RMarkdown_For_Brian_Meeting
Noah R. Wolkowicz
2022-10-19
<style type="text/css">
div.main-container {
max-width: 100%;
margin-left: auto;
margin-right: auto;
}
</style>1. JP_Data
-Note: Large chunks of code have been hidden to improve readibility.
2. Data Read-In
-Code hidden
3. Data Pre-processing
Creating Trial Number Column
-Code Hidden
Probe accuracy, incorrect trials, improbable reaction times
-The code below when run illustrates that there are no administrations -Code Hidden #With <50% accuracy. Nothing excluded here.
-The code below indicates the overall # of trials and the counts/percents for 1) incorrect trials to be excluded, and 2) improbable reaction time trials (i.e., <200ms OR >1500ms) to be excluded. The code immediately following then excludes this data.
-Filtering out improbable reaction times
-Maximum & Minimum RT after exclusions
-Code for Median Absolute Deviation and SD calculations for Reaction Times hidden
-MAD & SD Calculation Results
-Excluding Folks based on SD for Reaction Times
3. Descriptive Statistics
-Code Hidden
4. Attentional Bias (AB) Calculations
-Code for AB calculations hidden
-Attentional Bias Metrics Per Subject, Condition, Session, & Cue -Code hidden
5. ANOVA’s
-Changing session variable to represent just pre-post -Code Hidden
-Something to note: There are several subjects who appear to have no data for at least Condition/1 time-point:
Folks Missing 1 pre/post data point: #Condition A: 15 #Condition B: 02 #Condition C: 14
Folks Missing both pre/post data points #Condition A: 03, 11, 13, 14, 18, 22, 28 #Condition B: 03, 13, 17 #Condition C: 17,18,23
-R drops these folks from ANOVA analysis below. Listwise deletion results in a small subsample, with between 12-15/28 getting dropped, depending on condition.
Adding grouping by condition variable
-Code Hidden
-Changing Data into Long Format -Code Hidden
6. Mixed-Effects ANOVAs examining Attentional Biases by Session*Condition
6a. Mean Bias
-DV: Opioid Mean Bias
df_long_opioid_mean_bias <- df_long %>% filter(Cue == "opioid" & Bias_Type == "mean_bias")
afex::aov_car(formula = RT ~ Session*Condition + Error(Subject/Session*Condition), data = df_long_opioid_mean_bias,
type = 3, na.rm = T)## Warning: Missing values for following ID(s):
## 56002, 56003, 56011, 56013, 56014, 56015, 56017, 56018, 56022, 56023, 56028
## Removing those cases from the analysis.## Anova Table (Type 3 tests)
##
## Response: RT
## Effect df MSE F ges p.value
## 1 Session 1, 15 1730.29 0.70 .017 .417
## 2 Condition 1.87, 28.00 674.69 0.72 .013 .487
## 3 Session:Condition 1.71, 25.58 599.62 1.92 .027 .171
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG-DV: Pain Mean Bias
df_long_pain_mean_bias <- df_long %>% filter(Cue == "pain" & Bias_Type == "mean_bias")
afex::aov_car(formula = RT ~ Session*Condition + Error(Subject/Session*Condition), data = df_long_pain_mean_bias,
type = 3, na.rm = T)## Warning: Missing values for following ID(s):
## 56002, 56003, 56011, 56013, 56014, 56015, 56017, 56018, 56022, 56023, 56028
## Removing those cases from the analysis.## Anova Table (Type 3 tests)
##
## Response: RT
## Effect df MSE F ges p.value
## 1 Session 1, 15 899.92 0.37 .005 .552
## 2 Condition 1.27, 19.02 1477.83 0.28 .009 .657
## 3 Session:Condition 1.80, 26.94 418.87 0.17 .002 .825
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG6b. Mean Away Bias
-DV: Opioid Mean Away Bias
df_long_opioid_mean_away <- df_long %>% filter(Cue == "opioid" & Bias_Type == "mean_away")
afex::aov_car(formula = RT ~ Session*Condition + Error(Subject/Session*Condition), data = df_long_opioid_mean_away,
type = 3, na.rm = T)## Warning: Missing values for following ID(s):
## 56002, 56003, 56011, 56013, 56014, 56015, 56017, 56018, 56022, 56023, 56028
## Removing those cases from the analysis.## Anova Table (Type 3 tests)
##
## Response: RT
## Effect df MSE F ges p.value
## 1 Session 1, 15 1084.77 2.34 .010 .147
## 2 Condition 1.48, 22.25 1281.37 0.86 .007 .406
## 3 Session:Condition 1.98, 29.77 373.76 0.42 .001 .660
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG-DV: Pain Mean Away Bias
df_long_pain_mean_away <- df_long %>% filter(Cue == "pain" & Bias_Type == "mean_away")
afex::aov_car(formula = RT ~ Session*Condition + Error(Subject/Session*Condition), data = df_long_pain_mean_away,
type = 3, na.rm = T)## Warning: Missing values for following ID(s):
## 56002, 56003, 56011, 56013, 56014, 56015, 56017, 56018, 56022, 56023, 56028
## Removing those cases from the analysis.## Anova Table (Type 3 tests)
##
## Response: RT
## Effect df MSE F ges p.value
## 1 Session 1, 15 697.86 1.10 .003 .310
## 2 Condition 1.66, 24.90 1342.32 0.07 <.001 .899
## 3 Session:Condition 1.51, 22.72 809.59 0.07 <.001 .881
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG6c. Mean Toward Bias
-DV: Opioid Mean Toward Bias
df_long_opioid_mean_toward <- df_long %>% filter(Cue == "opioid" & Bias_Type == "mean_toward")
afex::aov_car(formula = RT ~ Session*Condition + Error(Subject/Session*Condition), data = df_long_opioid_mean_toward,
type = 3, na.rm = T)## Warning: Missing values for following ID(s):
## 56002, 56003, 56011, 56013, 56014, 56015, 56017, 56018, 56022, 56023, 56028
## Removing those cases from the analysis.## Anova Table (Type 3 tests)
##
## Response: RT
## Effect df MSE F ges p.value
## 1 Session 1, 15 331.77 1.33 .002 .266
## 2 Condition 1.83, 27.44 1097.91 0.65 .007 .515
## 3 Session:Condition 1.20, 18.06 1727.81 0.21 .002 .696
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG-DV: Pain Mean Toward Bias
df_long_pain_mean_toward <- df_long %>% filter(Cue == "pain" & Bias_Type == "mean_toward")
afex::aov_car(formula = RT ~ Session*Condition + Error(Subject/Session*Condition), data = df_long_pain_mean_toward,
type = 3, na.rm = T)## Warning: Missing values for following ID(s):
## 56002, 56003, 56011, 56013, 56014, 56015, 56017, 56018, 56022, 56023, 56028
## Removing those cases from the analysis.## Anova Table (Type 3 tests)
##
## Response: RT
## Effect df MSE F ges p.value
## 1 Session 1, 15 1623.54 0.22 .002 .643
## 2 Condition 1.91, 28.68 845.07 1.05 .008 .359
## 3 Session:Condition 1.69, 25.36 433.06 0.60 .002 .527
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG6d. Peak Away Bias
-DV: Opioid Peak Away Bias
df_long_opioid_peak_away <- df_long %>% filter(Cue == "opioid" & Bias_Type == "peak_away")
afex::aov_car(formula = RT ~ Session*Condition + Error(Subject/Session*Condition), data = df_long_opioid_peak_away,
type = 3, na.rm = T)## Warning: Missing values for following ID(s):
## 56002, 56003, 56011, 56013, 56014, 56015, 56017, 56018, 56022, 56023, 56028
## Removing those cases from the analysis.## Anova Table (Type 3 tests)
##
## Response: RT
## Effect df MSE F ges p.value
## 1 Session 1, 15 5111.51 1.17 .002 .296
## 2 Condition 1.64, 24.59 16820.23 0.80 .007 .438
## 3 Session:Condition 1.72, 25.80 7437.94 0.00 <.001 .993
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG-DV: Pain Peak Away Bias
df_long_pain_peak_away <- df_long %>% filter(Cue == "pain" & Bias_Type == "peak_away")
afex::aov_car(formula = RT ~ Session*Condition + Error(Subject/Session*Condition), data = df_long_pain_peak_away,
type = 3, na.rm = T)## Warning: Missing values for following ID(s):
## 56002, 56003, 56011, 56013, 56014, 56015, 56017, 56018, 56022, 56023, 56028
## Removing those cases from the analysis.## Anova Table (Type 3 tests)
##
## Response: RT
## Effect df MSE F ges p.value
## 1 Session 1, 15 10744.89 1.46 .006 .246
## 2 Condition 1.98, 29.73 6103.29 0.37 .002 .693
## 3 Session:Condition 1.55, 23.30 9044.77 0.71 .004 .468
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG6e. Peak Toward Bias
-DV: Opioid Peak Toward Bias
df_long_opioid_peak_toward <- df_long %>% filter(Cue == "opioid" & Bias_Type == "peak_toward")
afex::aov_car(formula = RT ~ Session*Condition + Error(Subject/Session*Condition), data = df_long_opioid_peak_toward,
type = 3, na.rm = T)## Warning: Missing values for following ID(s):
## 56002, 56003, 56011, 56013, 56014, 56015, 56017, 56018, 56022, 56023, 56028
## Removing those cases from the analysis.## Anova Table (Type 3 tests)
##
## Response: RT
## Effect df MSE F ges p.value
## 1 Session 1, 15 7247.98 3.83 + .012 .069
## 2 Condition 1.62, 24.32 11543.84 0.13 <.001 .840
## 3 Session:Condition 1.20, 17.99 19441.12 0.25 .002 .668
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG-DV: Pain Peak Toward Bias
df_long_pain_peak_toward <- df_long %>% filter(Cue == "pain" & Bias_Type == "peak_toward")
afex::aov_car(formula = RT ~ Session*Condition + Error(Subject/Session*Condition), data = df_long_pain_peak_toward,
type = 3, na.rm = T)## Warning: Missing values for following ID(s):
## 56002, 56003, 56011, 56013, 56014, 56015, 56017, 56018, 56022, 56023, 56028
## Removing those cases from the analysis.## Anova Table (Type 3 tests)
##
## Response: RT
## Effect df MSE F ges p.value
## 1 Session 1, 15 12315.63 0.04 <.001 .853
## 2 Condition 1.32, 19.84 14603.39 0.45 .004 .562
## 3 Session:Condition 1.35, 20.18 5994.07 0.31 .001 .649
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG6f. Variability Bias
