Random auditory stimulation during sleep disturbs traveling slow waves and declarative memory.

Code ↔ Paper

20 matches between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 20 matches

[1] § STAR★Methods › Method details › Task material › Hearing test and auditory stimulation ↔ analysis/ana27_hearingthreshold_calculation.m, lines 1–26 · score 0.89 · supra aural, biological norm, insert earphones, hearing threshold, ISO, physical
[2] § STAR★Methods › Quantification and statistical analysis › Linear mixed-effect modeling and mediation analysis ↔ analysis/ana25_correlations_behaviour_PCA_LMM_mediation.R, lines 1064–1111 · score 0.86 · linear mixed, full model, lme4, PCA, ANOVA, mediation
[3] § STAR★Methods › Quantification and statistical analysis › Spindle peak-frequency ↔ analysis/ana18_spindle_frequency.m, lines 83–114 · score 0.77 · power spectral density, log transformed, pwelch, PSD, 45 Hz, spindle
[4] § STAR★Methods › Quantification and statistical analysis › Spindle and slow wave detection ↔ analysis/detectSpindles.m, lines 1–29 · score 0.68 · N1 sleep, amplitude criterion, envelope, sleep stages, events, spindle
[5] § STAR★Methods › Quantification and statistical analysis › Spindle and slow wave detection ↔ analysis/ana11_spindle_SO_detection.m, lines 18–35 · score 0.67 · bandpass filtered, 11–17 Hz, Detection, concatenated, spindle, 11 Hz
[6] § Results › Stimulation decreased physiological markers of slow waves but not sleep spindles ↔ analysis/ana09_frequency_statistics.m, lines 36–94 · score 0.65 · 12–16 Hz, 0.5–30 Hz, 0.5–4 Hz, delta, spindles, Cluster
[7] § Results › Stimulation triggered a broad cortical response with a lasting low-frequency reduction ↔ analysis/ana22_ERP_stats.m, lines 626–689 · score 0.61 · ERPs, onset, REM, 30 Hz, Cluster, channels
[8] § STAR★Methods › Method details › Task material › Hearing test and auditory stimulation ↔ analysis/ana27_hearingthreshold_calculation.m, lines 1–26 · score 0.60 · insert earphones, Hearing threshold, volume, stimulation
[9] § Results › Traveling slow wave dynamics drive the sleep effect on memory › N2 duration ↔ analysis/ana25_correlations_behaviour_PCA_LMM_mediation.R, lines 1064–1111 · score 0.59 · linear mixed, N2 sleep, VIF, mediation, fit, predictor
[10] § Results › Stimulation reduced slow-wave sleep, but left sleep duration and general alertness mostly intact ↔ analysis/ana10_sleep_architecture.m, lines 26–125 · score 0.58 · sleep latency, falling asleep, awake, N1, REM, N2
[11] § Results › Traveling slow wave dynamics drive the sleep effect on memory ↔ analysis/ana25_correlations_behaviour_PCA_LMM_mediation.R, lines 897–936 · score 0.58 · principal component, linear mixed, mediation, correlated, models, distance
[12] § Results › Stimulation decreased physiological markers of slow waves but not sleep spindles ↔ analysis/ana18_spindle_frequency.m, lines 188–237 · score 0.56 · fast spindles, slow spindles, spindle frequency, amplitude, density, peak
[13] § STAR★Methods › Method details › Task material › Learning-and-memory-task (Lern-und-Gedächtnistest, LGT-3) ↔ analysis/ana24_behavioural_mean.m, lines 270–273 · score 0.56 · verbal score, figural score, DMS, LGT
[14] § Results › Traveling slow wave dynamics drive the sleep effect on memory › SWS duration ↔ analysis/ana16_TSW_analysis_stats_SWS.R, lines 49–133 · score 0.55 · maximal distance, TSW features, peak duration, electrodes, SE, SWS
[15] § STAR★Methods › Quantification and statistical analysis › Spindle and slow wave detection ↔ analysis/coupling_analysis/prep1.m, lines 43–102 · score 0.53 · Spindle events, cycle, SOs, inflection, troughs, coupling
[16] § STAR★Methods › Method details › Task material › Electrophysiology recordings and preprocessing ↔ analysis/signal_processing.m, lines 67–129 · score 0.53 · notch, detrending, demeaning, preprocessing, filtering, derivations
[17] § STAR★Methods › Method details › Task material › Finger tapping ↔ analysis/ana23_behavioural_readin.m, lines 138–265 · score 0.52 · finger tapping, accuracy, block, sequence
[18] § STAR★Methods › Quantification and statistical analysis › Wave parameters and statistical analysis ↔ analysis/ana17_TSW_electrodes.m, lines 78–95 · score 0.51 · permuted, diagonal, shuffling, heatmap, matrices, vector
[19] § Results › Stimulation decreased physiological markers of slow waves but not sleep spindles ↔ analysis/coupling_analysis/statistics_v4.m, lines 1–36 · score 0.51 · coupled spindle, auditory stimulation altered, Slow wave, phase, channels, sham
[20] § Results › Stimulation altered the nature and trajectory of traveling slow waves ↔ analysis/ana15_travelSOs.m, lines 242–256 · score 0.51 · maximal distance covered, peak duration, speed, travel, trajectory, TSW

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The paper is loaded by your browser, with JavaScript: read it at doi.org or on Europe PMC.

