Self-paced reading experiments on explicit and implicit contrastive and temporal discourse relations in Czech

Zikánová, Šárka

Self-paced reading experiments on explicit and implicit contrastive and temporal discourse relations in Czech

Please use the following text to cite this item or export to a predefined format:

Zikánová, Šárka and Smolík, Filip, 2022, Self-paced reading experiments on explicit and implicit contrastive and temporal discourse relations in Czech, LINDAT/CLARIAH-CZ digital library at the Institute of Formal and Applied Linguistics (ÚFAL), http://hdl.handle.net/11234/1-5006.

Share

Authors

Zikánová, Šárka and Smolík, Filip

Item identifier

http://hdl.handle.net/11234/1-5006

Project URL

https://ufal.mff.cuni.cz/grants/lusyd

Date issued

2022-12-16

Type

languageDescription,

text

Language(s)

Czech

Description

Supplementary materials for the paper “Processing of explicit and implicit contrastive and temporal discourse relations in Czech” (submitted to Discourse Processes)

Publisher

Charles University, Faculty of Mathematics and Physics, Institute of Formal and Applied Linguistics (UFAL)

Acknowledgement

Subject(s)

discourse

psycholinguistic experiments

explicit discourse relations

implicit discourse relations

self-paced reading

Collections

LINDAT / CLARIAH-CZ Data & Tools

Show full item record

This item isPublicly Available

and licensed under:

Creative Commons - Attribution 4.0 International (CC BY 4.0)

Files in this item

Download instructions for command line Download all files in item (13.04 MB)

Name: E2pub.html
Size: 690.65 KB
Format: text/html
Description: Supplementary materials for the paper “Processing of explicit and implicit contrastive and temporal discourse relations in Czech”: code for analyses and plots for Experiment 2
MD5: 54e66d1a2b8b8d2648d5d92b80f06e3a

Download file Preview

File Preview

Experiment 2 combined

library(bestNormalize)
library(sjPlot)
library(lmerTest)
library(ggplot2)
 
Sys.setlocale("LC_ALL", "Czech")
options(encoding = "UTF-8")

aqsync<-read.csv(file="aqsync.csv",fileEncoding = "UTF-16LE")

aqsync$cleanrt<-aqsync$rt
aqsync$cleanrt[aqsync$rt>4000|aqsync$rt<100]<-NA
aqsync$normclrt<-bestNormalize(aqsync$cleanrt,allow_orderNorm=F)$x.t

aqprec<-read.csv(file="aqprec.csv",fileEncoding = "UTF-16LE")

aqprec$cleanrt<-aqprec$rt
aqprec$cleanrt[aqprec$rt>4000|aqprec$rt<100]<-NA
aqprec$normclrt<-bestNormalize(aqprec$cleanrt,allow_orderNorm=F)$x.t

Auxiliary function

The following function calculates cell means and standard errors to be used in plots.

summarySE<- function(data=NULL, measurevar, groupvars=NULL, na.rm=FALSE,
                      conf.interval=.95, .drop=TRUE) {
    library(plyr)

    # New version of length which can handle NA's: if na.rm==T, don't count them
    length2 <- function (x, na.rm=FALSE) {
        if (na.rm) sum(!is.na(x))
        else       length(x)
    }

    # This does the summary. For each group's data frame, return a vector with
    # N, mean, and sd
    datac <- ddply(data, groupvars, .drop=.drop,
      .fun = function(xx, col) {
        c(N    = length2(xx[[col]], na.rm=na.rm),
          mean = mean   (xx[[col]], na.rm=na.rm),
          sd   = sd     (xx[[col]], na.rm=na.rm)
        )
      },
      measurevar
    )

    # Rename the "mean" column    
    datac <- rename(datac, c("mean" = measurevar))

    datac$se <- datac$sd / sqrt(datac$N)  # Calculate standard error of the mean

    # Confidence interval multiplier for standard error
    # Calculate t-statistic for confidence interval: 
    # e.g., if conf.interval is .95, use .975 (above/below), and use df=N-1
    ciMult <- qt(conf.interval/2 + .5, datac$N-1)
    datac$ci <- datac$se * ciMult

    return(datac)
}

pd <- position_dodge(0.1)