-DV: Opioid Variability Bias
df_long_opioid_variability <- df_long %>% filter(Cue == "opioid" & Bias_Type == "variability")
afex::aov_car(formula = RT ~ Session*Condition + Error(Subject/Session*Condition), data = df_long_opioid_variability,
type = 3, na.rm = T)## Warning: Missing values for following ID(s):
## 56002, 56003, 56011, 56013, 56014, 56015, 56017, 56018, 56022, 56023, 56028
## Removing those cases from the analysis.## Anova Table (Type 3 tests)
##
## Response: RT
## Effect df MSE F ges p.value
## 1 Session 1, 15 305.22 1.86 .004 .192
## 2 Condition 1.86, 27.85 623.94 0.82 .006 .441
## 3 Session:Condition 1.89, 28.32 283.68 0.62 .002 .534
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG-DV: Pain Variability Bias
df_long_pain_variability <- df_long %>% filter(Cue == "pain" & Bias_Type == "variability")
afex::aov_car(formula = RT ~ Session*Condition + Error(Subject/Session*Condition), data = df_long_pain_variability,
type = 3, na.rm = T)## Warning: Missing values for following ID(s):
## 56002, 56003, 56011, 56013, 56014, 56015, 56017, 56018, 56022, 56023, 56028
## Removing those cases from the analysis.## Anova Table (Type 3 tests)
##
## Response: RT
## Effect df MSE F ges p.value
## 1 Session 1, 15 646.84 0.46 .002 .510
## 2 Condition 1.96, 29.36 390.98 0.80 .004 .457
## 3 Session:Condition 1.85, 27.72 370.63 0.04 <.001 .950
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG7. Muiltilevel Models examining Attentional Bias regressed on session*condition (fixed effects). Random intercepts per subject, random slopes across sessions (pre-post)
7a. Mean Bias
-DV: Opioid Mean Bias
re_O_MB <- lmer(RT ~ (1|Subject), data = df_long_opioid_mean_bias)
performance::icc(re_O_MB)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.088
## Conditional ICC: 0.088MLM_O_MB <- lmer(RT ~ Session*Condition + (1 + Session|Subject), data = df_long_opioid_mean_bias)
#The code below then obtains mltiple Degree-of-Freedom F-Tests for the above-specified model
anova(MLM_O_MB, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 204.29 204.29 1 25.962 0.3180 0.5777
## Condition 1456.11 728.06 2 84.311 1.1332 0.3268
## Session:Condition 1513.27 756.63 2 83.174 1.1777 0.3131#Model Parameters
summary(MLM_O_MB)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_opioid_mean_bias
##
## REML criterion at convergence: 1244.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.21980 -0.47485 -0.01903 0.39442 3.14079
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 348.7 18.67
## SessionPost 613.7 24.77 -0.85
## Residual 642.5 25.35
## Number of obs: 134, groups: Subject, 27
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -3.3981 7.0389 74.7171 -0.483 0.631
## SessionPost 8.6312 9.6716 75.8654 0.892 0.375
## ConditionB 9.1184 8.0190 85.2335 1.137 0.259
## ConditionC -5.5859 8.0191 85.2271 -0.697 0.488
## SessionPost:ConditionB -14.4242 11.1918 83.9354 -1.289 0.201
## SessionPost:ConditionC -0.3397 11.2622 84.4068 -0.030 0.976
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.754
## ConditionB -0.649 0.475
## ConditionC -0.649 0.475 0.573
## SssnPst:CnB 0.467 -0.650 -0.719 -0.413
## SssnPst:CnC 0.464 -0.647 -0.410 -0.715 0.559#Means/SDs
df_long_opioid_mean_bias %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT))## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A -4.26 31.4
## 2 Pre B 4.98 26.2
## 3 Pre C -10.7 33.1
## 4 Post A 5.55 34.4
## 5 Post B 0.433 23.7
## 6 Post C -0.937 31.5#Actual Means
AM <- ggplot(df_long_opioid_mean_bias, aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot", position = position_dodge(width = 0.5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_opioid_mean_bias, aes(x=Session,y=predict(MLM_O_MB), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot", position = position_dodge(width = 0.5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
-DV: Pain Mean Bias (NON-CONVERGENCE)
re_P_MB <- lmer(RT ~ (1|Subject), data = df_long_pain_mean_bias)## boundary (singular) fit: see help('isSingular')performance::icc(re_P_MB) #Error suggesting singularity## Warning: Can't compute random effect variances. Some variance components equal
## zero. Your model may suffer from singularity (see '?lme4::isSingular'
## and '?performance::check_singularity').
## Solution: Respecify random structure! You may also decrease the
## 'tolerance' level to enforce the calculation of random effect variances.## [1] NAMLM_P_MB <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_mean_bias,
control = lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e5))) ## boundary (singular) fit: see help('isSingular')#^Control liner above added to assist with convergence after receiving the following error:
#Warning: Model failed to converge with 1 negative eigenvalue: -1.1e+01
#Afterwards, the error "boundary (singular) fit: see help('isSingular')" suggests our random effects here are very small or close to 0
isSingular(MLM_P_MB) #Verifies at least 1 random effect is small or close to 0## [1] TRUE#Multiple Degree-of-Freedom F-Tests
anova(MLM_P_MB, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 205.71 205.71 1 22.746 0.3232 0.5753
## Condition 925.42 462.71 2 100.496 0.7270 0.4859
## Session:Condition 993.78 496.89 2 106.258 0.7807 0.4607#Means/SDs
df_long_pain_mean_bias %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition, Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A -4.69 22.8
## 2 Post A -7.40 36.3
## 3 Pre B -6.71 20.8
## 4 Post B 3.85 32.3
## 5 Pre C 0.104 26.5
## 6 Post C 1.59 21.6#Parameter Estimates
summary(MLM_P_MB)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_mean_bias
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 1224.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.4227 -0.4652 0.0981 0.4275 3.4955
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 34.55 5.878
## SessionPost 384.34 19.605 -1.00
## Residual 636.45 25.228
## Number of obs: 134, groups: Subject, 27
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -5.005 5.926 91.776 -0.845 0.401
## SessionPost -1.634 9.103 83.223 -0.180 0.858
## ConditionB -1.979 7.775 99.414 -0.255 0.800
## ConditionC 5.396 7.776 99.474 0.694 0.489
## SessionPost:ConditionB 12.585 11.035 106.703 1.140 0.257
## SessionPost:ConditionC 2.309 11.114 108.016 0.208 0.836
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.719
## ConditionB -0.734 0.487
## ConditionC -0.734 0.487 0.560
## SssnPst:CnB 0.526 -0.678 -0.713 -0.402
## SssnPst:CnC 0.523 -0.675 -0.399 -0.709 0.555
## optimizer (bobyqa) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')#Actual Means
AM <- ggplot(df_long_pain_mean_bias, aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot", position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_pain_mean_bias, aes(x=Session,y=predict(MLM_P_MB), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot", position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM, PM)
7b. Mean Away Bias
-DV: Opioid Mean Away Bias
re_O_MA <- lmer(RT ~ (1|Subject), data = df_long_opioid_mean_away)
performance::icc(re_O_MA)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.635
## Conditional ICC: 0.635MLM_O_MA <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_mean_away)
#Multiple Degree-of-Freedom F-Tests
anova(MLM_O_MA, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 378.16 378.16 1 19.678 0.4061 0.5313
## Condition 2073.29 1036.65 2 73.367 1.1133 0.3340
## Session:Condition 1003.54 501.77 2 78.812 0.5389 0.5855#Means/SDs
df_long_opioid_mean_away %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition, Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 106. 57.7
## 2 Post A 104. 63.5
## 3 Pre B 109. 41.6
## 4 Post B 115. 57.5
## 5 Pre C 114. 53.9
## 6 Post C 115. 54.2#Parameter Estimates
summary(MLM_O_MA)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_opioid_mean_away
##
## REML criterion at convergence: 1331.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.35516 -0.51551 -0.03073 0.32946 2.74247
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 1662.8 40.78
## SessionPost 560.4 23.67 0.21
## Residual 931.1 30.51
## Number of obs: 134, groups: Subject, 27
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 113.691 10.727 52.602 10.599 1.16e-14 ***
## SessionPost -2.904 11.013 76.200 -0.264 0.793
## ConditionB -9.120 9.685 78.400 -0.942 0.349
## ConditionC 2.920 9.688 78.215 0.301 0.764
## SessionPost:ConditionB 13.874 13.386 80.066 1.036 0.303
## SessionPost:ConditionC 8.388 13.462 80.923 0.623 0.535
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.368
## ConditionB -0.515 0.479
## ConditionC -0.515 0.479 0.574
## SssnPst:CnB 0.358 -0.678 -0.702 -0.398
## SssnPst:CnC 0.356 -0.675 -0.395 -0.695 0.553#Actual Means
AM <- ggplot(df_long_opioid_mean_away, aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_opioid_mean_away, aes(x=Session,y=predict(MLM_O_MA), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM, PM)
-DV: Pain Mean Away Bias (NON-CONVERGENCE)
re_P_MA <- lmer(RT ~ (1|Subject), data = df_long_pain_mean_away)