The authors' code

R · 1,327 lines · 69 KB · no license · 3 matches

rm(list = ls())
# correlate behavioural outputs with sleep architecture, SO features, and TSW
MAINPATH <- ("D:/Nora/SPIN/data/")
PATH_IN_behavioural <- "D:/Nora/SPIN/results/behavioural/"
PATH_IN_behavioural <- "D:/Nora/SPIN/results_clean/behavioural/"
PATH_IN_architecture <- "D:/Nora/SPIN/results_clean/sleep_architecture"
PATH_IN_SO <- "D:/Nora/SPIN/results_clean/so_features/"
PATH_IN_TSW <- "D:/Nora/SPIN/results_clean/traveling_waves/"
PATH_IN_coupling <- "D:/Nora/SPIN/results_clean/coupling/"
PATH_OUT <- ("D:/Nora/SPIN/results_clean/correlation/")
# Set cd
setwd(MAINPATH)
# initiate libraries
library(tidyr)
library(dplyr)
library(rstatix)
library(plotrix)
library(ggpubr)
library(viridis)
library(svglite)
#install.packages('rmcorr')
library(rmcorr)
library(ggplot2)
library(car)
### define colors according to 'coolwarm' Matlab
color_sham <- rgb(0.1276, 0.3162, 0.8584) # old: "#0085BCFF"
color_stim <- rgb(0.7524, 0.0084, 0.0218) # old: '#FF9E61FF'
axes_color <- rgb(0.1500, 0.1500, 0.1500) # to match matlab plots
axisLineWidth <- 0.5 # to match matlab plots
#### Behavioural: Data read in & transformation ####
# Set cd
setwd(MAINPATH)
# read in data behavioural
subject_notes <- read.csv("subject_notes.csv")
subject_notes_20 <- subject_notes[c(1:10, 12:13, 15:19, 22:23, 25),]
colnames(subject_notes_20) <- as.factor(c('ID', 'cond_time', 'sex', 'age'))
# Set cd
setwd(PATH_IN_behavioural)
# LGT
LGT_categories_stim_20 <- read.csv("LGT_data_categories_stim_20.csv")
LGT_categories_stim_20 <- cbind(subject_notes_20,LGT_categories_stim_20)
LGT_tests_stim_20 <- read.csv("LGT_data_tests_stim_20.csv")
LGT_tests_stim_20 <- cbind(subject_notes_20,LGT_tests_stim_20)
colnames(LGT_tests_stim_20)[6] <- 'stim_Turk'
colnames(LGT_tests_stim_20)[12] <- 'sham_Turk'
LGT_categories_stim_20 <- LGT_categories_stim_20 %>%
gather(conditions, measure, stim_LGS:sham_VG) %>%
separate(conditions, c('stim', 'category')) %>%
arrange(ID, stim)
LGT_tests_stim_20 <- LGT_tests_stim_20 %>%
gather(conditions, measure, stim_Stadt:sham_Zeich) %>%
separate(conditions, c('stim', 'test')) %>%
arrange(ID, stim)
#### end ####
#### Sleep architecture: data read in & transformation ####
# Set cd
setwd(PATH_IN_architecture)
# sorted by stimulation
sleep_stim_20 <- read.csv("sleep_architecture_stim_20.csv")
sleep_stim_20 <- cbind(subject_notes_20,sleep_stim_20)
abs_sleep_stim_20 <- sleep_stim_20[,c(1:14, 23:32)]
colnames(abs_sleep_stim_20) <- c('ID', 'cond_time', 'sex', 'age','stim_duration', 'stim_sleep', 'stim_awake', 'stim_pure','stim_N1', 'stim_N2', 'stim_N3', 'stim_N4', 'stim_SWS', 'stim_REM','sham_duration', 'sham_sleep', 'sham_awake', 'sham_pure','sham_N1', 'sham_N2', 'sham_N3', 'sham_N4', 'sham_SWS', 'sham_REM')
rel_sleep_stim_20 <- sleep_stim_20[,c(1:4, 15:22, 33:40)]
colnames(rel_sleep_stim_20) <- c('ID', 'cond_time', 'sex', 'age','stim_sleep','stim_N1', 'stim_N2', 'stim_N3', 'stim_N4', 'stim_SWS', 'stim_REM','stim_sleeponset','sham_sleep','sham_N1', 'sham_N2', 'sham_N3', 'sham_N4', 'sham_SWS', 'sham_REM','sham_sleeponset')
# transform to long format
abs_sleep_stim_20 <- abs_sleep_stim_20 %>%
gather(conditions, duration, stim_duration:sham_REM) %>%
separate(conditions, c('stim', 'type')) %>%
arrange(ID, stim)
rel_sleep_stim_20 <- rel_sleep_stim_20 %>%
gather(conditions, proportion, stim_sleep:sham_sleeponset) %>%
separate(conditions, c('stim', 'type')) %>%
arrange(ID, stim)
#### end ####
#### SO count: Data read in & transformation ####
# Set cd
setwd(PATH_IN_SO)
sleep_so_20 <- read.csv("spin_sleep_so.csv")
sleep_so_20$id <- as.factor(sleep_so_20$id)
# transform to long format
sleep_so_20 <- sleep_so_20 %>%
gather(conditions, output, stim_counttotal:sham_ampNREM) %>%
separate(conditions, c('stim','type')) %>%
arrange(id, stim)
#### end ####
#### TSW: Data read in & transformation ####
# Set cd
setwd(PATH_IN_TSW)
# read in data
TSW_DS_20 <- read.csv("spin_TSW_DS.csv")
TSW_DS_20$id <- as.factor(TSW_DS_20$id)
TSW_relative_DS_20 <- read.csv("spin_TSW_relative_DS.csv")
TSW_relative_DS_20$id <- as.factor(TSW_relative_DS_20$id)
# transform to long format
TSW_DS_20 <- TSW_DS_20 %>%
gather(conditions, output, clustercount_sham:speedmax_stim) %>%
separate(conditions, c('type','stim')) %>%
arrange(id, stim)
TSW_relative_DS_20 <- TSW_relative_DS_20 %>%
gather(conditions, output, countbyduration_sham:countbysocount_stim) %>%
separate(conditions, c('type','stim')) %>%
arrange(id, stim)
#### end ####
#### SO spindle coupling: Data read in & transformation ####
# Set cd
setwd(PATH_IN_coupling)
coupling_20 <- read.csv("proportion_SOs_summary.csv")
coupling_20$id <- as.factor(coupling_20$id)
# transform to long format
coupling_20 <- coupling_20 %>%
gather(conditions, output, frontal_stim:parietal_sham) %>%
separate(conditions, c('type','stim')) %>%
arrange(id, stim)
#### end ####
#### Normality and data setup ####
# Set cd
setwd(PATH_OUT)
# check normal distribution
normality_travel_peakduration <- list()
normality_travel_duration <- list()
normality_travel_distance <- list()
normality_travel_distancemax <- list()
normality_travel_speed <- list()
normality_travel_speedmax <- list()
normality_travel_clustersize <- list()
normality_LGT_fig <- list()
normality_Zeich <- list()
normality_so_count <- list()
normality_cluster_count <- list()
normality_cluster_count_DS <- list()
normality_cluster_count_rel <- list()
normality_sleep_N2 <- list()
normality_sleep_SWS <- list()
normality_coupling_frontal <- list()
normality_coupling_parietal <- list()
for (types in 1:4){
criteria <- LGT_categories_stim_20$stim[types]
print(criteria)
normality <- shapiro.test(TSW_DS_20[which(TSW_DS_20$stim == criteria & TSW_DS_20$type == 'clustercount'),4])
normality_cluster_count_DS[criteria] <- normality$p.value
normality <- shapiro.test(TSW_DS_20[which(TSW_DS_20$stim == criteria & TSW_DS_20$type == 'peakduration'),4])
normality_travel_peakduration[criteria] <- normality$p.value
normality <- shapiro.test(TSW_DS_20[which(TSW_DS_20$stim == criteria & TSW_DS_20$type == 'distancemax'),4])
normality_travel_distancemax[criteria] <- normality$p.value
normality <- shapiro.test(TSW_DS_20[which(TSW_DS_20$stim == criteria & TSW_DS_20$type == 'speedmax'),4])
normality_travel_speedmax[criteria] <- normality$p.value
normality <- shapiro.test(TSW_DS_20[which(TSW_DS_20$stim == criteria & TSW_DS_20$type == 'clustersize'),4])
