library(lme4)## 載入需要的套件：Matrixlibrary(lmerTest)##
## 載入套件：'lmerTest'## 下列物件被遮斷自 'package:lme4':
##
## lmer## 下列物件被遮斷自 'package:stats':
##
## steplibrary(readxl)
Exp2_SCJT_ANOVA_MEM_final <- read_excel("C:/Users/openw/Desktop/CCUPSY/Research/exp4(exp2)/Data/MEM/Exp2 SCJT_ANOVA_MEM_final.xls")
View(Exp2_SCJT_ANOVA_MEM_final)library(readxl)
Exp2_Naming_ANOVA_MEM_final <- read_excel("C:/Users/openw/Desktop/CCUPSY/Research/exp4(exp2)/Data/MEM/Exp2 Naming_ANOVA_MEM_final.xls")
View(Exp2_Naming_ANOVA_MEM_final)##Exp2(enemy) SCJT_ACC
#隨機截距&隨機斜率
m9_lme = lmer(ACC~Task*Condition+(1+Condition|Subject)+(1|ID),data=Exp2_SCJT_ANOVA_MEM_final)
summary(m9_lme)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: ACC ~ Task * Condition + (1 + Condition | Subject) + (1 | ID)
## Data: Exp2_SCJT_ANOVA_MEM_final
##
## REML criterion at convergence: -550.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.5859 -0.0299 0.1363 0.3504 1.7933
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 0.002116 0.04600
## Conditionunlearned 0.001861 0.04314 -0.55
## ID (Intercept) 0.004737 0.06882
## Residual 0.046654 0.21600
## Number of obs: 3080, groups: Subject, 28; ID, 22
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 9.443e-01 2.139e-02 7.110e+01 44.136 <2e-16 ***
## Task 6.494e-04 3.892e-03 3.002e+03 0.167 0.868
## Conditionunlearned 5.998e-03 2.000e-02 2.482e+02 0.300 0.765
## Task:Conditionunlearned -3.571e-03 5.504e-03 3.002e+03 -0.649 0.516
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Task Cndtnn
## Task -0.546
## Condtnnlrnd -0.481 0.584
## Tsk:Cndtnnl 0.386 -0.707 -0.826#只有隨機截距
m10_lme = lmer(ACC~Task*Condition+(1|Subject)+(1|ID),data=Exp2_SCJT_ANOVA_MEM_final)
summary(m10_lme)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: ACC ~ Task * Condition + (1 | Subject) + (1 | ID)
## Data: Exp2_SCJT_ANOVA_MEM_final
##
## REML criterion at convergence: -541.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.5739 -0.0068 0.1356 0.3426 1.7982
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.001520 0.03898
## ID (Intercept) 0.004723 0.06873
## Residual 0.047087 0.21700
## Number of obs: 3080, groups: Subject, 28; ID, 22
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 9.442e-01 2.091e-02 7.069e+01 45.157 <2e-16 ***
## Task 6.494e-04 3.910e-03 3.028e+03 0.166 0.868
## Conditionunlearned 6.000e-03 1.835e-02 3.030e+03 0.327 0.744
## Task:Conditionunlearned -3.571e-03 5.530e-03 3.028e+03 -0.646 0.518
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Task Cndtnn
## Task -0.561
## Condtnnlrnd -0.439 0.639
## Tsk:Cndtnnl 0.397 -0.707 -0.904anova(m9_lme,m10_lme)## refitting model(s) with ML (instead of REML)## Data: Exp2_SCJT_ANOVA_MEM_final
## Models:
## m10_lme: ACC ~ Task * Condition + (1 | Subject) + (1 | ID)
## m9_lme: ACC ~ Task * Condition + (1 + Condition | Subject) + (1 | ID)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## m10_lme 7 -560.33 -518.10 287.17 -574.33
## m9_lme 9 -564.83 -510.53 291.41 -582.83 8.4943 2 0.0143 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#選擇m9，隨機截距&隨機斜率##Exp2(enemy) SCJT_RT
#隨機截距&隨機斜率
m11_lme = lmer(RT~Task*Condition+(1+Condition|Subject)+(1|ID),data=Exp2_SCJT_ANOVA_MEM_final)
summary(m11_lme)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Task * Condition + (1 + Condition | Subject) + (1 | ID)
## Data: Exp2_SCJT_ANOVA_MEM_final
##
## REML criterion at convergence: 38830.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9265 -0.4675 -0.1531 0.2223 20.3959
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 8299 91.10