Synchrony plot

aqsyncsum<-summarySE(aqsync[aqsync$region1%in%c(6:9)&aqsync$subj%in%factor(c(1:30))&aqsync$is_correct==1&!is.na(aqsync$cleanrt),], measurevar="cleanrt", groupvars=c("condition","region1"))
outplot<-ggplot(aqsyncsum, aes(x=factor(region1), y=cleanrt, group=factor(condition),linetype=factor(condition))) + 
    geom_errorbar(aes(ymin=cleanrt-se, ymax=cleanrt+se), width=.1,position=pd) +
    geom_line(position=pd) +
    geom_point(position=pd) +
    scale_y_continuous("Reading time (ms)")+
    scale_x_discrete("Region",breaks=c("6", "7","8", "9"),
        labels=c("\nMezitím\nMeanwhile", "Z rozhlasu\nz rozhlasu\nfrom radio","vyhrávala\nvyhrávala\nplayed", "hudba\nhudba\nmusic"))+
    scale_linetype_manual(name="", 
                         labels = c("Explicit", 
                                   "Implicit"),
                                    values = c("0"="solid",
                                    "1"="dashed"))+
    theme_bw()

outplot

Synchrony modeling

aqsync7<-aqsync[aqsync$region1%in%c(7),]
aqsync8<-aqsync[aqsync$region1%in%c(8),]
aqsync9<-aqsync[aqsync$region1%in%c(9),]

x<-lmer(normclrt~condition+(1|subjv)+(1|item),data=aqsync7)
xa<-lmer(normclrt~condition+(1|subjv)+(1|item),data=aqsync8)
xb<-lmer(normclrt~condition+(1|subjv)+(1|item),data=aqsync9)
tab_model(x,xa,xb,show.std=T,show.est=F,show.ci=F)

	normclrt		normclrt		normclrt
Predictors	std. Beta	p	std. Beta	p	std. Beta	p
(Intercept)	0.01	<0.001	0.01	<0.001	-0.01	0.343
condition	0.13	<0.001	-0.06	<0.001	-0.02	0.235
Random Effects
σ²	0.31		0.25		0.35
τ₀₀	0.41 _subjv		0.39 _subjv		0.62 _subjv
	0.02 _item		0.01 _item		0.04 _item
ICC	0.58		0.61		0.65
N	109 _subjv		109 _subjv		109 _subjv
	20 _item		20 _item		20 _item
Observations	1902		1901		1881
Marginal R² / Conditional R²	0.017 / 0.584		0.004 / 0.616		0.000 / 0.650

Precedence plot

aqprecsum<-summarySE(aqprec[aqprec$region1%in%c(6:9)&aqprec$subj%in%factor(c(1:30))&aqprec$is_correct==1&!is.na(aqprec$cleanrt),], measurevar="cleanrt", groupvars=c("condition","region1"))
outplot<-ggplot(aqprecsum, aes(x=factor(region1), y=cleanrt, group=factor(condition),linetype=factor(condition))) + 
    geom_errorbar(aes(ymin=cleanrt-se, ymax=cleanrt+se), width=.1,position=pd) +
    geom_line(position=pd) +
    geom_point(position=pd) +
    scale_y_continuous("Reading time (ms)")+
    scale_x_discrete("Region",breaks=c("6", "7","8", "9"),
        labels=c("\nPotom\nThen", "Do třídy\ndo třídy\nin classroom","přišel\npřišel\ncame", "ředitel\nředitel\ndirector"))+
    scale_linetype_manual(name="", 
                         labels = c("Explicit", 
                                   "Implicit"),
                                    values = c("0"="solid",
                                    "1"="dashed"))+
    theme_bw()

outplot

# Precedence modeling

aqprec7<-aqprec[aqprec$region1%in%c(7),]
aqprec8<-aqprec[aqprec$region1%in%c(8),]
aqprec9<-aqprec[aqprec$region1%in%c(9),]