normality_travel_clustersize[criteria] <- normality$p.value
normality <- shapiro.test(TSW_relative_DS_20[which(TSW_relative_DS_20$stim == criteria & TSW_relative_DS_20$type == 'countbyduration'),4])
normality_cluster_count_rel[criteria] <- normality$p.value
normality <- shapiro.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == criteria & LGT_tests_stim_20$test == 'Zeich'),7])
normality_Zeich[criteria] <- normality$p.value
normality <- shapiro.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == criteria & LGT_categories_stim_20$category == 'FG'),7])
normality_LGT_fig[criteria] <- normality$p.value
normality <- shapiro.test(sleep_so_20[which(sleep_so_20$stim == criteria & sleep_so_20$type == 'counttotal'),4])
normality_so_count[criteria] <- normality$p.value
normality <- shapiro.test(rel_sleep_stim_20[which(rel_sleep_stim_20$stim == criteria & rel_sleep_stim_20$type == 'N2'),7])
normality_sleep_N2[criteria] <- normality$p.value
normality <- shapiro.test(rel_sleep_stim_20[which(rel_sleep_stim_20$stim == criteria & rel_sleep_stim_20$type == 'SWS'),7])
normality_sleep_SWS[criteria] <- normality$p.value
normality <- shapiro.test(coupling_20[which(coupling_20$stim == criteria & coupling_20$type == 'frontal'),4])
normality_coupling_frontal[criteria] <- normality$p.value
normality <- shapiro.test(coupling_20[which(coupling_20$stim == criteria & coupling_20$type == 'parietal'),4])
normality_coupling_parietal[criteria] <- normality$p.value}
# correlations
df.regression_stim = data.frame(id = as.character(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"&LGT_categories_stim_20$stim == "stim"]),
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'stim' & LGT_categories_stim_20$category == 'FG'),7],
Zeich = LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'stim' & LGT_tests_stim_20$test == 'Zeich'),7],
N2 = rel_sleep_stim_20[which(rel_sleep_stim_20$stim == 'stim' & rel_sleep_stim_20$type == 'N2'),7],
SWS = rel_sleep_stim_20[which(rel_sleep_stim_20$stim == 'stim' & rel_sleep_stim_20$type == 'SWS'),7],
socount = sleep_so_20[which(sleep_so_20$stim == 'stim' & sleep_so_20$type == 'counttotal'),4],
clustercountDS = TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'clustercount'),4],
clustercountrel = TSW_relative_DS_20[which(TSW_relative_DS_20$stim == 'stim' & TSW_relative_DS_20$type == 'countbyduration'),4],
clustersize = TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'clustersize'),4],
travelpeakdur = TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'peakduration'),4],
traveldistancemax = TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'distancemax'),4],
travelspeedmax = TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'speedmax'),4],
coupling_frontal = coupling_20[which(coupling_20$stim == 'stim' & coupling_20$type == 'frontal'),4],
coupling_parietal = coupling_20[which(coupling_20$stim == 'stim' & coupling_20$type == 'parietal'),4]
)
df.regression_sham = data.frame(id = as.character(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"&LGT_categories_stim_20$stim == "stim"]),
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'sham' & LGT_categories_stim_20$category == 'FG'),7],
Zeich = LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'sham' & LGT_tests_stim_20$test == 'Zeich'),7],
N2 = rel_sleep_stim_20[which(rel_sleep_stim_20$stim == 'sham' & rel_sleep_stim_20$type == 'N2'),7],
SWS = rel_sleep_stim_20[which(rel_sleep_stim_20$stim == 'sham' & rel_sleep_stim_20$type == 'SWS'),7],
socount = sleep_so_20[which(sleep_so_20$stim == 'sham' & sleep_so_20$type == 'counttotal'),4],
clustercountDS = TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'clustercount'),4],
clustercountrel = TSW_relative_DS_20[which(TSW_relative_DS_20$stim == 'sham' & TSW_relative_DS_20$type == 'countbyduration'),4],
clustersize = TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'clustersize'),4],
travelpeakdur = TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'peakduration'),4],
traveldistancemax = TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'distancemax'),4],
travelspeedmax = TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'speedmax'),4],
coupling_frontal = coupling_20[which(coupling_20$stim == 'sham' & coupling_20$type == 'frontal'),4],
coupling_parietal = coupling_20[which(coupling_20$stim == 'sham' & coupling_20$type == 'parietal'),4]
)
#### end ####
#### Correlations figural memory ####
# behaviour FG & N2 duration
# repeated correlations: FG & N2
df.regression_FG_N2 = data.frame(id = as.factor(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"]),
N2 = rel_sleep_stim_20[which(rel_sleep_stim_20$type == 'N2'),7],
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7]
)
corr_overall_FG_N2 <- rmcorr(participant = "id", measure1 = 'FG', measure2 = 'N2', dataset = df.regression_FG_N2)
# individual correlations
results_FG_N2_stim <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'stim' & LGT_categories_stim_20$category == 'FG'),7], rel_sleep_stim_20[which(rel_sleep_stim_20$stim == 'stim' & rel_sleep_stim_20$type == 'N2'),7], alternative = 'two.sided',method = 'pearson')
results_FG_N2_sham <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'sham' & LGT_categories_stim_20$category == 'FG'),7], rel_sleep_stim_20[which(rel_sleep_stim_20$stim == 'sham' & rel_sleep_stim_20$type == 'N2'),7], alternative = 'two.sided',method = 'pearson')
# overall plot: FG & N2
df <- rbind.data.frame(
data.frame(x = df.regression_stim$FG, y = df.regression_stim$N2, id = "stim"),
data.frame(x = df.regression_sham$FG, y = df.regression_sham$N2, id = "sham")
)
plot_FG_N2 <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "pearson",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "% in N2") +
theme(text = element_text(size = 20, family = "arial"))#,
print(plot_FG_N2)
ggsave('FG_N2.svg', plot = plot_FG_N2, width = 4, height = 6, dpi = 300)
ggsave('FG_N2_wider.svg', plot = plot_FG_N2, width = 5.5, height = 6, dpi = 300)
# behaviour FG & SWS duration
# repeated correlations: FG & SWS
df.regression_FG_SWS = data.frame(id = as.factor(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"]),
SWS = rel_sleep_stim_20[which(rel_sleep_stim_20$type == 'SWS'),7],
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7]
)
corr_overall_FG_SWS <- rmcorr(participant = "id", measure1 = 'FG', measure2 = 'SWS', dataset = df.regression_FG_SWS)