performance::icc(re_P_MA)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.745
## Conditional ICC: 0.745MLM_P_MA <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_mean_away)## boundary (singular) fit: see help('isSingular')isSingular(MLM_P_MA)## [1] TRUE#Multiple Degree-of-Freedom F-Tests
anova(MLM_P_MA, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 1036.10 1036.10 1 53.137 1.1259 0.2935
## Condition 2778.59 1389.30 2 103.619 1.5097 0.2258
## Session:Condition 8.41 4.21 2 101.765 0.0046 0.9954#Means/SDs
df_long_pain_mean_away %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition, Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 110. 60.3
## 2 Post A 114. 69.2
## 3 Pre B 112. 47.9
## 4 Post B 111. 48.7
## 5 Pre C 112. 62.3
## 6 Post C 115. 70.9#Parameter Estimates
summary(MLM_P_MA)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_mean_away
##
## REML criterion at convergence: 1327.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.6290 -0.4581 -0.1128 0.3960 2.9067
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 2370.6 48.69
## SessionPost 123.1 11.10 1.00
## Residual 920.3 30.34
## Number of obs: 134, groups: Subject, 27
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 117.5566 11.8121 45.3740 9.952 5.51e-13 ***
## SessionPost 5.5747 9.9752 99.1065 0.559 0.578
## ConditionB -10.8324 9.5488 102.4076 -1.134 0.259
## ConditionC -2.3703 9.5563 102.4536 -0.248 0.805
## SessionPost:ConditionB 0.2646 13.1255 101.7211 0.020 0.984
## SessionPost:ConditionC 1.1766 13.1818 101.7196 0.089 0.929
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.239
## ConditionB -0.458 0.506
## ConditionC -0.458 0.506 0.570
## SssnPst:CnB 0.314 -0.724 -0.696 -0.389
## SssnPst:CnC 0.312 -0.721 -0.387 -0.689 0.547
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')#Actual Means
AM <- ggplot(df_long_pain_mean_away, aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_pain_mean_away, aes(x=Session,y=predict(MLM_P_MA), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
7c. Mean Toward Bias
-DV: Opioid Mean Toward Bias (NON-CONVERGENCE)
re_O_MT <- lmer(RT ~ (1|Subject), data = df_long_opioid_mean_toward)
performance::icc(re_O_MT)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.684
## Conditional ICC: 0.684MLM_O_MT <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_mean_toward)## boundary (singular) fit: see help('isSingular')isSingular(MLM_O_MT)## [1] TRUE#Multiple Degree-of-Freedom F-Tests
anova(MLM_O_MT, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 46.89 46.89 1 100.02 0.0567 0.8123
## Condition 1626.97 813.48 2 104.62 0.9838 0.3773
## Session:Condition 459.86 229.93 2 102.02 0.2781 0.7578#Means/SDs
df_long_opioid_mean_toward %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition, Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 101. 52.4
## 2 Post A 111. 51.6
## 3 Pre B 113. 40.0
## 4 Post B 107. 37.8
## 5 Pre C 115. 63.1
## 6 Post C 110. 52.9#Parameter Estimates
summary(MLM_O_MT)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_opioid_mean_toward
##
## REML criterion at convergence: 1304.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.4295 -0.4645 -0.0749 0.3621 4.2082
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 1849.12 43.00
## SessionPost 2.34 1.53 -1.00
## Residual 826.85 28.75
## Number of obs: 134, groups: Subject, 27
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 108.3702 10.7536 47.9798 10.078 1.97e-13 ***
## SessionPost 6.5845 9.2397 101.8382 0.713 0.478
## ConditionB 0.6354 9.1018 103.5472 0.070 0.944
## ConditionC 9.7676 9.1089 103.6244 1.072 0.286
## SessionPost:ConditionB -7.3799 12.4482 102.0260 -0.593 0.555
## SessionPost:ConditionC -8.7908 12.4953 101.9863 -0.704 0.483
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.468
## ConditionB -0.481 0.525
## ConditionC -0.481 0.524 0.573
## SssnPst:CnB 0.333 -0.742 -0.701 -0.394
## SssnPst:CnC 0.331 -0.739 -0.392 -0.694 0.549
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')#Actual Means
AM <- ggplot(df_long_opioid_mean_toward, aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_opioid_mean_toward, aes(x=Session,y=predict(MLM_O_MT), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
-DV: Pain Mean Toward Bias
re_P_MT <- lmer(RT ~ (1|Subject), data = df_long_pain_mean_toward)
performance::icc(re_P_MT)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.713
## Conditional ICC: 0.713MLM_P_MT <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_mean_toward)
#Multiple Degree-of-Freedom F-Tests
anova(MLM_P_MT, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 821.77 821.77 1 26.945 1.2168 0.2797
## Condition 1851.76 925.88 2 82.606 1.3710 0.2596
## Session:Condition 1972.03 986.01 2 86.744 1.4600 0.2379#Means/SDs
df_long_pain_mean_toward %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition,Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 104. 48.4
## 2 Post A 98.7 50.2
## 3 Pre B 109. 40.3
## 4 Post B 117. 52.4
## 5 Pre C 112. 52.6
## 6 Post C 116. 63.5#Parameter Estimates
summary(MLM_P_MT)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_mean_toward
##
## REML criterion at convergence: 1295.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.86247 -0.50958 -0.05566 0.35560 2.93915
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 1567.1 39.59
## SessionPost 406.0 20.15 0.64
## Residual 675.3 25.99
## Number of obs: 134, groups: Subject, 27
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 110.895 9.816 46.666 11.298 5.96e-15 ***
## SessionPost -3.413 9.317 83.654 -0.366 0.715
## ConditionB -5.783 8.208 87.169 -0.705 0.483
## ConditionC 4.279 8.211 86.990 0.521 0.604
## SessionPost:ConditionB 19.381 11.342 87.833 1.709 0.091 .
## SessionPost:ConditionC 10.753 11.402 88.443 0.943 0.348
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.189
## ConditionB -0.475 0.473
## ConditionC -0.475 0.473 0.572
## SssnPst:CnB 0.328 -0.676 -0.698 -0.393
## SssnPst:CnC 0.325 -0.672 -0.390 -0.691 0.550#Actual Means
AM <- ggplot(df_long_pain_mean_toward, aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_pain_mean_toward, aes(x=Session,y=predict(MLM_P_MT), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
7d. Peak Away Bias
-DV: Opioid Peak Away Bias (NON-CONVERGENCE)
re_O_PA <- lmer(RT ~ (1|Subject), data = df_long_opioid_peak_away)
performance::icc(re_O_PA)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.617
## Conditional ICC: 0.617MLM_O_PA <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_peak_away)## boundary (singular) fit: see help('isSingular')isSingular(MLM_O_PA)## [1] TRUE#Multiple Degree-of-Freedom F-Tests
anova(MLM_O_PA, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 135.7 135.7 1 75.837 0.0105 0.9185
## Condition 26491.3 13245.7 2 105.491 1.0291 0.3609
## Session:Condition 7437.8 3718.9 2 102.397 0.2889 0.7497#Means/SDs
df_long_opioid_peak_away %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition,Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 329. 170.
## 2 Post A 324. 193.
## 3 Pre B 350. 163.
## 4 Post B 361. 197.
## 5 Pre C 372. 204.
## 6 Post C 351. 179.#Parameter Estimates
summary(MLM_O_PA)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_opioid_peak_away
##
## REML criterion at convergence: 1649.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.8671 -0.4296 -0.1134 0.4154 2.9405
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 17569.4 132.5
## SessionPost 615.2 24.8 1.00
## Residual 12871.3 113.5
## Number of obs: 134, groups: Subject, 27
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 340.128 37.039 58.896 9.183 5.76e-13 ***
## SessionPost -5.845 36.742 101.766 -0.159 0.874
## ConditionB -5.125 35.664 103.712 -0.144 0.886
## ConditionC 37.061 35.688 103.792 1.038 0.301
## SessionPost:ConditionB 28.445 49.077 102.345 0.580 0.563
## SessionPost:ConditionC -4.643 49.285 102.324 -0.094 0.925
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.407
## ConditionB -0.546 0.516
## ConditionC -0.546 0.516 0.570
## SssnPst:CnB 0.376 -0.735 -0.698 -0.390
## SssnPst:CnC 0.373 -0.732 -0.388 -0.691 0.548
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')#Actual Means
AM <- ggplot(df_long_opioid_peak_away, aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_opioid_peak_away, aes(x=Session,y=predict(MLM_O_PA), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
-DV: Pain Peak Away Bias
re_P_PA <- lmer(RT ~ (1|Subject), data = df_long_pain_peak_away)
performance::icc(re_P_PA)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.711
## Conditional ICC: 0.711MLM_P_PA <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_peak_away)
#Multiple Degree-of-Freedom F-Tests
anova(MLM_P_PA, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 5979.7 5979.7 1 23.462 0.6931 0.4135
## Condition 3272.4 1636.2 2 79.920 0.1897 0.8276
## Session:Condition 8763.9 4381.9 2 84.768 0.5079 0.6036#Means/SDs
df_long_pain_peak_away %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition, Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 334. 189.