x<-lmer(normclrt~condition+(1|subjv)+(1|item),data=aqprec7)
xa<-lmer(normclrt~condition+(1|subjv)+(1|item),data=aqprec8)
xb<-lmer(normclrt~condition+(1|subjv)+(1|item),data=aqprec9)
tab_model(x,xa,xb,show.std=T,show.est=F,show.ci=F)

	normclrt		normclrt		normclrt
Predictors	std. Beta	p	std. Beta	p	std. Beta	p
(Intercept)	0.01	<0.001	0.01	<0.001	0.00	0.528
condition	0.14	<0.001	-0.07	<0.001	-0.00	0.912
Random Effects
σ²	0.28		0.24		0.23
τ₀₀	0.60 _subjv		0.52 _subjv		0.67 _subjv
	0.01 _item		0.00 _item		0.01 _item
ICC	0.68		0.69		0.75
N	109 _subjv		109 _subjv		108 _subjv
	20 _item		20 _item		20 _item
Observations	1950		1954		1927
Marginal R² / Conditional R²	0.021 / 0.691		0.005 / 0.690		0.000 / 0.747

Name: aqconf.csv
Size: 3.02 MB
Format: application/octet-stream
Description: Supplementary materials for the paper “Processing of explicit and implicit contrastive and temporal discourse relations in Czech”: the data file for the confrontation relation
MD5: 41a2f497778c155edcf2d6c4f4c181c8

Download file

File Preview

The file preview has not been generated yet. Please try again later or contact the system administrator lindat-help@ufal.mff.cuni.cz

Name: E1pub.html
Size: 697.84 KB
Format: text/html
Description: Supplementary materials for the paper “Processing of explicit and implicit contrastive and temporal discourse relations in Czech”: code for analyses and plots for Experiment 1
MD5: 0831bdf0931b25ae6f3dd0d02cedfcfe

Download file Preview

File Preview

Experiment 1 combined

Sys.setlocale("LC_ALL", "Czech")
options(encoding = "UTF-8")

library(bestNormalize)
library(sjPlot)
library(lmerTest)
library(ggplot2)

Load concession data, trim the response times 200 ms and below and 4000 and above, normalize the response times.

aqconc<-read.csv(file="aqconc.csv",fileEncoding = "UTF-16LE")

aqconc$cleanrt<-aqconc$rt
aqconc$cleanrt[aqconc$rt>4000|aqconc$rt<200]<-NA
aqconc$normclrt<-bestNormalize(aqconc$cleanrt,allow_orderNorm=F)$x.t

Load confrontation data, trim the response times 200 ms and below and 4000 and above, normalize the response times.

aqconf<-read.csv(file="aqconf.csv",fileEncoding = "UTF-16LE")

aqconf$cleanrt<-aqconf$rt
aqconf$cleanrt[aqconf$rt>4000|aqconf$rt<200]<-NA
aqconf$normclrt<-bestNormalize(aqconf$cleanrt,allow_orderNorm=F)$x.t

Auxiliary function

The following function calculates cell means and standard errors to be used in plots.

summarySE<- function(data=NULL, measurevar, groupvars=NULL, na.rm=FALSE,
                      conf.interval=.95, .drop=TRUE) {
    library(plyr)

    # New version of length which can handle NA's: if na.rm==T, don't count them
    length2 <- function (x, na.rm=FALSE) {
        if (na.rm) sum(!is.na(x))
        else       length(x)
    }

    # This does the summary. For each group's data frame, return a vector with
    # N, mean, and sd
    datac <- ddply(data, groupvars, .drop=.drop,
      .fun = function(xx, col) {
        c(N    = length2(xx[[col]], na.rm=na.rm),
          mean = mean   (xx[[col]], na.rm=na.rm),
          sd   = sd     (xx[[col]], na.rm=na.rm)
        )
      },
      measurevar
    )

    # Rename the "mean" column    
    datac <- rename(datac, c("mean" = measurevar))

    datac$se <- datac$sd / sqrt(datac$N)  # Calculate standard error of the mean

    # Confidence interval multiplier for standard error
    # Calculate t-statistic for confidence interval: 
    # e.g., if conf.interval is .95, use .975 (above/below), and use df=N-1
    ciMult <- qt(conf.interval/2 + .5, datac$N-1)
    datac$ci <- datac$se * ciMult

    return(datac)
}

pd <- position_dodge(0.1)