# individual correlations
results_FG_SWS_stim <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'stim' & LGT_categories_stim_20$category == 'FG'),7], rel_sleep_stim_20[which(rel_sleep_stim_20$stim == 'stim' & rel_sleep_stim_20$type == 'SWS'),7], alternative = 'two.sided',method = 'spearman')
results_FG_SWS_sham <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'sham' & LGT_categories_stim_20$category == 'FG'),7], rel_sleep_stim_20[which(rel_sleep_stim_20$stim == 'sham' & rel_sleep_stim_20$type == 'SWS'),7], alternative = 'two.sided',method = 'pearson')
# overall plot: FG & SWS
df <- rbind.data.frame(
data.frame(x = df.regression_stim$FG, y = df.regression_stim$SWS, id = "stim"),
data.frame(x = df.regression_sham$FG, y = df.regression_sham$SWS, id = "sham")
)
plot_FG_SWS <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "spearman",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "% in SWS") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_FG_SWS)
ggsave('FG_SWS.svg', plot = plot_FG_SWS, width = 4, height = 6, dpi = 300)
ggsave('FG_SWS_wider.svg', plot = plot_FG_SWS, width = 5.5, height = 6, dpi = 300)
# behaviour FG & SO count total
# repeated correlations: FG & SO count
df.regression_FG_socount = data.frame(id = as.factor(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"]),
socount = sleep_so_20[which(sleep_so_20$type == 'counttotal'),4],
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7]
)
corr_overall_FG_socount <- rmcorr(participant = "id", measure1 = 'FG', measure2 = 'socount', dataset = df.regression_FG_socount)
# individual correlations
results_FG_socount_stim <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'stim' & LGT_categories_stim_20$category == 'FG'),7], sleep_so_20[which(sleep_so_20$stim == 'stim' & sleep_so_20$type == 'counttotal'),4], alternative = 'two.sided',method = 'pearson')
results_FG_socount_sham <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'sham' & LGT_categories_stim_20$category == 'FG'),7], sleep_so_20[which(sleep_so_20$stim == 'sham' & sleep_so_20$type == 'counttotal'),4], alternative = 'two.sided',method = 'pearson')
# overall plot: FG & socount
df <- rbind.data.frame(
data.frame(x = df.regression_stim$FG, y = df.regression_stim$socount, id = "stim"),
data.frame(x = df.regression_sham$FG, y = df.regression_sham$socount, id = "sham")
)
plot_FG_socount <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "pearson",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "SO count") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_FG_socount)
ggsave('FG_socount.svg', plot = plot_FG_socount, width = 4, height = 6, dpi = 300)
ggsave('FG_socount_wider.svg', plot = plot_FG_socount, width = 5.5, height = 6, dpi = 300)
# behaviour FG & travelling peak duration
# repeated correlations: FG & travel peak duration
df.regression_FG_travelpeakdur = data.frame(id = as.factor(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"]),
travelpeakdur = TSW_DS_20[which(TSW_DS_20$type == 'peakduration'),4],
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7]
)
corr_overall_FG_travelpeakdur <- rmcorr(participant = "id", measure1 = 'FG', measure2 = 'travelpeakdur', dataset = df.regression_FG_travelpeakdur)
# individual correlations
results_FG_travelpeakduration_stim <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'stim' & LGT_categories_stim_20$category == 'FG'),7], TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'peakduration'),4], alternative = 'two.sided',method = 'pearson')
results_FG_travelpeakduration_sham <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'sham' & LGT_categories_stim_20$category == 'FG'),7], TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'peakduration'),4], alternative = 'two.sided',method = 'pearson')
# overall plot: FG & travelling peakduration
df <- rbind.data.frame(
data.frame(x = df.regression_stim$FG, y = df.regression_stim$travelpeakdur, id = "stim"),
data.frame(x = df.regression_sham$FG, y = df.regression_sham$travelpeakdur, id = "sham")
)
plot_FG_travelpeakduration <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "pearson",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "peak to peak duration") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_FG_travelpeakduration)
ggsave('FG_travelpeakduration.svg', plot = plot_FG_travelpeakduration, width = 4, height = 6, dpi = 300)
# behaviour FG & travelling distancemax
# repeated correlations: FG & travel distancemax
df.regression_FG_traveldistancemax = data.frame(id = as.factor(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"]),
traveldistancemax = TSW_DS_20[which(TSW_DS_20$type == 'distancemax'),4],
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7]
)
corr_overall_FG_traveldistancemax <- rmcorr(participant = "id", measure1 = 'FG', measure2 = 'traveldistancemax', dataset = df.regression_FG_traveldistancemax)
# individual correlations
results_FG_traveldistancemax_stim <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'stim' & LGT_categories_stim_20$category == 'FG'),7], TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'distancemax'),4], alternative = 'two.sided',method = 'pearson')
results_FG_traveldistancemax_sham <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'sham' & LGT_categories_stim_20$category == 'FG'),7], TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'distancemax'),4], alternative = 'two.sided',method = 'pearson')
# overall plot: FG & travelling distancemax
df <- rbind.data.frame(
data.frame(x = df.regression_stim$FG, y = df.regression_stim$traveldistancemax, id = "stim"),
data.frame(x = df.regression_sham$FG, y = df.regression_sham$traveldistancemax, id = "sham")
)
plot_FG_traveldistancemax <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "pearson",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "SO travel distancemax") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_FG_traveldistancemax)
ggsave('FG_traveldistancemax.svg', plot = plot_FG_traveldistancemax, width = 4, height = 6, dpi = 300)
# behaviour FG & travelling speedmax
# repeated correlations: FG & travel speedmax
df.regression_FG_travelspeedmax = data.frame(id = as.factor(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"]),
travelspeedmax = TSW_DS_20[which(TSW_DS_20$type == 'speedmax'),4],
