Analysis logbook: cue-expectancy
1
url: your book url like https://bookdown.org/yihui/bookdown
1.1
Usage
1.2
Render book
1.3
Preview book
2
[ beh] expectation ~ cue
2.1
Pain
2.2
Vicarious
2.3
Cognitive
2.4
Individual difference analysis
3
[ beh ] outcome ~ cue
3.1
Pain
3.2
Vicarious
3.3
Cognitive
3.4
Individual differences analysis: random cue effects
3.5
Individual differences analysis 2: random intercept + random slopes of cue effect
4
[ beh ] outcome ~ stimulus_intensity
4.1
Pain
4.2
Vicarious
4.3
Cognitive
4.4
for loop
4.5
Lineplot
4.6
individual differences in outcome rating cue effect
5
outcome_rating ~ cue * stim
5.1
What is the purpose of this notebook?
5.2
model 03 iv-cuecontrast dv-actual
5.2.1
model 03 3-2. individual difference
5.2.2
model 04 iv-cue-stim dv-actual
5.2.3
model 04 4-2 individual differences in cue effects
5.2.4
model 04 4-3 scatter plot
5.2.5
model 04 4-4 lineplot
6
expect-actual ~ cue * trial
6.1
Overview
6.2
plot 1 - one run, average across participants
6.3
plot 2 - average across participant, but spread all 6 runs in one x axis
6.4
Do current expectation ratings predict outcome ratings?
6.5
Additional analysis
7
RT ~ cue
7.0.1
parameters
7.0.2
1) plot RT data
7.0.3
plot RT distribution per participant
7.0.4
exclude participants with RT of 5 seconds
7.1
model 1:
7.2
model 1-1
7.3
model 1-2:
7.4
model 2:
8
RT ~ cue * stim
8.1
Overview model 05 iv-cue dv-RT summary
8.2
Prepare data and preprocess
8.3
model 1:
8.4
model 1-1
8.5
model 1-2:
8.6
model 2: Log transformation
8.7
Conclusion across model 1 and 2
9
cognitive RT tradeoff ~ cue * stim (withinsubject)
9.1
Overview
9.2
Why use multilevel models?
9.3
Terminology
9.4
Model versions
9.5
Method 1 one-sample t
9.6
Method 1-1 aov
9.7
Method 1-2 aov contrast-coding
9.8
Method 1 effectsize
9.9
Method 2 matlab
9.10
Method 3 multilevel modeling
Conclusion: Method 1 vs Method 3
9.11
References
9.12
Other links
10
outcome_rating ~ session (“behavioral ICC”)
10.1
Functions
10.2
TODO:
11
[beh] N-1 outcome rating ~ N expectation rating
11.1
expectation_rating ~ N-1_outcome_rating
11.2
Current expectation_rating ~ N-1_outcomerating * cue
11.3
Let’s demean the ratings.
11.4
DEMEAN AND THEN DISCRETIZE
12
(N-2) shifted outcome ratings ~ (N) expectation ratings; Jayazeri (2018)
12.1
Do previous outcome ratings predict current expectation ratings?
12.2
Do these models differ as a function of cue?
12.3
Demean and discretize
13
outcome ~ expect Jayazeri (2018)
13.1
Overview
13.2
Do expectation ratings predict current outcome ratings? Does this differ as a function of cue?
13.3
task-pain, HLM modeling
13.4
Fig. Expectation ratings predict outcome ratings
13.5
binned expectation ratings per task
13.5.1
Pain: binned expectation ratings
13.5.2
Vicarious: binned expectation ratings
13.5.3
Cognitive: binned expectation ratings
13.6
not splitting into cue groups
14
[ beh ] outcome ~ cue * stim * expectrating * n-1outcomerating
14.1
Original motivation:
14.2
Pain
14.2.1
pain plot parameters
14.2.2
loess
14.3
Vicarious
14.4
Cognitive
14.4.1
cognitive parameters
15
[ beh ] outcome_demean ~ cue * stim * expectrating * n-1outcomerating
15.1
linear model
15.2
Q. Are those overestimating for high cues also underestimators for low cues?
15.3
pain run, collapsed across stimulus intensity
15.4
vicarious
15.5
cognitive
15.6
across tasks (PVC), is the slope for (highvslow cue) the same?Tor question
16
[ beh ] outcome_demean_per_run ~ cue * stim * expectrating * n-1outcomerating
16.1
Linear model with three factors: cue X stim X expectation rating
16.2
Pain run, collapsed across stimulus intensity
16.3
vicarious
16.4
cognitive
16.5
across tasks (PVC), is the slope for (highvslow cue) the same?Tor question
17
[beh] Mediation outcome ~ cue * stim * expectrating * n-1outcomerating
17.1
mediation
17.2
mediation 2
17.3
mediation 3: Test same model using mediation() from MBESS
17.4
mediation 4: Test library mediation
18
NPS_contrast_notscaled ~ cue * stim
18.1
Overview
18.2
regressors and contrasts
18.3
main effect: stim-linear high > low
18.4
main_effect: stim-quadratic med > high&low
18.5
interaction: cue X stim-linear
18.6
interaction: cue X stim-quadratic
19
[fMRI] NPS_contrast ~ cue * stim
19.1
Overview
19.2
regressors and contrasts
19.3
main effect: stim-linear high > low
19.4
main_effect: stim-quadratic med > high&low
19.5
interaction: cue X stim-linear
19.6
interaction: cue X stim-quadratic
20
[fMRI] nps_contrast ~ cue * stim (“error, contrast not scaled”)
20.1
Overview
20.2
contrast plot
21
[fMRI] nps_contrast ~ cue * stim
21.1
Overview
21.2
‘P_simple_stimlin_high_gt_low’, ‘V_simple_stimlin_high_gt_low’, ‘C_simple_stimlin_high_gt_low’,…
21.3
‘P_simple_stimquad_med_gt_other’, ‘V_simple_stimquad_med_gt_other’, ‘C_simple_stimquad_med_gt_other’,…
21.4
For loop for all the pvc dummy codes
22
nps_dummy ~ stim
22.1
TODO
22.2
regressors and contrasts
22.3
Functions
22.4
Pain
22.5
Vicarious
22.6
Cognitive
23
[fMRI] biomarker NPS ~ cue * stim (2022)
23.1
load libraries
23.1.1
NPS cue effect
23.2
NPS stim effect
23.3
VPS
23.4
VPS cue effect
23.5
VPS stim effect
24
[fMRI] NPSdummy ~ stim * task (contrast-notscaled-error)
25
NPSdummy ~ stim * task (contrast-scaled)
25.1
Raincloud plots
25.2
Line plots
26
[fMRI] NPS ~ singletrial
26.1
NPS ~ paintask: 2 cue x 3 stimulus_intensity
26.1.1
Linear model results (NPS ~ paintask: 2 cue x 3 stimulus_intensity)
26.1.2
2 cue * 3 stimulus_intensity * expectation_rating
26.2
NPS_ses01 ~ SES * CUE * STIM
26.2.1
Here are the stats models: NPS~session * cue * stimulus_intensity
26.3
26.4
OUTCOME ~ NPS
26.4.1
outcome_rating * cue
26.4.2
outcome_ratings * stimulus_intensity * cue
26.4.3
demeaned outcome rating * cue
26.4.4
demeaned_outcome_ratings * stimulus_intensity * cue
26.4.5
Is this statistically significant?
26.5
NPS ~ expectation_rating
26.5.1
NPS ~ expect * cue
26.5.2
NPS ~ expect * cue * stim
26.5.3
NPS ~ demeaned_expect * cue
26.5.4
NPS ~ demeaned_expect * cue * stim
26.5.5
ACCURATE Is this statistically significant?
27
https://stats.stackexchange.com/questions/586748/calculating-trends-with-emtrends-for-three-way-interaction-model-results-in-sa
28
emtrends(model.npsexpectdemean, var = ‘EXPECT_demean’, lmer.df = “asymptotic”)
29
[fMRI] NPSses01ses03 ~ singletrial
29.1
NPS ~ paintask: 2 cue x 3 stimulus_intensity
29.1.1
Linear model results (NPS ~ paintask: 2 cue x 3 stimulus_intensity)
29.1.2
2 cue * 3 stimulus_intensity * expectation_rating
29.2
NPS_ses01 ~ SES * CUE * STIM
29.2.1
Here are the stats models: NPS~session * cue * stimulus_intensity
29.3
OUTCOME ~ NPS
29.3.1
outcome_rating * cue
29.3.2
outcome_ratings * stimulus_intensity * cue
29.3.3
demeaned outcome rating * cue
29.3.4
demeaned_outcome_ratings * stimulus_intensity * cue
29.3.5
Is this statistically significant?
29.4
NPS ~ expectation_rating
29.4.1
NPS ~ expect * cue
29.4.2
NPS ~ expect * cue * stim
29.4.3
NPS ~ demeaned_expect * cue
29.4.4
NPS ~ demeaned_expect * cue * stim
29.4.5
ACCURATE Is this statistically significant?
30
[fMRI] NPSses01ses03 ~ singletrial
30.1
NPS ~ paintask: 2 cue x 3 stimulus_intensity
30.1.1
Linear model results (NPS ~ paintask: 2 cue x 3 stimulus_intensity)
30.1.2
2 cue * 3 stimulus_intensity * expectation_rating
30.2
NPS_ses01 ~ SES * CUE * STIM
30.2.1
Here are the stats models: NPS~session * cue * stimulus_intensity
30.3
OUTCOME ~ NPS
30.3.1
outcome_rating * cue
30.3.2
outcome_ratings * stimulus_intensity * cue
30.3.3
demeaned outcome rating * cue
30.3.4
demeaned_outcome_ratings * stimulus_intensity * cue
30.3.5
Is this statistically significant?