Concession plot

aqconcsum<-summarySE(aqconc[aqconc$region1%in%c(6:10)&!is.na(aqconc$cleanrt),], measurevar="cleanrt", groupvars=c("condition","region1"))
outplot<-ggplot(aqconcsum, aes(x=factor(region1), y=cleanrt, group=factor(condition),linetype=factor(condition))) + 
    geom_errorbar(aes(ymin=cleanrt-se, ymax=cleanrt+se), width=.1,position=pd) +
    geom_line(position=pd) +
    geom_point(position=pd) +
    scale_x_discrete("Region",breaks=c("6", "7","8", "9", "10"),
        labels=c("\nPřesto\nYet", "Dál\ndál\nfurther","kupoval\nkupoval\nhe bought", "drahé\ndrahé\nexpensive","zbytečnosti\nzbytečnosti\nfutilities"))+
    theme_bw()+
    scale_linetype_manual(name="", 
                         labels = c("Explicit", 
                                   "Implicit"),
                                    values = c("0"="solid",
                                    "1"="dashed"))+
    labs(y="Reading time (ms)",x="")

outplot

Concession modeling

The following code shows the results of mixed models:

aqconc7<-subset(aqconc, region1=="7")
aqconc8<-subset(aqconc, region1=="8")
aqconc9<-subset(aqconc, region1=="9")
aqconc10<-subset(aqconc, region1=="10")

 xa<-lmer(normclrt~condition+(condition|subjv)+(1|item),data=aqconc7)
 xb<-lmer(normclrt~condition+(condition|subjv)+(1|item),data=aqconc8)
 xc<-lmer(normclrt~condition+(condition|subjv)+(1|item),data=aqconc9)
 xd<-lmer(normclrt~condition+(condition|subjv)+(1|item),data=aqconc10)
tab_model(xa,xb,xc,xd,show.std=T,show.ci=F,show.est=F)

	normclrt		normclrt		normclrt		normclrt
Predictors	std. Beta	p	std. Beta	p	std. Beta	p	std. Beta	p
(Intercept)	-0.02	<0.001	-0.02	<0.001	-0.04	<0.001	-0.02	0.430
condition	0.21	<0.001	-0.01	0.624	0.02	0.225	0.04	0.008
Random Effects
σ²	0.33		0.32		0.28		0.34
τ₀₀	0.38 _subjv		0.35 _subjv		0.36 _subjv		0.48 _subjv
	0.02 _item		0.02 _item		0.06 _item		0.03 _item
τ₁₁	0.04 _{subjv.condition}		0.02 _{subjv.condition}		0.02 _{subjv.condition}		0.01 _{subjv.condition}
ρ₀₁	-0.36 _subjv		0.09 _subjv		-0.37 _subjv		-0.26 _subjv
ICC	0.54		0.55		0.59		0.59
N	115 _subjv		115 _subjv		115 _subjv		115 _subjv
	20 _item		20 _item		20 _item		20 _item
Observations	2024		2028		2026		2010
Marginal R² / Conditional R²	0.043 / 0.556		0.000 / 0.547		0.000 / 0.589		0.002 / 0.591

Confrontation plot

aqconfsum<-summarySE(aqconf[aqconf$region1%in%c(6:10)&!is.na(aqconf$cleanrt),], measurevar="cleanrt", groupvars=c("condition","region1"))

outplot<-ggplot(aqconfsum, aes(x=factor(region1), y=cleanrt, group=factor(condition),linetype=factor(condition))) + 
    geom_errorbar(aes(ymin=cleanrt-se, ymax=cleanrt+se), width=.1,position=pd) +
    geom_line(position=pd) +
    geom_point(position=pd) +
    scale_x_discrete("Region",breaks=c("6", "7","8", "9", "10"),
      labels=c("\nZato\nBut", "V Praze\nv Praze\nin Prague","trvale\ntrvale\nsteadily", "mírně\nmírně\nslightly","stoupá\nstoupá\ngrows"))+
    theme_bw()+
    scale_linetype_manual(name="", 
                         labels = c("Explicit", 
                                   "Implicit"),
                                    values = c("0"="solid",
                                    "1"="dashed"))+
    labs(y="Reading time (ms)", x="")
    