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7]
)
corr_overall_FG_travelspeedmax <- rmcorr(participant = "id", measure1 = 'FG', measure2 = 'travelspeedmax', dataset = df.regression_FG_travelspeedmax)
# individual correlations
results_FG_travelspeedmax_stim <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'stim' & LGT_categories_stim_20$category == 'FG'),7], TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'speedmax'),4], alternative = 'two.sided',method = 'spearman')
results_FG_travelspeedmax_sham <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'sham' & LGT_categories_stim_20$category == 'FG'),7], TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'speedmax'),4], alternative = 'two.sided',method = 'spearman')
# overall plot: FG & travelling speedmax
df <- rbind.data.frame(
data.frame(x = df.regression_stim$FG, y = df.regression_stim$travelspeedmax, id = "stim"),
data.frame(x = df.regression_sham$FG, y = df.regression_sham$travelspeedmax, id = "sham")
)
plot_FG_travelspeedmax <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "spearman",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "SO travel speedmax") +
theme(text = element_text(size = 20, family = "arial"))
ggsave('FG_travelspeedmax.svg', plot = plot_FG_travelspeedmax, width = 4, height = 6, dpi = 300)
# behaviour FG & travelling clustersize
# repeated correlations: FG & clustersize
df.regression_FG_clustersize = data.frame(id = as.factor(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"]),
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7],
clustersize = TSW_DS_20[which(TSW_DS_20$type == 'clustersize'),4]
)
corr_overall_FG_clustersize <- rmcorr(participant = "id", measure1 = 'FG', measure2 = 'clustersize', dataset = df.regression_FG_clustersize)
# individual correlations
results_FG_clustersize_stim <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'stim' & LGT_categories_stim_20$category == 'FG'),7], TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'clustersize'),4], alternative = 'two.sided',method = 'pearson')
results_FG_clustersize_sham <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'sham' & LGT_categories_stim_20$category == 'FG'),7], TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'clustersize'),4], alternative = 'two.sided',method = 'pearson')
# overall plot: FG & clustersize
df <- rbind.data.frame(
data.frame(x = df.regression_stim$FG, y = df.regression_stim$clustersize, id = "stim"),
data.frame(x = df.regression_sham$FG, y = df.regression_sham$clustersize, id = "sham")
)
plot_FG_clustersize <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "pearson",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "SO cluster size") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_FG_clustersize)
ggsave('FG_clustersize.svg', plot = plot_FG_clustersize, width = 4, height = 6, dpi = 300)
# behaviour FG & SO cluster count DS
# repeated correlations: FG & cluster count DS
df.regression_FG_clustercountDS = data.frame(id = as.factor(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"]),
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7],
clustercountDS = TSW_DS_20[which(TSW_DS_20$type == 'clustercount'),4]
)
corr_overall_FG_clustercountDS <- rmcorr(participant = "id", measure1 = 'FG', measure2 = 'clustercountDS', dataset = df.regression_FG_clustercountDS)
# individual correlation
results_FG_clustercountDS_stim <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'stim' & LGT_categories_stim_20$category == 'FG'),7], TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'clustercount'),4], alternative = 'two.sided',method = 'pearson')
results_FG_clustercountDS_sham <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'sham' & LGT_categories_stim_20$category == 'FG'),7], TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'clustercount'),4], alternative = 'two.sided',method = 'pearson')
# overall plot: FG & clustercountDS
df <- rbind.data.frame(
data.frame(x = df.regression_stim$FG, y = df.regression_stim$clustercountDS, id = "stim"),
data.frame(x = df.regression_sham$FG, y = df.regression_sham$clustercountDS, id = "sham")
)
plot_FG_clustercountDS <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "pearson",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "SO cluster count") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_FG_clustercountDS)
ggsave('FG_clustercountDS.svg', plot = plot_FG_clustercountDS, width = 4, height = 6, dpi = 300)
# behaviour FG & SO cluster count relative
# repeated correlations: FG & clustercount rel
df.regression_FG_clustercountrel = data.frame(id = as.factor(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"]),
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7],
clustercountrel = TSW_relative_DS_20[which(TSW_relative_DS_20$type == 'countbyduration'),4]
)
corr_overall_FG_clustercountrel <- rmcorr(participant = "id", measure1 = 'FG', measure2 = 'clustercountrel', dataset = df.regression_FG_clustercountrel)
# individual correlations
results_FG_clustercountrel_stim <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'stim' & LGT_categories_stim_20$category == 'FG'),7], TSW_relative_DS_20[which(TSW_relative_DS_20$stim == 'stim' & TSW_relative_DS_20$type == 'countbyduration'),4], alternative = 'two.sided',method = 'pearson')
results_FG_clustercountrel_sham <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'sham' & LGT_categories_stim_20$category == 'FG'),7], TSW_relative_DS_20[which(TSW_relative_DS_20$stim == 'sham' & TSW_relative_DS_20$type == 'countbyduration'),4], alternative = 'two.sided',method = 'spearman')
# overall plot: FG & clustercount relative
df <- rbind.data.frame(
data.frame(x = df.regression_stim$FG, y = df.regression_stim$clustercountrel, id = "stim"),
data.frame(x = df.regression_sham$FG, y = df.regression_sham$clustercountrel, id = "sham")
)
plot_FG_clustercountrel <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "spearman",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "SO cluster count (rel)") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_FG_clustercountrel)
ggsave('FG_clustercountrel.svg', plot = plot_FG_clustercountrel, width = 4, height = 6, dpi = 300)
# behaviour FG & coupling frontal
# repeated correlations: FG & coupling
df.regression_FG_couplingfrontal = data.frame(id = as.factor(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"]),