## Conditionunlearned 2705 52.01 -0.58
## ID (Intercept) 2691 51.87
## Residual 35367 188.06
## Number of obs: 2907, groups: Subject, 28; ID, 22
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 709.685 23.504 64.536 30.195 < 2e-16 ***
## Task -20.413 3.480 2830.051 -5.866 4.97e-09 ***
## Conditionunlearned -41.548 19.068 149.021 -2.179 0.0309 *
## Task:Conditionunlearned 6.535 4.924 2830.137 1.327 0.1846
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Task Cndtnn
## Task -0.444
## Condtnnlrnd -0.516 0.548
## Tsk:Cndtnnl 0.314 -0.707 -0.774#只有隨機截距
m12_lme = lmer(RT~Task*Condition+(1|Subject)+(1|ID),data=Exp2_SCJT_ANOVA_MEM_final)
summary(m12_lme)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Task * Condition + (1 | Subject) + (1 | ID)
## Data: Exp2_SCJT_ANOVA_MEM_final
##
## REML criterion at convergence: 38852.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.1615 -0.4801 -0.1529 0.2203 20.5076
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 6356 79.73
## ID (Intercept) 2834 53.24
## Residual 35958 189.63
## Number of obs: 2907, groups: Subject, 28; ID, 22
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 709.731 22.172 78.058 32.010 < 2e-16 ***
## Task -20.395 3.508 2855.025 -5.814 6.79e-09 ***
## Conditionunlearned -40.659 16.471 2855.268 -2.469 0.0136 *
## Task:Conditionunlearned 6.546 4.965 2854.978 1.319 0.1874
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Task Cndtnn
## Task -0.475
## Condtnnlrnd -0.372 0.639
## Tsk:Cndtnnl 0.336 -0.707 -0.903anova(m11_lme,m12_lme)## refitting model(s) with ML (instead of REML)## Data: Exp2_SCJT_ANOVA_MEM_final
## Models:
## m12_lme: RT ~ Task * Condition + (1 | Subject) + (1 | ID)
## m11_lme: RT ~ Task * Condition + (1 + Condition | Subject) + (1 | ID)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## m12_lme 7 38889 38931 -19437 38875
## m11_lme 9 38871 38925 -19427 38853 21.45 2 2.198e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#選擇m11，隨機截距&隨機斜率##Exp2(enemy) Naming_ACC
#隨機截距&隨機斜率
m13_lme = lmer(ACC~Task*Condition+(1+Condition|Subject)+(1|ID),data=Exp2_Naming_ANOVA_MEM_final)
summary(m13_lme)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: ACC ~ Task * Condition + (1 + Condition | Subject) + (1 | ID)
## Data: Exp2_Naming_ANOVA_MEM_final
##
## REML criterion at convergence: -4254.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -8.3418 -0.0537 0.0099 0.0934 2.0492
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 0.0005228 0.02287
## Conditionunlearned 0.0015437 0.03929 -0.61
## ID (Intercept) 0.0016852 0.04105
## Residual 0.0139271 0.11801
## Number of obs: 3080, groups: Subject, 28; ID, 22
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 9.702e-01 1.205e-02 5.966e+01 80.541 <2e-16 ***
## Task 4.221e-03 2.126e-03 3.000e+03 1.985 0.0472 *
## Conditionunlearned -2.262e-03 1.244e-02 1.119e+02 -0.182 0.8561
## Task:Conditionunlearned 9.740e-04 3.007e-03 3.000e+03 0.324 0.7460
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Task Cndtnn
## Task -0.530
## Condtnnlrnd -0.463 0.513
## Tsk:Cndtnnl 0.374 -0.707 -0.725#只有隨機截距
m14_lme = lmer(ACC~Task*Condition+(1|Subject)+(1|ID),data=Exp2_Naming_ANOVA_MEM_final)
summary(m14_lme)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: ACC ~ Task * Condition + (1 | Subject) + (1 | ID)
## Data: Exp2_Naming_ANOVA_MEM_final
##
## REML criterion at convergence: -4212.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -8.1835 -0.0682 0.0051 0.2065 1.7901
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.0002928 0.01711
## ID (Intercept) 0.0017559 0.04190
## Residual 0.0142995 0.11958
## Number of obs: 3080, groups: Subject, 28; ID, 22
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 9.704e-01 1.189e-02 5.731e+01 81.619 <2e-16 ***
