Analysis E1A Inner Speech Talk and Type

Matt

5/14/2019

Load Libraries

library(papaja)
library(TalkTyping)
library(dplyr)
library(Crump)
library(xtable)
library(ggplot2)
library(ggpubr)

E1A IKSI Analysis

#load E1A data
E1_data <- talk_type_E1A_data

# IKSI analysis

E1_data <- E1_data %>%
            mutate(subject = as.factor(subject)) %>%
            filter(errors == 1,
                   iksis < 5000,
                   LetterType != "Space") %>%
            group_by(subject,linguistic_unit,LetterType) %>%
            summarise(mean_iksi = mean(modified_recursive_moving(iksis)$restricted),
                      prop_removed = modified_recursive_moving(iksis)$prop_removed)
                  
E1_aov_out <- aov(mean_iksi ~ linguistic_unit*LetterType + 
                    Error(subject/(linguistic_unit*LetterType)), E1_data)

knitr::kable(xtable(summary(E1_aov_out)))
Df Sum Sq Mean Sq F value Pr(>F)
Residuals 39 1474756.256 37814.2630 NA NA
linguistic_unit 1 76739.011 76739.0113 24.891295 0.0000130
Residuals 39 120235.666 3082.9658 NA NA
LetterType 1 151196.259 151196.2589 22.284563 0.0000301
Residuals 39 264607.123 6784.7980 NA NA
linguistic_unit:LetterType 1 1547.208 1547.2079 3.083524 0.0869404
Residuals 39 19568.881 501.7662 NA NA 
E1_apa_print <- apa_print(E1_aov_out)
E1_means <- model.tables(E1_aov_out,"means")
E1A_iksi_table <- E1_data %>%
              group_by(linguistic_unit,LetterType) %>%
              summarize(mIKSI = mean(mean_iksi),
                        sem = sd(mean_iksi)/sqrt(length(mean_iksi)))

levels(E1A_iksi_table$linguistic_unit) <- c("Say Letter", "Say Word")

E1A_graph_iksi <- ggplot(E1A_iksi_table, aes(x=linguistic_unit,
                                             y=mIKSI, 
                                             group=LetterType,
                                             fill=LetterType))+
  geom_bar(stat="identity",position="dodge")+
  geom_errorbar(aes(ymin=mIKSI-sem,
                    ymax=mIKSI+sem), width=.1,
                linetype="solid", position=position_dodge(.9))+
  scale_fill_grey(start = 0.6, end = 0.8, na.value = "red",
  aesthetics = "fill")+
  theme_classic(base_size=12)+
  theme(legend.position = "top",
        legend.title = element_blank())+
  ylab("Mean IKSI (ms)")+
  xlab("Linguistic Unit")

knitr::kable(E1A_iksi_table)
linguistic_unit LetterType mIKSI sem
Say Letter First 247.9339 22.368227
Say Letter Middle 180.2336 11.767037
Say Word First 197.9142 21.590340
Say Word Middle 142.6526 9.982549

accuracy

# Accuracy

E1acc_data <- talk_type_E1A_data


E1acc_data <- E1acc_data %>%
                mutate(subject = as.factor(subject)) %>%
                filter(LetterType != "Space") %>%
                group_by(subject, linguistic_unit, LetterType) %>%
                summarise(mean_acc = mean(errors))
                  
E1acc_aov_out <- aov(mean_acc ~ linguistic_unit*LetterType + Error(subject/(linguistic_unit*LetterType)), E1acc_data)

E1acc_apa_print <- apa_print(E1acc_aov_out)
E1acc_means <- model.tables(E1acc_aov_out,"means")

knitr::kable(xtable(summary(E1acc_aov_out)))
Df Sum Sq Mean Sq F value Pr(>F)
Residuals 39 0.4066274 0.0104263 NA NA
linguistic_unit 1 0.0039057 0.0039057 0.6985677 0.4083570
Residuals 39 0.2180470 0.0055909 NA NA
LetterType 1 0.0008383 0.0008383 1.0114875 0.3207462
Residuals 39 0.0323229 0.0008288 NA NA
linguistic_unit:LetterType 1 0.0000369 0.0000369 0.0807627 0.7777700
Residuals 39 0.0178246 0.0004570 NA NA 
E1A_acc_table <- E1acc_data %>%
              group_by(linguistic_unit,LetterType) %>%
              summarize(mAcc = mean(mean_acc),
                        sem = sd(mean_acc)/sqrt(length(mean_acc)))

levels(E1A_acc_table$linguistic_unit) <- c("Say Letter", "Say Word")

E1A_graph_acc <- ggplot(E1A_acc_table, 
                         aes(x=linguistic_unit,
                             y=mAcc, 
                             group=LetterType,
                             fill=LetterType))+
  geom_bar(stat="identity",position="dodge")+
  geom_errorbar(aes(ymin=mAcc-sem,
                    ymax=mAcc+sem), width=.1,
                linetype="solid", position=position_dodge(.9))+
  scale_fill_grey(start = 0.6, end = 0.8, na.value = "red",
  aesthetics = "fill")+
  theme_classic(base_size=12)+
  theme(legend.position = "top",
        legend.title = element_blank())+
  ylab("Mean Accuracy")+
  xlab("Linguistic Unit")+
  coord_cartesian(ylim=c(.8,1))

knitr::kable(E1A_acc_table)
linguistic_unit LetterType mAcc sem
Say Letter First 0.9587045 0.0116924
Say Letter Middle 0.9550872 0.0125089
Say Word First 0.9497838 0.0086318
Say Word Middle 0.9442452 0.0080552

Figure 1

ggarrange(E1A_graph_iksi,E1A_graph_acc)

Demographic information

E1A_dmg <- talk_type_E1A_dmg
proportion_word <- E1A_dmg %>%
                    group_by(InnerVoice) %>%
                    summarize(p_word = length(InnerVoice)/dim(E1A_dmg)[1])

inner_voice_props<-c(
  round(proportion_word[proportion_word$InnerVoice=="Words",]$p_word, digits=2),
round(proportion_word[proportion_word$InnerVoice=="Letters",]$p_word,digits=2),
round(proportion_word[proportion_word$InnerVoice=="undefined",]$p_word,digits=2))

Save all

save.image(file="E1A_workspace.RData")