coupling = coupling_20[which(coupling_20$type == 'frontal'),4],
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7]
)
corr_overall_FG_couplingfrontal <- rmcorr(participant = "id", measure1 = 'FG', measure2 = 'coupling', dataset = df.regression_FG_couplingfrontal)
# individual correlations
results_FG_couplingfrontal_stim <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'stim' & LGT_categories_stim_20$category == 'FG'),7], coupling_20[which(coupling_20$stim == 'stim' & coupling_20$type == 'frontal'),4], alternative = 'two.sided',method = 'spearman')
results_FG_couplingfrontal_sham <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'sham' & LGT_categories_stim_20$category == 'FG'),7], coupling_20[which(coupling_20$stim == 'sham' & coupling_20$type == 'frontal'),4], alternative = 'two.sided',method = 'spearman')
# overall plot: FG & coupling frontal
df <- rbind.data.frame(
data.frame(x = df.regression_stim$FG, y = df.regression_stim$coupling_frontal, id = "stim"),
data.frame(x = df.regression_sham$FG, y = df.regression_sham$coupling_frontal, id = "sham")
)
plot_FG_couplingfrontal <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "spearman",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "% in coupling") +
theme(text = element_text(size = 20, family = "arial"))#,
print(plot_FG_couplingfrontal)
ggsave('FG_couplingfrontal.svg', plot = plot_FG_couplingfrontal, width = 4, height = 6, dpi = 300)
ggsave('FG_couplingfrontal_wider.svg', plot = plot_FG_couplingfrontal, width = 5.5, height = 6, dpi = 300)
# behaviour FG & coupling parietal
# repeated correlations: FG & coupling
df.regression_FG_couplingparietal = data.frame(id = as.factor(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"]),
coupling = coupling_20[which(coupling_20$type == 'parietal'),4],
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7]
)
corr_overall_FG_couplingparietal <- rmcorr(participant = "id", measure1 = 'FG', measure2 = 'coupling', dataset = df.regression_FG_couplingparietal)
# individual correlations
results_FG_couplingparietal_stim <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'stim' & LGT_categories_stim_20$category == 'FG'),7], coupling_20[which(coupling_20$stim == 'stim' & coupling_20$type == 'parietal'),4], alternative = 'two.sided',method = 'spearman')
results_FG_couplingparietal_sham <- cor.test(LGT_categories_stim_20[which(LGT_categories_stim_20$stim == 'sham' & LGT_categories_stim_20$category == 'FG'),7], coupling_20[which(coupling_20$stim == 'sham' & coupling_20$type == 'parietal'),4], alternative = 'two.sided',method = 'spearman')
# overall plot: FG & coupling parietal
df <- rbind.data.frame(
data.frame(x = df.regression_stim$FG, y = df.regression_stim$coupling_parietal, id = "stim"),
data.frame(x = df.regression_sham$FG, y = df.regression_sham$coupling_parietal, id = "sham")
)
plot_FG_couplingparietal <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "spearman",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "% in coupling") +
theme(text = element_text(size = 20, family = "arial"))#,
print(plot_FG_couplingparietal)
ggsave('FG_couplingparietal.svg', plot = plot_FG_couplingparietal, width = 4, height = 6, dpi = 300)
ggsave('FG_couplingparietal_wider.svg', plot = plot_FG_couplingparietal, width = 5.5, height = 6, dpi = 300)
#### end ####
#### correlations with Zeichen subtest ####
# behaviour Zeich & N2 duration
# repeated correlations: Zeich & N2
df.regression_Zeich_N2 = data.frame(id = as.factor(LGT_tests_stim_20$ID[LGT_tests_stim_20$test == "Zeich"]),
N2 = rel_sleep_stim_20[which(rel_sleep_stim_20$type == 'N2'),7],
Zeich = LGT_tests_stim_20[which(LGT_tests_stim_20$test == 'Zeich'),7]
)
corr_overall_Zeich_N2 <- rmcorr(participant = "id", measure1 = 'Zeich', measure2 = 'N2', dataset = df.regression_Zeich_N2)
# individual correlations
results_Zeich_N2_stim <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'stim' & LGT_tests_stim_20$test == 'Zeich'),7], rel_sleep_stim_20[which(rel_sleep_stim_20$stim == 'stim' & rel_sleep_stim_20$type == 'N2'),7], alternative = 'two.sided',method = 'pearson')
results_Zeich_N2_sham <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'sham' & LGT_tests_stim_20$test == 'Zeich'),7], rel_sleep_stim_20[which(rel_sleep_stim_20$stim == 'sham' & rel_sleep_stim_20$type == 'N2'),7], alternative = 'two.sided',method = 'pearson')
# overall plot: Zeich & N2
df <- rbind.data.frame(
data.frame(x = df.regression_stim$Zeich, y = df.regression_stim$N2, id = "stim"),
data.frame(x = df.regression_sham$Zeich, y = df.regression_sham$N2, id = "sham")
)
plot_Zeich_N2 <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "pearson",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "% in N2") +
theme(text = element_text(size = 20, family = "arial"))#,
print(plot_Zeich_N2)
ggsave('Zeich_N2.svg', plot = plot_Zeich_N2, width = 4, height = 6, dpi = 300)
ggsave('Zeich_N2_wider.svg', plot = plot_Zeich_N2, width = 5.5, height = 6, dpi = 300)
# behaviour Zeich & SWS duration
# repeated correlations: Zeich & SWS
df.regression_Zeich_SWS = data.frame(id = as.factor(LGT_tests_stim_20$ID[LGT_tests_stim_20$test == "Zeich"]),
SWS = rel_sleep_stim_20[which(rel_sleep_stim_20$type == 'SWS'),7],
Zeich = LGT_tests_stim_20[which(LGT_tests_stim_20$test == 'Zeich'),7]
)
corr_overall_Zeich_SWS <- rmcorr(participant = "id", measure1 = 'Zeich', measure2 = 'SWS', dataset = df.regression_Zeich_SWS)
# individual correlations
results_Zeich_SWS_stim <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'stim' & LGT_tests_stim_20$test == 'Zeich'),7], rel_sleep_stim_20[which(rel_sleep_stim_20$stim == 'stim' & rel_sleep_stim_20$type == 'SWS'),7], alternative = 'two.sided',method = 'spearman')
results_Zeich_SWS_sham <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'sham' & LGT_tests_stim_20$test == 'Zeich'),7], rel_sleep_stim_20[which(rel_sleep_stim_20$stim == 'sham' & rel_sleep_stim_20$type == 'SWS'),7], alternative = 'two.sided',method = 'pearson')
# overall plot: Zeich & SWS
df <- rbind.data.frame(
data.frame(x = df.regression_stim$Zeich, y = df.regression_stim$SWS, id = "stim"),
data.frame(x = df.regression_sham$Zeich, y = df.regression_sham$SWS, id = "sham")
)
plot_Zeich_SWS <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "spearman",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "% in SWS") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_Zeich_SWS)