outplot

Confrontation modeling

aqconf7<-subset(aqconf, region1=="7")
aqconf8<-subset(aqconf, region1=="8") 
aqconf9<-subset(aqconf, region1=="9") 
aqconf10<-subset(aqconf, region1=="10")
 
 xa<-lmer(normclrt~factor(condition)+(condition|subjv)+(condition|item),data=aqconf7)
 xb<-lmer(normclrt~factor(condition)+(condition|subjv)+(condition|item),data=aqconf8)
 xc<-lmer(normclrt~factor(condition)+(condition|subjv)+(condition|item),data=aqconf9)
 xd<-lmer(normclrt~factor(condition)+(1|subjv)+(condition|item),data=aqconf10)
tab_model(xa,xb,xc,xd,show.std=T,show.est=F,show.ci=F)

	normclrt		normclrt			normclrt		normclrt
Predictors	std. Beta	p	std. Beta	p	std. p	std. Beta	p	std. Beta	p
(Intercept)	-0.23	<0.001	0.04	<0.001	0.680	-0.01	<0.001	-0.07	0.015
condition [1]	0.43	<0.001	-0.11	0.006	0.006	-0.02	0.630	0.10	0.003
Random Effects
σ²	0.33		0.33			0.30		0.41
τ₀₀	0.40 _subjv		0.32 _subjv			0.28 _subjv		0.47 _subjv
	0.04 _item		0.05 _item			0.03 _item		0.03 _item
τ₁₁	0.05 _{subjv.condition}		0.01 _{subjv.condition}			0.02 _{subjv.condition}		0.00 _{item.condition}
	0.01 _{item.condition}		0.01 _{item.condition}			0.01 _{item.condition}
ρ₀₁	-0.48 _subjv		0.09 _subjv			-0.05 _subjv		0.03 _item
	-0.51 _item		-0.52 _item			-0.26 _item
ICC	0.57		0.53			0.51		0.55
N	115 _subjv		115 _subjv			115 _subjv		115 _subjv
	20 _item		20 _item			20 _item		20 _item
Observations	2122		2118			2120		2104
Marginal R² / Conditional R²	0.043 / 0.587		0.003 / 0.532			0.000 / 0.513		0.002 / 0.554

Name: aqconc.csv
Size: 2.81 MB
Format: application/octet-stream
Description: Supplementary materials for the paper “Processing of explicit and implicit contrastive and temporal discourse relations in Czech”: the data file for the concession relation
MD5: 765de88a5c26a2b748f78ebf63cde2ff

Download file

File Preview

The file preview has not been generated yet. Please try again later or contact the system administrator lindat-help@ufal.mff.cuni.cz

Name: aqsync.csv
Size: 2.38 MB
Format: application/octet-stream
Description: Supplementary materials for the paper “Processing of explicit and implicit contrastive and temporal discourse relations in Czech”: the data file for the synchrony relation
MD5: bb5f902f3897a855e92d6b5cce59c35d

Download file

File Preview

The file preview has not been generated yet. Please try again later or contact the system administrator lindat-help@ufal.mff.cuni.cz

Name: aqprec.csv
Size: 3.47 MB
Format: application/octet-stream
Description: Supplementary materials for the paper “Processing of explicit and implicit contrastive and temporal discourse relations in Czech”: the data file for the asynchrony relation
MD5: 9338ba45e93c3e84e0711cdde7590690

Download file

File Preview

The file preview has not been generated yet. Please try again later or contact the system administrator lindat-help@ufal.mff.cuni.cz

Name: experimental_stimuli.pdf
Size: 673.69 KB
Format: application/pdf
Description: Supplementary materials for the paper “Processing of explicit and implicit contrastive and temporal discourse relations in Czech”: experimental stimuli
MD5: 990f8b384e5e107acbb1de3ff988e324

Download file

File Preview

The file preview has not been generated yet. Please try again later or contact the system administrator lindat-help@ufal.mff.cuni.cz