## 2 Post A 372. 180.
## 3 Pre B 362. 153.
## 4 Post B 356. 173.
## 5 Pre C 353. 173.
## 6 Post C 342. 187.#Parameter Estimates
summary(MLM_P_PA)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_peak_away
##
## REML criterion at convergence: 1614.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.27661 -0.56489 -0.03049 0.41850 2.33591
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 20760 144.08
## SessionPost 1958 44.25 0.24
## Residual 8627 92.88
## Number of obs: 134, groups: Subject, 27
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 348.594 35.534 46.996 9.810 5.93e-13 ***
## SessionPost 36.423 31.343 87.597 1.162 0.248
## ConditionB -1.559 29.440 83.961 -0.053 0.958
## ConditionC 15.289 29.456 83.712 0.519 0.605
## SessionPost:ConditionB -22.266 40.465 85.894 -0.550 0.584
## SessionPost:ConditionC -40.975 40.655 86.696 -1.008 0.316
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.368
## ConditionB -0.472 0.505
## ConditionC -0.472 0.504 0.574
## SssnPst:CnB 0.327 -0.715 -0.701 -0.395
## SssnPst:CnC 0.324 -0.712 -0.393 -0.694 0.551#Actual Means
AM <- ggplot(df_long_pain_peak_away, aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_pain_peak_away, aes(x=Session,y=predict(MLM_P_PA), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
7e. Peak Toward Bias
-DV: Opioid Peak Toward Bias
re_O_PT <- lmer(RT ~ (1|Subject), data = df_long_opioid_peak_toward)
performance::icc(re_O_PA)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.617
## Conditional ICC: 0.617MLM_O_PT <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_peak_toward)## boundary (singular) fit: see help('isSingular')#Multiple Degree-of-Freedom F-Tests
anova(MLM_O_PA, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 135.7 135.7 1 75.837 0.0105 0.9185
## Condition 26491.3 13245.7 2 105.491 1.0291 0.3609
## Session:Condition 7437.8 3718.9 2 102.397 0.2889 0.7497#Means/SDs
df_long_opioid_peak_toward %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition,Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 312. 149.
## 2 Post A 362. 179.
## 3 Pre B 370. 170.
## 4 Post B 380. 162.
## 5 Pre C 358. 176.
## 6 Post C 345 188.#Parameter Estimates
summary(MLM_O_PA)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_opioid_peak_away
##
## REML criterion at convergence: 1649.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.8671 -0.4296 -0.1134 0.4154 2.9405
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 17569.4 132.5
## SessionPost 615.2 24.8 1.00
## Residual 12871.3 113.5
## Number of obs: 134, groups: Subject, 27
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 340.128 37.039 58.896 9.183 5.76e-13 ***
## SessionPost -5.845 36.742 101.766 -0.159 0.874
## ConditionB -5.125 35.664 103.712 -0.144 0.886
## ConditionC 37.061 35.688 103.792 1.038 0.301
## SessionPost:ConditionB 28.445 49.077 102.345 0.580 0.563
## SessionPost:ConditionC -4.643 49.285 102.324 -0.094 0.925
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.407
## ConditionB -0.546 0.516
## ConditionC -0.546 0.516 0.570
## SssnPst:CnB 0.376 -0.735 -0.698 -0.390
## SssnPst:CnC 0.373 -0.732 -0.388 -0.691 0.548
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')#Actual Means
AM <- ggplot(df_long_opioid_mean_toward, aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_opioid_mean_toward, aes(x=Session,y=predict(MLM_O_PT), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
-DV: Pain Peak Toward Bias
re_P_PT <- lmer(RT ~ (1|Subject), data = df_long_pain_peak_toward)
performance::icc(re_P_PT)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.699
## Conditional ICC: 0.699MLM_P_PT <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_peak_toward)
#Multiple Degree-of-Freedom F-Tests
anova(MLM_P_PT, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 2324.3 2324.3 1 25.267 0.3161 0.5789
## Condition 10172.3 5086.2 2 81.690 0.6917 0.5037
## Session:Condition 9286.3 4643.1 2 86.139 0.6314 0.5343#Means/SDs
df_long_pain_peak_toward %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition,Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 335. 150.
## 2 Post A 332. 163.
## 3 Pre B 347. 123.
## 4 Post B 360. 142.
## 5 Pre C 360. 173.
## 6 Post C 347. 204.#Parameter Estimates
summary(MLM_P_PT)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_peak_toward
##
## REML criterion at convergence: 1592.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.73146 -0.52681 -0.01424 0.43289 2.40516
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 14577 120.73
## SessionPost 2432 49.31 0.71
## Residual 7354 85.75
## Number of obs: 134, groups: Subject, 27
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 344.405 30.890 49.966 11.149 3.68e-15 ***
## SessionPost 1.814 29.338 87.748 0.062 0.951
## ConditionB -6.812 27.016 86.422 -0.252 0.802
## ConditionC 25.741 27.031 86.187 0.952 0.344
## SessionPost:ConditionB 31.058 37.270 87.350 0.833 0.407
## SessionPost:ConditionC -6.469 37.452 88.046 -0.173 0.863
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.256
## ConditionB -0.496 0.492
## ConditionC -0.496 0.491 0.571
## SssnPst:CnB 0.342 -0.702 -0.697 -0.391
## SssnPst:CnC 0.339 -0.699 -0.388 -0.690 0.549#Actual Means
AM <- ggplot(df_long_pain_peak_toward, aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_pain_peak_toward, aes(x=Session,y=predict(MLM_P_PT), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
7f. Variability Bias
-DV: Opioid Variability Bias (NON-CONVERGENCE)
re_O_V <- lmer(RT ~ (1|Subject), data = df_long_opioid_variability)
performance::icc(re_O_V)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.784
## Conditional ICC: 0.784MLM_O_V <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_variability)## boundary (singular) fit: see help('isSingular')isSingular(MLM_O_V)## [1] TRUE#Multiple Degree-of-Freedom F-Tests
anova(MLM_O_V, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 211.20 211.20 1 80.418 0.5456 0.4623
## Condition 1593.49 796.74 2 103.656 2.0582 0.1329
## Session:Condition 5.46 2.73 2 101.998 0.0071 0.9930#Means/SDs
df_long_opioid_variability %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition,Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 88.6 42.7
## 2 Post A 92.5 43.4
## 3 Pre B 98.1 30.7
## 4 Post B 95.9 39.5
## 5 Pre C 98.2 48.5
## 6 Post C 97.7 47.3#Parameter Estimates
summary(MLM_O_V)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_opioid_variability
##
## REML criterion at convergence: 1220.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.39724 -0.51084 -0.07423 0.45530 2.68996
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 1317.97 36.30
## SessionPost 13.84 3.72 1.00
## Residual 387.10 19.67
## Number of obs: 134, groups: Subject, 27
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 93.7981 8.4125 41.4107 11.150 4.78e-14 ***
## SessionPost 3.1267 6.3582 101.4801 0.492 0.624
## ConditionB 0.4783 6.2151 102.7932 0.077 0.939
## ConditionC 7.7974 6.2203 102.8389 1.254 0.213
## SessionPost:ConditionB -1.0091 8.5141 101.9806 -0.119 0.906
## SessionPost:ConditionC -0.6090 8.5484 101.9658 -0.071 0.943
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.288
## ConditionB -0.420 0.517
## ConditionC -0.420 0.517 0.572
## SssnPst:CnB 0.289 -0.737 -0.698 -0.391
## SssnPst:CnC 0.286 -0.734 -0.389 -0.690 0.548
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')#Actual Means
AM <- ggplot(df_long_opioid_variability, aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_opioid_variability, aes(x=Session,y=predict(MLM_O_V), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
-DV: Pain Variability Bias