ggsave('Zeich_SWS.svg', plot = plot_Zeich_SWS, width = 4, height = 6, dpi = 300)
ggsave('Zeich_SWS_wider.svg', plot = plot_Zeich_SWS, width = 5.5, height = 6, dpi = 300)
# behaviour Zeich & SO count total
# repeated correlations: Zeich & SO count
df.regression_Zeich_socount = data.frame(id = as.factor(LGT_tests_stim_20$ID[LGT_tests_stim_20$test == "Zeich"]),
socount = sleep_so_20[which(sleep_so_20$type == 'counttotal'),4],
Zeich = LGT_tests_stim_20[which(LGT_tests_stim_20$test == 'Zeich'),7]
)
corr_overall_Zeich_socount <- rmcorr(participant = "id", measure1 = 'Zeich', measure2 = 'socount', dataset = df.regression_Zeich_socount)
# individual correlations
results_Zeich_socount_stim <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'stim' & LGT_tests_stim_20$test == 'Zeich'),7], sleep_so_20[which(sleep_so_20$stim == 'stim' & sleep_so_20$type == 'counttotal'),4], alternative = 'two.sided',method = 'pearson')
results_Zeich_socount_sham <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'sham' & LGT_tests_stim_20$test == 'Zeich'),7], sleep_so_20[which(sleep_so_20$stim == 'sham' & sleep_so_20$type == 'counttotal'),4], alternative = 'two.sided',method = 'pearson')
# overall plot: Zeich & socount
df <- rbind.data.frame(
data.frame(x = df.regression_stim$Zeich, y = df.regression_stim$socount, id = "stim"),
data.frame(x = df.regression_sham$Zeich, y = df.regression_sham$socount, id = "sham")
)
plot_Zeich_socount <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "pearson",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "SO count") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_Zeich_socount)
ggsave('Zeich_socount.svg', plot = plot_Zeich_socount, width = 4, height = 6, dpi = 300)
ggsave('Zeich_socount_wider.svg', plot = plot_Zeich_socount, width = 5.5, height = 6, dpi = 300)
# behaviour Zeich & travelling peak duration
# repeated correlations: Zeich & travel peak duration
df.regression_Zeich_travelpeakdur = data.frame(id = as.factor(LGT_tests_stim_20$ID[LGT_tests_stim_20$test == "Zeich"]),
travelpeakdur = TSW_DS_20[which(TSW_DS_20$type == 'peakduration'),4],
Zeich = LGT_tests_stim_20[which(LGT_tests_stim_20$test == 'Zeich'),7]
)
corr_overall_Zeich_travelpeakdur <- rmcorr(participant = "id", measure1 = 'Zeich', measure2 = 'travelpeakdur', dataset = df.regression_Zeich_travelpeakdur)
# individual correlations
results_Zeich_travelpeakduration_stim <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'stim' & LGT_tests_stim_20$test == 'Zeich'),7], TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'peakduration'),4], alternative = 'two.sided',method = 'pearson')
results_Zeich_travelpeakduration_sham <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'sham' & LGT_tests_stim_20$test == 'Zeich'),7], TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'peakduration'),4], alternative = 'two.sided',method = 'pearson')
# overall plot: Zeich & travelling peakduration
df <- rbind.data.frame(
data.frame(x = df.regression_stim$Zeich, y = df.regression_stim$travelpeakdur, id = "stim"),
data.frame(x = df.regression_sham$Zeich, y = df.regression_sham$travelpeakdur, id = "sham")
)
plot_Zeich_travelpeakduration <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "pearson",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "peak to peak duration") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_Zeich_travelpeakduration)
ggsave('Zeich_travelpeakduration.svg', plot = plot_Zeich_travelpeakduration, width = 4, height = 6, dpi = 300)
# behaviour Zeich & travelling distancemax
# repeated correlations: Zeich & travel distancemax
df.regression_Zeich_traveldistancemax = data.frame(id = as.factor(LGT_tests_stim_20$ID[LGT_tests_stim_20$test == "Zeich"]),
traveldistancemax = TSW_DS_20[which(TSW_DS_20$type == 'distancemax'),4],
Zeich = LGT_tests_stim_20[which(LGT_tests_stim_20$test == 'Zeich'),7]
)
corr_overall_Zeich_traveldistancemax <- rmcorr(participant = "id", measure1 = 'Zeich', measure2 = 'traveldistancemax', dataset = df.regression_Zeich_traveldistancemax)
# individual correlations
results_Zeich_traveldistancemax_stim <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'stim' & LGT_tests_stim_20$test == 'Zeich'),7], TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'distancemax'),4], alternative = 'two.sided',method = 'pearson')
results_Zeich_traveldistancemax_sham <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'sham' & LGT_tests_stim_20$test == 'Zeich'),7], TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'distancemax'),4], alternative = 'two.sided',method = 'pearson')
# overall plot: Zeich & travelling distancemax
df <- rbind.data.frame(
data.frame(x = df.regression_stim$Zeich, y = df.regression_stim$traveldistancemax, id = "stim"),
data.frame(x = df.regression_sham$Zeich, y = df.regression_sham$traveldistancemax, id = "sham")
)
plot_Zeich_traveldistancemax <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "pearson",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "SO travel distancemax") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_Zeich_traveldistancemax)
ggsave('Zeich_traveldistancemax.svg', plot = plot_Zeich_traveldistancemax, width = 4, height = 6, dpi = 300)
# behaviour Zeich & travelling speedmax
# repeated correlations: Zeich & travel speedmax
df.regression_Zeich_travelspeedmax = data.frame(id = as.factor(LGT_tests_stim_20$ID[LGT_tests_stim_20$test == "Zeich"]),
travelspeedmax = TSW_DS_20[which(TSW_DS_20$type == 'speedmax'),4],
Zeich = LGT_tests_stim_20[which(LGT_tests_stim_20$test == 'Zeich'),7]
)
corr_overall_Zeich_travelspeedmax <- rmcorr(participant = "id", measure1 = 'Zeich', measure2 = 'travelspeedmax', dataset = df.regression_Zeich_travelspeedmax)
# individual correlations
results_Zeich_travelspeedmax_stim <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'stim' & LGT_tests_stim_20$test == 'Zeich'),7], TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'speedmax'),4], alternative = 'two.sided',method = 'spearman')
results_Zeich_travelspeedmax_sham <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'sham' & LGT_tests_stim_20$test == 'Zeich'),7], TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'speedmax'),4], alternative = 'two.sided',method = 'spearman')
# overall plot: Zeich & travelling speedmax
df <- rbind.data.frame(