## Task 4.221e-03 2.155e-03 3.028e+03 1.959 0.0502 .
## Conditionunlearned -2.228e-03 1.011e-02 3.031e+03 -0.220 0.8256
## Task:Conditionunlearned 9.740e-04 3.047e-03 3.028e+03 0.320 0.7493
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Task Cndtnn
## Task -0.544
## Condtnnlrnd -0.425 0.639
## Tsk:Cndtnnl 0.384 -0.707 -0.904anova(m13_lme,m14_lme)## refitting model(s) with ML (instead of REML)## Data: Exp2_Naming_ANOVA_MEM_final
## Models:
## m14_lme: ACC ~ Task * Condition + (1 | Subject) + (1 | ID)
## m13_lme: ACC ~ Task * Condition + (1 + Condition | Subject) + (1 | ID)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## m14_lme 7 -4236.0 -4193.7 2125.0 -4250.0
## m13_lme 9 -4272.5 -4218.2 2145.3 -4290.5 40.556 2 1.561e-09 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#選擇m13，隨機截距&隨機斜率##Exp2(enemy) Naming_RT
#隨機截距&隨機斜率
m15_lme = lmer(RT~Task*Condition+(1+Condition|Subject)+(1|ID),data=Exp2_Naming_ANOVA_MEM_final)
summary(m15_lme)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Task * Condition + (1 + Condition | Subject) + (1 | ID)
## Data: Exp2_Naming_ANOVA_MEM_final
##
## REML criterion at convergence: 35627.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.0334 -0.4807 -0.0774 0.3755 9.7755
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## Subject (Intercept) 2429.9 49.29
## Conditionunlearned 487.5 22.08 -0.58
## ID (Intercept) 1140.9 33.78
## Residual 10224.3 101.12
## Number of obs: 2940, groups: Subject, 28; ID, 22
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 516.662 13.322 65.430 38.783 < 2e-16 ***
## Task -6.374 1.867 2861.599 -3.414 0.000650 ***
## Conditionunlearned -28.304 9.756 212.439 -2.901 0.004110 **
## Task:Conditionunlearned 8.722 2.637 2863.052 3.307 0.000953 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Task Cndtnn
## Task -0.423
## Condtnnlrnd -0.473 0.578
## Tsk:Cndtnnl 0.300 -0.708 -0.818#只有隨機截距
m16_lme = lmer(RT~Task*Condition+(1|Subject)+(1|ID),data=Exp2_Naming_ANOVA_MEM_final)
summary(m16_lme)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RT ~ Task * Condition + (1 | Subject) + (1 | ID)
## Data: Exp2_Naming_ANOVA_MEM_final
##
## REML criterion at convergence: 35640.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.7373 -0.4925 -0.0754 0.3747 9.8523
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 1869 43.24
## ID (Intercept) 1188 34.47
## Residual 10335 101.66
## Number of obs: 2940, groups: Subject, 28; ID, 22
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 516.278 12.648 74.661 40.818 < 2e-16 ***
## Task -6.374 1.877 2888.028 -3.395 0.000694 ***
## Conditionunlearned -27.975 8.863 2889.137 -3.156 0.001613 **
## Task:Conditionunlearned 8.715 2.651 2887.962 3.287 0.001023 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Task Cndtnn
## Task -0.448
## Condtnnlrnd -0.350 0.639
## Tsk:Cndtnnl 0.317 -0.708 -0.905anova(m15_lme,m16_lme)## refitting model(s) with ML (instead of REML)## Data: Exp2_Naming_ANOVA_MEM_final
## Models:
## m16_lme: RT ~ Task * Condition + (1 | Subject) + (1 | ID)
## m15_lme: RT ~ Task * Condition + (1 + Condition | Subject) + (1 | ID)
## npar AIC BIC logLik deviance Chisq Df Pr(>Chisq)
## m16_lme 7 35672 35714 -17829 35658
## m15_lme 9 35664 35718 -17823 35646 12.053 2 0.002414 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#選擇m15，隨機截距&隨機斜率#分割資料
fixC <- split(Exp2_Naming_ANOVA_MEM_final,Exp2_Naming_ANOVA_MEM_final$Task)
fixD <- split(Exp2_Naming_ANOVA_MEM_final,Exp2_Naming_ANOVA_MEM_final$Condition)#在不同檢測階段的簡單主要效果
fixC1.aov <- aov(RT ~ Condition, data = fixC$`1`)
fixC2.aov <- aov(RT ~ Condition, data = fixC$`2`)
fixC3.aov <- aov(RT ~ Condition, data = fixC$`3`)
fixC4.aov <- aov(RT ~ Condition, data = fixC$`4`)
fixC5.aov <- aov(RT ~ Condition, data = fixC$`5`)summary(fixC1.aov)## Df Sum Sq Mean Sq F value Pr(>F)
## Condition 1 79981 79981 3.659 0.0563 .
## Residuals 573 12523498 21856
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 因為不存在，97 個觀察量被刪除了summary(fixC2.aov)## Df Sum Sq Mean Sq F value Pr(>F)
## Condition 1 36175 36175 2.581 0.109
## Residuals 581 8142628 14015
## 因為不存在，89 個觀察量被刪除了summary(fixC3.aov)## Df Sum Sq Mean Sq F value Pr(>F)
## Condition 1 9649 9649 1.11 0.293
## Residuals 585 5085840 8694
## 因為不存在，85 個觀察量被刪除了summary(fixC4.aov)## Df Sum Sq Mean Sq F value Pr(>F)
## Condition 1 12687 12687 1.286 0.257
## Residuals 590 5818679 9862
## 因為不存在，80 個觀察量被刪除了summary(fixC5.aov)## Df Sum Sq Mean Sq F value Pr(>F)
## Condition 1 13847 13847 1.277 0.259
## Residuals 601 6515881 10842
## 因為不存在，69 個觀察量被刪除了#在不同同音鄰群類型的簡單主要效果
fixD1.aov <- aov(RT ~ Task, data = fixD$`learned`)
fixD2.aov <- aov(RT ~ Task, data = fixD$`unlearned`)summary(fixD1.aov)## Df Sum Sq Mean Sq F value Pr(>F)
## Task 1 111023 111023 7.944 0.00489 **
## Residuals 1473 20585701 13975
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 因為不存在，205 個觀察量被刪除了summary(fixD2.aov)## Df Sum Sq Mean Sq F value Pr(>F)
## Task 1 25668 25668 2.07 0.15
## Residuals 1463 18138210 12398
## 因為不存在，215 個觀察量被刪除了