re_P_V <- lmer(RT ~ (1|Subject), data = df_long_pain_variability)
performance::icc(re_P_V)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.804
## Conditional ICC: 0.804MLM_P_V <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_variability)
#Multiple Degree-of-Freedom F-Tests
anova(MLM_P_V, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 515.24 515.24 1 23.801 1.3752 0.2525
## Condition 815.83 407.91 2 78.248 1.0887 0.3417
## Session:Condition 15.49 7.74 2 83.947 0.0207 0.9796#Means/SDs
df_long_pain_variability %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition,Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 89.4 45.7
## 2 Post A 93.2 44.6
## 3 Pre B 98.2 35.7
## 4 Post B 98.3 43.2
## 5 Pre C 97.3 45.9
## 6 Post C 98.4 54.6#Parameter Estimates
summary(MLM_P_V)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_variability
##
## REML criterion at convergence: 1228.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.7438 -0.4918 -0.1046 0.4235 3.1886
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 1480.5 38.48
## SessionPost 149.3 12.22 0.52
## Residual 374.7 19.36
## Number of obs: 134, groups: Subject, 27
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 95.2362 8.7356 39.4413 10.902 1.81e-13 ***
## SessionPost 4.2747 6.7269 84.8506 0.635 0.527
## ConditionB -0.7355 6.1395 83.3470 -0.120 0.905
## ConditionC 5.2924 6.1429 83.0851 0.862 0.391
## SessionPost:ConditionB 1.5426 8.4454 85.1775 0.183 0.856
## SessionPost:ConditionC 0.2217 8.4874 85.9665 0.026 0.979
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.191
## ConditionB -0.400 0.490
## ConditionC -0.400 0.489 0.574
## SssnPst:CnB 0.277 -0.696 -0.699 -0.395
## SssnPst:CnC 0.275 -0.693 -0.392 -0.692 0.551#Actual Means
AM <- ggplot(df_long_pain_variability, aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_pain_variability, aes(x=Session,y=predict(MLM_P_V), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
8. Muiltilevel Models Separated by Dose examining Attentional Bias regressed on session*condition (fixed effects). Random intercepts per subject, random slopes across sessions (pre-post)
8a. Mean Bias
-DV: Opioid Mean Bias
re_O_MB_LD <- lmer(RT ~ (1|Subject), data = df_long_opioid_mean_bias %>% filter(Dose == "Low"))
re_O_MB_HD <- lmer(RT ~ (1|Subject), data = df_long_opioid_mean_bias %>% filter(Dose == "High"))
performance::icc(re_O_MB_LD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.043
## Conditional ICC: 0.043performance::icc(re_O_MB_HD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.231
## Conditional ICC: 0.231MLM_O_MB_LD <- lmer(RT ~ Session*Condition + (1 + Session|Subject), data = df_long_opioid_mean_bias %>% filter(Dose == "Low"))
MLM_O_MB_HD <- lmer(RT ~ Session*Condition + (1 + Session|Subject), data = df_long_opioid_mean_bias %>% filter(Dose == "High"))## boundary (singular) fit: see help('isSingular')#The code below then obtains mltiple Degree-of-Freedom F-Tests for the above-specified model
anova(MLM_O_MB_LD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 187.78 187.78 1 15.885 0.2607 0.6166
## Condition 199.17 99.58 2 51.463 0.1383 0.8712
## Session:Condition 1545.79 772.89 2 50.360 1.0731 0.3496anova(MLM_O_MB_HD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 193.58 193.58 1 11.849 0.3312 0.5757
## Condition 2612.92 1306.46 2 38.010 2.2353 0.1208
## Session:Condition 156.79 78.39 2 35.556 0.1341 0.8749#Model Parameters
summary(MLM_O_MB_LD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_opioid_mean_bias %>% filter(Dose == "Low")
##
## REML criterion at convergence: 769.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.8170 -0.6311 -0.0437 0.4117 3.4516
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 507.9 22.54
## SessionPost 840.4 28.99 -0.96
## Residual 720.2 26.84
## Number of obs: 84, groups: Subject, 16
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -5.894 9.514 39.086 -0.620 0.539
## SessionPost 8.037 13.003 41.673 0.618 0.540
## ConditionB 10.530 10.402 51.274 1.012 0.316
## ConditionC -2.869 10.504 50.044 -0.273 0.786
## SessionPost:ConditionB -14.898 14.530 50.744 -1.025 0.310
## SessionPost:ConditionC 5.206 14.929 49.672 0.349 0.729
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.798
## ConditionB -0.597 0.442
## ConditionC -0.578 0.426 0.529
## SssnPst:CnB 0.432 -0.619 -0.721 -0.382
## SssnPst:CnC 0.408 -0.589 -0.374 -0.707 0.527summary(MLM_O_MB_HD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_opioid_mean_bias %>% filter(Dose == "High")
##
## REML criterion at convergence: 426
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.18606 -0.50718 -0.09435 0.55229 3.02205
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 53.0 7.28
## SessionPost 103.1 10.15 1.00
## Residual 584.5 24.18
## Number of obs: 50, groups: Subject, 11
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 2.210 10.208 30.630 0.217 0.830
## SessionPost 9.670 13.911 30.831 0.695 0.492
## ConditionB 3.511 12.844 35.533 0.273 0.786
## ConditionC -12.442 12.600 35.525 -0.987 0.330
## SessionPost:ConditionB -9.120 17.963 35.504 -0.508 0.615
## SessionPost:ConditionC -6.626 17.387 35.396 -0.381 0.705
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.628
## ConditionB -0.758 0.538
## ConditionC -0.775 0.545 0.618
## SssnPst:CnB 0.527 -0.731 -0.698 -0.425
## SssnPst:CnC 0.538 -0.763 -0.426 -0.699 0.587
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')#Means/SDs
df_long_opioid_mean_bias %>% filter(Dose == "Low") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT))## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A -6.36 35.2
## 2 Pre B 4.70 30.8
## 3 Pre C -10.6 32.4
## 4 Post A 2.36 37.3
## 5 Post B -2.23 21.0
## 6 Post C 4.81 35.5df_long_opioid_mean_bias %>% filter(Dose == "High") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT))## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 0.291 23.2
## 2 Pre B 5.44 17.5
## 3 Pre C -10.7 35.9
## 4 Post A 11.5 30.1
## 5 Post B 5.75 29.1
## 6 Post C -8.41 25.2#Actual Means
AM_LD <- ggplot(df_long_opioid_mean_bias %>% filter(Dose == "Low"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot", position = position_dodge(width = 0.5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
AM_HD <- ggplot(df_long_opioid_mean_bias %>% filter(Dose == "High"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot", position = position_dodge(width = 0.5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM_LD <- ggplot(df_long_opioid_mean_bias %>% filter(Dose == "Low"), aes(x=Session,y=predict(MLM_O_MB_LD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot", position = position_dodge(width = 0.5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
PM_HD <- ggplot(df_long_opioid_mean_bias %>% filter(Dose == "High"), aes(x=Session,y=predict(MLM_O_MB_HD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot", position = position_dodge(width = 0.5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM_LD,AM_HD)
cowplot::plot_grid(PM_LD,PM_HD)
-DV: Pain Mean Bias (NON-CONVERGENCE PRESENT IN AT LEAST 1 MODEL)
re_P_MB_LD <- lmer(RT ~ (1|Subject), data = df_long_pain_mean_bias %>% filter(Dose == "Low"))## boundary (singular) fit: see help('isSingular')re_P_MB_HD <- lmer(RT ~ (1|Subject), data = df_long_pain_mean_bias %>% filter(Dose == "High"))## boundary (singular) fit: see help('isSingular')isSingular(re_P_MB_LD)## [1] TRUEisSingular(re_P_MB_HD) #non-convergence of random effects, suggests insufficient variance explained by grouping by subject## [1] TRUEperformance::icc(re_P_MB_HD)## Warning: Can't compute random effect variances. Some variance components equal