30.4
NPS ~ expectation_rating
30.4.1
NPS ~ expect * cue
30.4.2
NPS ~ expect * cue * stim
30.4.3
NPS ~ demeaned_expect * cue
30.4.4
NPS ~ demeaned_expect * cue * stim
30.4.5
NPS_demean ~ demeaned_expect * cue
30.4.6
NPS ~ demeaned_expect * cue * stim
30.4.7
ACCURATE Is this statistically significant?
31
[fMRI] NPSses04 ~ singletrial
31.1
NPS ~ paintask: 2 cue x 3 stimulus_intensity
31.1.1
Linear model results (NPS ~ paintask: 2 cue x 3 stimulus_intensity)
31.1.2
2 cue * 3 stimulus_intensity * expectation_rating
31.2
NPS_ses01 ~ SES * CUE * STIM
31.2.1
Here are the stats models: NPS~session * cue * stimulus_intensity
31.3
OUTCOME ~ NPS
31.3.1
outcome_rating * cue
31.3.2
outcome_ratings * stimulus_intensity * cue
31.3.3
demeaned outcome rating * cue
31.3.4
demeaned_outcome_ratings * stimulus_intensity * cue
31.3.5
Is this statistically significant?
31.4
NPS ~ expectation_rating
31.4.1
NPS ~ expect * cue
31.4.2
NPS ~ expect * cue * stim
31.4.3
NPS ~ demeaned_expect * cue
31.4.4
NPS ~ demeaned_expect * cue * stim
31.4.5
NPS_demean ~ demeaned_expect * cue
31.4.6
NPS ~ demeaned_expect * cue * stim
31.4.7
Is this statistically significant?
32
[fMRI] NPS ~ singletrial
32.1
NPS ~ 3 task * 3 stimulus_intensity
32.2
NPS ~ paintask: 2 cue x 3 stimulus_intensity
32.2.1
Linear model results (NPS ~ paintask: 2 cue x 3 stimulus_intensity)
32.2.2
2 cue * 3 stimulus_intensity * expectation_rating
32.3
NPS ~ SES * CUE * STIM
32.3.1
Here are the stats models: NPS~session * cue * stimulus_intensity
32.4
OUTCOME ~ NPS
32.4.1
outcome_rating * cue
32.4.2
outcome_ratings * stimulus_intensity * cue
32.4.3
demeaned outcome rating * cue
32.4.4
demeaned_outcome_ratings * stimulus_intensity * cue
32.4.5
Is this statistically significant?
32.5
NPS ~ expectation_rating
32.5.1
demeaned expect rating * cue
32.5.2
Is this statistically significant?
33
fMRI Pain signature ~ single trial
33.1
PVC all task comparison
33.2
Vicarious only Stim x cue interaction
33.2.1
2x3 stimulus intensity * cue
33.2.2
Linear model
33.2.3
VPS stimulus intensity Cohen’s d = 0.2131521
33.2.4
VPS stimulus & cue effect size: stim_d = 0.217, cue_d = 0.013
33.2.5
Lineplots
33.2.6
Linear model with Stim x Cue x Expectation rating
33.3
Vicarious only: Outcome ratings & VPS
33.3.1
outcome ratings * cue
33.3.2
outcome ratings * stim * cue
33.4
Vicarious only: Expectation ratings & VPS
34
Cognitive signature ~ single trial
34.1
PVC all task comparison
34.2
Cognitive only Stim x cue interaction
34.2.1
2x3 stimulus intensity * cue
34.2.2
Linear model
34.2.3
Cog stimulus intensity Cohen’s d = 0.72
34.2.4
Cognitive stimulus & cue effect size: stim_d = 0.73, cue_d = 0.069
34.2.5
Lineplots
34.2.6
Linear model with Stim x Cue x Expectation rating
34.2.7
Session 1: 2x3 stimulus intensity * cue
34.2.8
Session 3: 2x3 stimulus intensity * cue
34.2.9
Session 4: 2x3 stimulus intensity * cue
34.3
Cognitive only: Outcome ratings & Kragel 2018
34.3.1
outcome ratings * cue
34.3.2
outcome ratings * stim * cue
34.4
Cognitive only: Expectation ratings & NPS
35
single trial correlation between cue and stim ~ cue x stim
35.1
Stack data
35.2
plot correlation (one-sample-t)
35.3
Lineplot
36
signature effect size ~ single trial
36.1
effeect size
36.2
contrastt (stim intensity)
36.3
layer in metadata
37
vif
Published with bookdown
A Minimal Book Example
Chapter 27
https://stats.stackexchange.com/questions/586748/calculating-trends-with-emtrends-for-three-way-interaction-model-results-in-sa