term:
- N trials of each type
- 4 types of trial: 2x2 (noise)x(TR)
(l/h)x(2/20)
- 1 DOF
- 2 min per trial -- might have to tweak target size
or noise level
* Questions:
* N at 7 (56min)
* relationship between noise variance and taking the mean
* case 1: generate noise at variance s
* case 2: generate 10 samples of noise at variance 10*s
and then take the mean
- Playback:
- N trials of each type
- 2 DOF
- 2 types of trial: traditional vs. replay
- 4 min per trial (tentative)
- 6 playback blocks per trial (tentative)
- ** try to make playback really really good
* Questions:
* which 'traditional' experiment type to choose? 2s TR and 'm' noise?
* how long to figure out 2 DOF target? Tweak trial length based on this.
* N at 8