## zero. Your model may suffer from singularity (see '?lme4::isSingular'
## and '?performance::check_singularity').
## Solution: Respecify random structure! You may also decrease the
## 'tolerance' level to enforce the calculation of random effect variances.## [1] NAMLM_P_MB_LD <- lmer(RT ~ Session*Condition + (1 + Session|Subject), data = df_long_pain_mean_bias %>% filter(Dose == "Low"),
control = lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e5)))## boundary (singular) fit: see help('isSingular')isSingular(MLM_P_MB_LD) #non-convergence## [1] TRUEMLM_P_MB_HD <- lmer(RT ~ Session*Condition + (1 + Session|Subject), data = df_long_pain_mean_bias %>% filter(Dose == "High"),
control = lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e5)))## boundary (singular) fit: see help('isSingular')## Warning: Model failed to converge with 1 negative eigenvalue: -8.1e-01isSingular(MLM_P_MB_HD) #non-convergence## [1] TRUE8b. Mean Away Bias
-DV: Opioid Mean Away Bias (NON-CONVERGENCE PRESENT IN AT LEAST 1 MODEL)
re_O_MA_LD <- lmer(RT ~ (1|Subject), data = df_long_opioid_mean_away %>% filter(Dose == "Low"))
re_O_MA_HD <- lmer(RT ~ (1|Subject), data = df_long_opioid_mean_away %>% filter(Dose == "High"))
performance::icc(re_O_MA_LD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.656
## Conditional ICC: 0.656performance::icc(re_O_MA_HD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.625
## Conditional ICC: 0.625MLM_O_MA_LD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_mean_away %>% filter(Dose == "Low"),
control = lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e5))) #non-convergence## boundary (singular) fit: see help('isSingular')isSingular(MLM_O_MA_LD) #non-convergence## [1] TRUEMLM_O_MA_HD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_mean_away %>% filter(Dose == "High"))
#Multiple Degree-of-Freedom F-Tests
anova(MLM_O_MA_HD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 68.561 68.561 1 10.219 0.0919 0.7678
## Condition 104.643 52.321 2 25.735 0.0701 0.9324
## Session:Condition 222.454 111.227 2 27.119 0.1491 0.8622#Means/SDs
df_long_opioid_mean_away %>% filter(Dose == "High") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition, Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 95.5 35.6
## 2 Post A 98.4 86.7
## 3 Pre B 116. 53.9
## 4 Post B 109. 43.1
## 5 Pre C 117. 54.7
## 6 Post C 114. 51.1#Parameter Estimates
summary(MLM_O_MA_HD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_opioid_mean_away %>% filter(Dose == "High")
##
## REML criterion at convergence: 464.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.2639 -0.4723 -0.1453 0.4648 2.7191
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 1965.8 44.34
## SessionPost 1570.9 39.63 -0.25
## Residual 745.9 27.31
## Number of obs: 50, groups: Subject, 11
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 114.5756 18.1553 20.3074 6.311 3.43e-06 ***
## SessionPost -10.2133 20.3298 25.6452 -0.502 0.620
## ConditionB -4.5673 15.6375 27.6238 -0.292 0.772
## ConditionC 0.8734 15.5171 28.1789 0.056 0.956
## SessionPost:ConditionB 11.5537 21.3601 27.6550 0.541 0.593
## SessionPost:ConditionC 5.7990 20.9829 28.8668 0.276 0.784
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.504
## ConditionB -0.534 0.463
## ConditionC -0.555 0.477 0.665
## SssnPst:CnB 0.381 -0.609 -0.714 -0.468
## SssnPst:CnC 0.394 -0.650 -0.469 -0.712 0.620#Actual Means
AM <- ggplot(df_long_opioid_mean_away %>% filter(Dose == "High"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_opioid_mean_away %>% filter(Dose == "High"), aes(x=Session,y=predict(MLM_O_MA_HD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM, PM)
-DV: Pain Mean Away Bias (NON-CONVERGENCE PRESENT IN AT LEAST 1 MODEL)
re_P_MA_LD <- lmer(RT ~ (1|Subject), data = df_long_pain_mean_away %>% filter(Dose == "Low"))
re_P_MA_HD <- lmer(RT ~ (1|Subject), data = df_long_pain_mean_away %>% filter(Dose == "High"))
performance::icc(re_P_MA_LD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.725
## Conditional ICC: 0.725performance::icc(re_P_MA_HD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.792
## Conditional ICC: 0.792MLM_P_MA_LD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_mean_away %>% filter(Dose == "Low"),
control = lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e5)))## boundary (singular) fit: see help('isSingular')isSingular(MLM_P_MA_LD) #non-convergence## [1] TRUEMLM_P_MA_HD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_mean_away %>% filter(Dose == "High"))
#Multiple Degree-of-Freedom F-Tests
anova(MLM_P_MA_HD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 573.40 573.40 1 6.8012 0.8775 0.3810
## Condition 998.02 499.01 2 20.4368 0.7636 0.4788
## Session:Condition 1876.09 938.05 2 23.8855 1.4355 0.2578#Means/SDs
df_long_pain_mean_away %>% filter(Dose == "High") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition, Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 92.6 26.5
## 2 Post A 118. 82.2
## 3 Pre B 120. 59.5
## 4 Post B 101. 45.9
## 5 Pre C 110. 61.3
## 6 Post C 122. 77.2#Parameter Estimates
summary(MLM_P_MA_HD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_mean_away %>% filter(Dose == "High")
##
## REML criterion at convergence: 458.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.73299 -0.55854 0.02361 0.37982 2.38923
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 2457.0 49.57
## SessionPost 895.2 29.92 0.77
## Residual 653.5 25.56
## Number of obs: 50, groups: Subject, 11
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 108.8460 18.6088 17.4479 5.849 1.74e-05 ***
## SessionPost 28.3623 17.2569 24.4171 1.644 0.113
## ConditionB 4.0595 14.1755 24.7819 0.286 0.777
## ConditionC -0.4901 14.0046 25.5106 -0.035 0.972
## SessionPost:ConditionB -32.7531 19.3309 25.1726 -1.694 0.103
## SessionPost:ConditionC -19.1714 18.7727 27.4501 -1.021 0.316
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.030
## ConditionB -0.467 0.468
## ConditionC -0.482 0.473 0.646
## SssnPst:CnB 0.322 -0.639 -0.692 -0.431
## SssnPst:CnC 0.326 -0.673 -0.430 -0.683 0.597#Actual Means
AM <- ggplot(df_long_pain_mean_away %>% filter(Dose == "High"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_pain_mean_away %>% filter(Dose == "High"), aes(x=Session,y=predict(MLM_P_MA_HD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
8c. Mean Toward Bias
-DV: Opioid Mean Toward Bias
re_O_MT_LD <- lmer(RT ~ (1|Subject), data = df_long_opioid_mean_toward %>% filter(Dose == "Low"))
re_O_MT_HD <- lmer(RT ~ (1|Subject), data = df_long_opioid_mean_toward %>% filter(Dose == "High"))
performance::icc(re_O_MT_LD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.598
## Conditional ICC: 0.598performance::icc(re_O_MT_HD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.871
## Conditional ICC: 0.871MLM_O_MT_LD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_mean_toward %>% filter(Dose == "Low"),
control = lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e5)))## boundary (singular) fit: see help('isSingular')isSingular(MLM_O_MT_LD)## [1] TRUEMLM_O_MT_HD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_mean_toward %>% filter(Dose == "High"))
#Multiple Degree-of-Freedom F-Tests
anova(MLM_O_MT_HD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 176.49 176.49 1 8.7816 0.6718 0.43412
## Condition 218.14 109.07 2 23.8751 0.4152 0.66491
## Session:Condition 1910.39 955.20 2 26.9879 3.6357 0.03998 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#Means/SDs
df_long_opioid_mean_toward %>% filter(Dose == "High") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition, Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 82.2 39.0
## 2 Post A 113. 60.8
## 3 Pre B 120. 49.7
## 4 Post B 108. 43.3
## 5 Pre C 113. 44.1
## 6 Post C 106. 54.9#Parameter Estimates
summary(MLM_O_MT_HD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_opioid_mean_toward %>% filter(Dose == "High")
##
## REML criterion at convergence: 421.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.9215 -0.4365 -0.0448 0.4367 2.3358
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 1817.9 42.64
## SessionPost 165.8 12.87 0.63
## Residual 262.7 16.21
## Number of obs: 50, groups: Subject, 11
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 98.736 14.688 14.455 6.722 8.28e-06 ***
## SessionPost 24.019 10.078 29.991 2.383 0.0237 *
## ConditionB 17.191 9.078 27.237 1.894 0.0689 .
## ConditionC 14.844 8.985 27.768 1.652 0.1098
## SessionPost:ConditionB -26.043 12.244 28.002 -2.127 0.0424 *
## SessionPost:ConditionC -30.837 11.864 29.793 -2.599 0.0144 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.103
## ConditionB -0.380 0.511
## ConditionC -0.393 0.518 0.652
## SssnPst:CnB 0.263 -0.693 -0.695 -0.436
## SssnPst:CnC 0.267 -0.728 -0.436 -0.688 0.597#Actual Means
AM <- ggplot(df_long_opioid_mean_toward %>% filter(Dose == "High"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_opioid_mean_toward %>% filter(Dose == "High"), aes(x=Session,y=predict(MLM_O_MT_HD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
-DV: Pain Mean Toward Bias (NON-CONVERGENCE PRESENT IN AT LEAST 1 MODEL)
re_P_MT_LD <- lmer(RT ~ (1|Subject), data = df_long_pain_mean_toward %>% filter(Dose == "Low"))
re_P_MT_HD <- lmer(RT ~ (1|Subject), data = df_long_pain_mean_toward %>% filter(Dose == "High"))
performance::icc(re_P_MT_LD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.663
## Conditional ICC: 0.663performance::icc(re_P_MT_HD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.795
## Conditional ICC: 0.795MLM_P_MT_LD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_mean_toward %>% filter(Dose == "Low"))
MLM_P_MT_HD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_mean_toward %>% filter(Dose == "High"),
control = lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e5)))## boundary (singular) fit: see help('isSingular')isSingular(MLM_P_MT_HD) #non-convergence## [1] TRUE#Multiple Degree-of-Freedom F-Tests
anova(MLM_P_MT_LD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 331.91 331.91 1 14.909 0.4974 0.4915
## Condition 1140.08 570.04 2 50.204 0.8543 0.4317
## Session:Condition 1839.20 919.60 2 51.894 1.3782 0.2611#Means/SDs
df_long_pain_mean_toward %>% filter(Dose == "Low") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition,Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 108. 55.3
## 2 Post A 97.8 48.3
## 3 Pre B 109. 38.1
## 4 Post B 117. 54.3
## 5 Pre C 106. 41.2
## 6 Post C 114. 64.4#Parameter Estimates
summary(MLM_P_MT_LD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_mean_toward %>% filter(Dose == "Low")
##
## REML criterion at convergence: 794
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.61649 -0.46796 -0.03134 0.37930 2.77210
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 1311.8 36.22
## SessionPost 695.8 26.38 0.36
## Residual 667.3 25.83
## Number of obs: 84, groups: Subject, 16
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 112.368 11.683 26.670 9.618 3.67e-10 ***
## SessionPost -7.360 12.276 41.859 -0.600 0.552
## ConditionB -6.879 10.012 52.003 -0.687 0.495
## ConditionC -2.025 10.109 50.990 -0.200 0.842
## SessionPost:ConditionB 18.493 13.941 52.591 1.326 0.190
## SessionPost:ConditionC 22.172 14.350 51.148 1.545 0.129
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.216
## ConditionB -0.468 0.429
## ConditionC -0.453 0.421 0.529
## SssnPst:CnB 0.325 -0.627 -0.704 -0.372
## SssnPst:CnC 0.316 -0.594 -0.369 -0.698 0.523#Actual Means
AM <- ggplot(df_long_pain_mean_toward %>% filter(Dose == "Low"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_pain_mean_toward %>% filter(Dose == "Low"), aes(x=Session,y=predict(MLM_P_MT_LD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
8d. Peak Away Bias
-DV: Opioid Peak Away Bias
re_O_PA_LD <- lmer(RT ~ (1|Subject), data = df_long_opioid_peak_away %>% filter(Dose == "Low"))
re_O_PA_HD <- lmer(RT ~ (1|Subject), data = df_long_opioid_peak_away %>% filter(Dose == "High"))
performance::icc(re_O_PA_LD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.609
## Conditional ICC: 0.609performance::icc(re_O_PA_HD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.670
## Conditional ICC: 0.670MLM_O_PA_LD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_peak_away %>% filter(Dose == "Low"),
control = lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e5)))## boundary (singular) fit: see help('isSingular')isSingular(MLM_O_PA_LD)## [1] TRUEMLM_O_PA_HD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_peak_away %>% filter(Dose == "High"))
#Multiple Degree-of-Freedom F-Tests
anova(MLM_O_PA_HD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 11388.0 11388.0 1 9.6686 1.5581 0.2413
## Condition 29635.4 14817.7 2 26.1029 2.0274 0.1519
## Session:Condition 1305.4 652.7 2 28.0549 0.0893 0.9148#Means/SDs
df_long_opioid_peak_away %>% filter(Dose == "High") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition,Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 277. 105.
## 2 Post A 269. 172.
## 3 Pre B 410. 206.
## 4 Post B 338. 185.
## 5 Pre C 346 148.
## 6 Post C 318. 155.#Parameter Estimates
summary(MLM_O_PA_HD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_opioid_peak_away %>% filter(Dose == "High")
##
## REML criterion at convergence: 558.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.95197 -0.40054 0.00381 0.25373 2.82412
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 16796 129.60
## SessionPost 5716 75.60 -0.07
## Residual 7309 85.49
## Number of obs: 50, groups: Subject, 11
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 320.10 54.39 21.50 5.885 6.99e-06 ***
## SessionPost -42.99 54.97 29.40 -0.782 0.440
## ConditionB 71.82 48.24 28.65 1.489 0.148
## ConditionC 20.13 47.77 29.26 0.421 0.677
## SessionPost:ConditionB -11.93 65.45 28.85 -0.182 0.857
## SessionPost:ConditionC 12.28 63.75 30.30 0.193 0.849
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.476
## ConditionB -0.548 0.514
## ConditionC -0.567 0.525 0.656
## SssnPst:CnB 0.386 -0.685 -0.706 -0.452
## SssnPst:CnC 0.396 -0.723 -0.453 -0.703 0.607#Actual Means
AM <- ggplot(df_long_opioid_peak_away %>% filter(Dose == "High"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_opioid_peak_away %>% filter(Dose == "High"), aes(x=Session,y=predict(MLM_O_PA_HD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM,PM)
-DV: Pain Peak Away Bias
re_P_PA_LD <- lmer(RT ~ (1|Subject), data = df_long_pain_peak_away %>% filter(Dose == "Low"))
re_P_PA_HD <- lmer(RT ~ (1|Subject), data = df_long_pain_peak_away %>% filter(Dose == "High"))
performance::icc(re_P_PA_LD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.715
## Conditional ICC: 0.715performance::icc(re_P_PA_HD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.722
## Conditional ICC: 0.722MLM_P_PA_LD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_peak_away %>% filter(Dose == "Low"))
MLM_P_PA_HD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_peak_away %>% filter(Dose == "High"))
#Multiple Degree-of-Freedom F-Tests
anova(MLM_P_PA_LD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 808.1 808.1 1 14.986 0.0949 0.7622
## Condition 2683.0 1341.5 2 51.348 0.1576 0.8546
## Session:Condition 7638.2 3819.1 2 53.361 0.4487 0.6409anova(MLM_P_PA_HD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 8279.8 8279.8 1 6.6396 0.8913 0.3782
## Condition 13618.7 6809.3 2 22.2618 0.7330 0.4917
## Session:Condition 24627.0 12313.5 2 25.3604 1.3255 0.2835#Means/SDs
df_long_pain_peak_away %>% filter(Dose == "Low") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition, Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 366. 215.
## 2 Post A 376. 177.
## 3 Pre B 371. 134.
## 4 Post B 387. 181.
## 5 Pre C 353. 157.
## 6 Post C 324. 177.df_long_pain_peak_away %>% filter(Dose == "High") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition, Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 264. 93.1
## 2 Post A 364. 200.
## 3 Pre B 347. 187.
## 4 Post B 293. 148.
## 5 Pre C 354. 203.
## 6 Post C 364 208.#Parameter Estimates
summary(MLM_P_PA_LD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_peak_away %>% filter(Dose == "Low")
##
## REML criterion at convergence: 985
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.10401 -0.56847 -0.02793 0.45852 1.91856
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 20438 142.96
## SessionPost 1558 39.47 0.06
## Residual 8512 92.26
## Number of obs: 84, groups: Subject, 16
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 369.33459 44.40999 25.29642 8.316 1.05e-08 ***
## SessionPost 8.96094 37.76617 53.74150 0.237 0.813
## ConditionB -11.45274 35.73417 52.87229 -0.320 0.750
## ConditionC 0.05451 36.08540 51.93424 0.002 0.999
## SessionPost:ConditionB 20.21971 49.37309 53.99170 0.410 0.684
## SessionPost:ConditionC -26.14935 51.00515 52.68777 -0.513 0.610
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.382
## ConditionB -0.440 0.491
## ConditionC -0.425 0.484 0.529
## SssnPst:CnB 0.303 -0.713 -0.704 -0.371
## SssnPst:CnC 0.296 -0.685 -0.369 -0.697 0.524summary(MLM_P_PA_HD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_peak_away %>% filter(Dose == "High")
##
## REML criterion at convergence: 566.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.68813 -0.38241 0.02226 0.47663 2.46423
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 21813 147.69
## SessionPost 2177 46.66 0.68
## Residual 9290 96.38
## Number of obs: 50, groups: Subject, 11
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 303.46 60.95 20.48 4.978 6.76e-05 ***
## SessionPost 96.79 56.25 30.02 1.721 0.0956 .
## ConditionB 26.12 53.29 26.11 0.490 0.6281
## ConditionC 51.11 52.64 26.79 0.971 0.3403
## SessionPost:ConditionB -113.70 72.07 26.68 -1.578 0.1265
## SessionPost:ConditionC -87.45 69.61 28.81 -1.256 0.2191
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.339
## ConditionB -0.535 0.533
## ConditionC -0.552 0.538 0.644
## SssnPst:CnB 0.369 -0.727 -0.692 -0.429
## SssnPst:CnC 0.373 -0.759 -0.429 -0.686 0.590#Actual Means
AM_LD <- ggplot(df_long_pain_peak_away %>% filter(Dose == "Low"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
AM_HD <- ggplot(df_long_pain_peak_away %>% filter(Dose == "High"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM_LD <- ggplot(df_long_pain_peak_away %>% filter(Dose == "Low"), aes(x=Session,y=predict(MLM_P_PA_LD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
PM_HD <- ggplot(df_long_pain_peak_away %>% filter(Dose == "High"), aes(x=Session,y=predict(MLM_P_PA_HD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM_LD,PM_LD, AM_HD,PM_HD)
8e. Peak Toward Bias
-DV: Opioid Peak Toward Bias (NON-CONVERGENCE PRESENT IN AT LEAST 1 MODEL)
re_O_PT_LD <- lmer(RT ~ (1|Subject), data = df_long_opioid_peak_toward %>% filter(Dose == "Low"))
re_O_PT_HD <- lmer(RT ~ (1|Subject), data = df_long_opioid_peak_toward %>% filter(Dose == "High"))
performance::icc(re_O_PA_LD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.609
## Conditional ICC: 0.609performance::icc(re_O_PA_HD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.670
## Conditional ICC: 0.670MLM_O_PT_LD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_peak_toward %>% filter(Dose == "Low"),
control = lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e5)))## boundary (singular) fit: see help('isSingular')isSingular(MLM_O_PT_LD) #non-convergence## [1] TRUEMLM_O_PT_HD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_peak_toward %>% filter(Dose == "High"),
control = lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e5)))## boundary (singular) fit: see help('isSingular')isSingular(MLM_O_PT_HD) #non-convergence## [1] TRUE-DV: Pain Peak Toward Bias (NON-CONVERGENCE PRESENT IN AT LEAST 1 MODEL)
re_P_PT_LD <- lmer(RT ~ (1|Subject), data = df_long_pain_peak_toward %>% filter(Dose == "Low"))
re_P_PT_HD <- lmer(RT ~ (1|Subject), data = df_long_pain_peak_toward %>% filter(Dose == "High"))
performance::icc(re_P_PT_LD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.662
## Conditional ICC: 0.662performance::icc(re_P_PT_HD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.786
## Conditional ICC: 0.786MLM_P_PT_LD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_peak_toward %>% filter(Dose == "Low"))
MLM_P_PT_HD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_peak_toward %>% filter(Dose == "High"),
control = lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e5)))## boundary (singular) fit: see help('isSingular')isSingular(MLM_P_PT_HD) #non-convergence## [1] TRUE#Multiple Degree-of-Freedom F-Tests
anova(MLM_P_PT_LD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 79.1 79.1 1 14.297 0.0094 0.9242
## Condition 11509.0 5754.5 2 50.472 0.6825 0.5100
## Session:Condition 5796.0 2898.0 2 52.509 0.3437 0.7107#Means/SDs
df_long_pain_peak_toward %>% filter(Dose == "Low") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition,Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 364. 166.
## 2 Post A 353. 180.
## 3 Pre B 349. 133.
## 4 Post B 365. 147.
## 5 Pre C 374. 156.
## 6 Post C 350 221.#Parameter Estimates
summary(MLM_P_PT_LD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_peak_toward %>% filter(Dose == "Low")
##
## REML criterion at convergence: 982.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.41907 -0.50753 -0.02376 0.42226 2.29480
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 13364 115.60
## SessionPost 3378 58.12 0.63
## Residual 8432 91.83
## Number of obs: 84, groups: Subject, 16
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 367.617 38.923 29.321 9.445 2.13e-10 ***
## SessionPost -7.565 39.126 50.144 -0.193 0.847
## ConditionB -27.920 35.398 52.488 -0.789 0.434
## ConditionC 19.030 35.805 51.432 0.531 0.597
## SessionPost:ConditionB 33.055 49.144 53.234 0.673 0.504
## SessionPost:ConditionC -3.090 50.757 51.805 -0.061 0.952
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.252
## ConditionB -0.495 0.468
## ConditionC -0.481 0.463 0.529
## SssnPst:CnB 0.340 -0.686 -0.701 -0.369
## SssnPst:CnC 0.334 -0.657 -0.368 -0.695 0.524#Actual Means
AM <- ggplot(df_long_pain_peak_toward %>% filter(Dose == "Low"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_pain_peak_toward %>% filter(Dose == "Low"), aes(x=Session,y=predict(MLM_P_PT_LD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM_LD,PM_LD)
8f. Variability Bias
-DV: Opioid Variability Bias (NON-CONVERGENCE PRESENT IN AT LEAST 1 MODEL)
re_O_V_LD <- lmer(RT ~ (1|Subject), data = df_long_opioid_variability %>% filter(Dose == "Low"))
re_O_V_HD <- lmer(RT ~ (1|Subject), data = df_long_opioid_variability %>% filter(Dose == "High"))
performance::icc(re_O_V_LD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.776
## Conditional ICC: 0.776performance::icc(re_O_V_HD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.827
## Conditional ICC: 0.827MLM_O_V_LD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_variability %>% filter(Dose == "Low"),
control = lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e5)))## boundary (singular) fit: see help('isSingular')isSingular(MLM_O_V_LD) #non-convergence## [1] TRUEMLM_O_V_HD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_opioid_variability %>% filter(Dose == "High"))
#Multiple Degree-of-Freedom F-Tests
anova(MLM_O_V_HD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 4.884 4.884 1 11.123 0.0216 0.8857
## Condition 183.934 91.967 2 25.837 0.4071 0.6698
## Session:Condition 66.589 33.294 2 28.069 0.1474 0.8636#Means/SDs
df_long_opioid_variability %>% filter(Dose == "High") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition,Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 80.2 31.5
## 2 Post A 88.7 56.0
## 3 Pre B 100. 36.8
## 4 Post B 91.1 31.9
## 5 Pre C 99.1 39.2
## 6 Post C 96.0 45.3#Parameter Estimates
summary(MLM_O_V_HD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_opioid_variability %>% filter(Dose == "High")
##
## REML criterion at convergence: 415.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.69102 -0.60651 0.04626 0.39193 2.38422
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 1175.6 34.29
## SessionPost 254.6 15.96 0.24
## Residual 225.9 15.03
## Number of obs: 50, groups: Subject, 11
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 91.173 12.314 15.913 7.404 1.54e-06 ***
## SessionPost 2.593 10.069 29.473 0.257 0.799
## ConditionB 5.192 8.527 28.385 0.609 0.547
## ConditionC 8.403 8.453 28.889 0.994 0.328
## SessionPost:ConditionB -4.687 11.555 28.771 -0.406 0.688
## SessionPost:ConditionC -6.010 11.272 30.126 -0.533 0.598
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.244
## ConditionB -0.428 0.494
## ConditionC -0.444 0.505 0.659
## SssnPst:CnB 0.301 -0.661 -0.704 -0.452
## SssnPst:CnC 0.308 -0.699 -0.452 -0.699 0.609#Actual Means
AM <- ggplot(df_long_opioid_variability %>% filter(Dose == "High"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM <- ggplot(df_long_opioid_variability %>% filter(Dose == "High"), aes(x=Session,y=predict(MLM_O_V_HD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM_HD,PM_HD)
-DV: Pain Variability Bias
re_P_V_LD <- lmer(RT ~ (1|Subject), data = df_long_pain_variability %>% filter(Dose == "Low"))
re_P_V_HD <- lmer(RT ~ (1|Subject), data = df_long_pain_variability %>% filter(Dose == "High"))
performance::icc(re_P_V_LD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.761
## Conditional ICC: 0.761performance::icc(re_P_V_HD)## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.889
## Conditional ICC: 0.889MLM_P_V_LD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_variability %>% filter(Dose == "Low"))
MLM_P_V_HD <- lmer(RT ~ Session*Condition + (1 + Session |Subject), data = df_long_pain_variability %>% filter(Dose == "High"))
#Multiple Degree-of-Freedom F-Tests
anova(MLM_P_V_LD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 174.37 174.371 1 13.149 0.3557 0.5610
## Condition 586.42 293.212 2 48.617 0.5982 0.5538
## Session:Condition 119.71 59.855 2 51.141 0.1221 0.8853anova(MLM_P_V_HD, type = "III")## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Session 341.56 341.56 1 9.4895 1.7757 0.2138
## Condition 219.77 109.89 2 24.1668 0.5713 0.5722
## Session:Condition 203.66 101.83 2 27.0957 0.5294 0.5949#Means/SDs
df_long_pain_variability %>% filter(Dose == "Low") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition,Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 94.4 52.4
## 2 Post A 93.8 45.1
## 3 Pre B 99.6 30.5
## 4 Post B 101. 45.0
## 5 Pre C 96.7 45.1
## 6 Post C 96.2 57.0df_long_pain_variability %>% filter(Dose == "High") %>% select(-Cue, -Bias_Type) %>% group_by(Session, Condition) %>% summarise(M = mean(RT), SD = sd(RT)) %>% arrange(Condition,Session)## `summarise()` has grouped output by 'Session'. You can override using the
## `.groups` argument.## # A tibble: 6 × 4
## # Groups: Session [2]
## Session Condition M SD
## <fct> <chr> <dbl> <dbl>
## 1 Pre A 78.5 26.6
## 2 Post A 92.0 47.3
## 3 Pre B 95.9 45.2
## 4 Post B 91.9 41.4
## 5 Pre C 98.1 49.5
## 6 Post C 101. 54.2#Parameter Estimates
summary(MLM_P_V_LD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_variability %>% filter(Dose == "Low")
##
## REML criterion at convergence: 767.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.5495 -0.5076 -0.1696 0.4355 2.7427
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 1427.4 37.78
## SessionPost 151.0 12.29 0.43
## Residual 490.2 22.14
## Number of obs: 84, groups: Subject, 16
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 98.08464 11.36428 23.35943 8.631 1e-08 ***
## SessionPost -0.01073 9.28602 50.23042 -0.001 0.999
## ConditionB -1.77172 8.56859 50.73274 -0.207 0.837
## ConditionC 3.55380 8.65490 49.64885 0.411 0.683
## SessionPost:ConditionB 5.16185 11.85777 51.93351 0.435 0.665
## SessionPost:ConditionC 5.23287 12.24614 50.34591 0.427 0.671
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.242
## ConditionB -0.412 0.478
## ConditionC -0.398 0.472 0.529
## SssnPst:CnB 0.283 -0.698 -0.702 -0.370
## SssnPst:CnC 0.278 -0.668 -0.369 -0.696 0.524summary(MLM_P_V_HD)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Session * Condition + (1 + Session | Subject)
## Data: df_long_pain_variability %>% filter(Dose == "High")
##
## REML criterion at convergence: 412
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.54854 -0.58499 0.05419 0.53893 1.79712
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 1719.1 41.46
## SessionPost 213.7 14.62 0.60
## Residual 192.4 13.87
## Number of obs: 50, groups: Subject, 11
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 89.816 13.922 13.525 6.452 1.79e-05 ***
## SessionPost 13.497 9.189 28.619 1.469 0.153
## ConditionB 1.707 7.817 27.278 0.218 0.829
## ConditionC 9.007 7.743 27.791 1.163 0.255
## SessionPost:ConditionB -5.752 10.569 27.973 -0.544 0.591
## SessionPost:ConditionC -10.532 10.276 29.629 -1.025 0.314
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) SssnPs CndtnB CndtnC SsP:CB
## SessionPost -0.015
## ConditionB -0.346 0.489
## ConditionC -0.358 0.497 0.656
## SssnPst:CnB 0.241 -0.659 -0.698 -0.443
## SssnPst:CnC 0.246 -0.695 -0.443 -0.692 0.602#Actual Means
AM_LD <- ggplot(df_long_pain_variability %>% filter(Dose == "Low"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
AM_HD <- ggplot(df_long_pain_variability %>% filter(Dose == "High"), aes(x=Session,y=RT, col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
#Predicted Means
PM_LD <- ggplot(df_long_pain_variability %>% filter(Dose == "Low"), aes(x=Session,y=predict(MLM_P_V_LD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
PM_HD <- ggplot(df_long_pain_variability %>% filter(Dose == "High"), aes(x=Session,y=predict(MLM_P_V_HD), col = Condition)) +
stat_summary(fun.data = "mean_cl_boot",position = position_dodge(width = .5)) + ggtitle("Group Means") +
theme_light() +
theme(plot.title = element_text(hjust = 0.5))
cowplot::plot_grid(AM_LD,PM_LD,AM_HD,PM_HD)