data.frame(x = df.regression_stim$Zeich, y = df.regression_stim$travelspeedmax, id = "stim"),
data.frame(x = df.regression_sham$Zeich, y = df.regression_sham$travelspeedmax, id = "sham")
)
plot_Zeich_travelspeedmax <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "spearman",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "SO travel speedmax") +
theme(text = element_text(size = 20, family = "arial"))
ggsave('Zeich_travelspeedmax.svg', plot = plot_Zeich_travelspeedmax, width = 4, height = 6, dpi = 300)
# behaviour Zeich & travelling clustersize
# repeated correlations: Zeich & clustersize
df.regression_Zeich_clustersize = data.frame(id = as.factor(LGT_tests_stim_20$ID[LGT_tests_stim_20$test == "Zeich"]),
Zeich = LGT_tests_stim_20[which(LGT_tests_stim_20$test == 'Zeich'),7],
clustersize = TSW_DS_20[which(TSW_DS_20$type == 'clustersize'),4]
)
corr_overall_Zeich_clustersize <- rmcorr(participant = "id", measure1 = 'Zeich', measure2 = 'clustersize', dataset = df.regression_Zeich_clustersize)
# individual correlations
results_Zeich_clustersize_stim <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'stim' & LGT_tests_stim_20$test == 'Zeich'),7], TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'clustersize'),4], alternative = 'two.sided',method = 'pearson')
results_Zeich_clustersize_sham <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'sham' & LGT_tests_stim_20$test == 'Zeich'),7], TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'clustersize'),4], alternative = 'two.sided',method = 'pearson')
# overall plot: Zeich & clustersize
df <- rbind.data.frame(
data.frame(x = df.regression_stim$Zeich, y = df.regression_stim$clustersize, id = "stim"),
data.frame(x = df.regression_sham$Zeich, y = df.regression_sham$clustersize, id = "sham")
)
plot_Zeich_clustersize <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "pearson",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "SO cluster size") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_Zeich_clustersize)
ggsave('Zeich_clustersize.svg', plot = plot_Zeich_clustersize, width = 4, height = 6, dpi = 300)
# behaviour Zeich & SO cluster count DS
# repeated correlations: Zeich & cluster count DS
df.regression_Zeich_clustercountDS = data.frame(id = as.factor(LGT_tests_stim_20$ID[LGT_tests_stim_20$test == "Zeich"]),
Zeich = LGT_tests_stim_20[which(LGT_tests_stim_20$test == 'Zeich'),7],
clustercountDS = TSW_DS_20[which(TSW_DS_20$type == 'clustercount'),4]
)
corr_overall_Zeich_clustercountDS <- rmcorr(participant = "id", measure1 = 'Zeich', measure2 = 'clustercountDS', dataset = df.regression_Zeich_clustercountDS)
# individual correlation
results_Zeich_clustercountDS_stim <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'stim' & LGT_tests_stim_20$test == 'Zeich'),7], TSW_DS_20[which(TSW_DS_20$stim == 'stim' & TSW_DS_20$type == 'clustercount'),4], alternative = 'two.sided',method = 'spearman')
results_Zeich_clustercountDS_sham <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'sham' & LGT_tests_stim_20$test == 'Zeich'),7], TSW_DS_20[which(TSW_DS_20$stim == 'sham' & TSW_DS_20$type == 'clustercount'),4], alternative = 'two.sided',method = 'pearson')
# overall plot: Zeich & clustercountDS
df <- rbind.data.frame(
data.frame(x = df.regression_stim$Zeich, y = df.regression_stim$clustercountDS, id = "stim"),
data.frame(x = df.regression_sham$Zeich, y = df.regression_sham$clustercountDS, id = "sham")
)
plot_Zeich_clustercountDS <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "spearman",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "SO cluster count") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_Zeich_clustercountDS)
ggsave('Zeich_clustercountDS.svg', plot = plot_Zeich_clustercountDS, width = 4, height = 6, dpi = 300)
# behaviour Zeich & SO cluster count relative
# repeated correlations: Zeich & clustercount rel
df.regression_Zeich_clustercountrel = data.frame(id = as.factor(LGT_tests_stim_20$ID[LGT_tests_stim_20$test == "Zeich"]),
Zeich = LGT_tests_stim_20[which(LGT_tests_stim_20$test == 'Zeich'),7],
clustercountrel = TSW_relative_DS_20[which(TSW_relative_DS_20$type == 'countbyduration'),4]
)
corr_overall_Zeich_clustercountrel <- rmcorr(participant = "id", measure1 = 'Zeich', measure2 = 'clustercountrel', dataset = df.regression_Zeich_clustercountrel)
# individual correlations
results_Zeich_clustercountrel_stim <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'stim' & LGT_tests_stim_20$test == 'Zeich'),7], TSW_relative_DS_20[which(TSW_relative_DS_20$stim == 'stim' & TSW_relative_DS_20$type == 'countbyduration'),4], alternative = 'two.sided',method = 'pearson')
results_Zeich_clustercountrel_sham <- cor.test(LGT_tests_stim_20[which(LGT_tests_stim_20$stim == 'sham' & LGT_tests_stim_20$test == 'Zeich'),7], TSW_relative_DS_20[which(TSW_relative_DS_20$stim == 'sham' & TSW_relative_DS_20$type == 'countbyduration'),4], alternative = 'two.sided',method = 'spearman')
# overall plot: Zeich & clustercount relative
df <- rbind.data.frame(
data.frame(x = df.regression_stim$Zeich, y = df.regression_stim$clustercountrel, id = "stim"),
data.frame(x = df.regression_sham$Zeich, y = df.regression_sham$clustercountrel, id = "sham")
)
plot_Zeich_clustercountrel <- ggplot(df, aes(x, y, color = id, fill = id)) +
geom_point(size = 3, alpha = 0.7) +
geom_smooth(method = "lm", alpha = 0.2, size = 1.5) +
stat_cor(method = "spearman",size = 7, family = 'arial') +
scale_color_manual(values = c(stim = color_stim, sham = color_sham), aesthetics = c("color", "fill")) +
theme_pubr() +
labs(x = "FS", y = "SO cluster count (rel)") +
theme(text = element_text(size = 20, family = "arial"))
print(plot_Zeich_clustercountrel)
ggsave('Zeich_clustercountrel.svg', plot = plot_Zeich_clustercountrel, width = 4, height = 6, dpi = 300)
#### end ####
#### Linear mixed-effects model (LMM) using lme4: Part 1####
df.regression = data.frame(id = as.factor(as.character(LGT_categories_stim_20$ID[LGT_categories_stim_20$category == "FG"])),
conditions = as.factor(LGT_categories_stim_20$stim[LGT_categories_stim_20$category == "FG"]),
FG = LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7],
Zeich = LGT_tests_stim_20[which(LGT_tests_stim_20$test == 'Zeich'),7],
N2 = rel_sleep_stim_20[which(rel_sleep_stim_20$type == 'N2'),7],
SWS = rel_sleep_stim_20[which(rel_sleep_stim_20$type == 'SWS'),7],
socount = sleep_so_20[which(sleep_so_20$type == 'counttotal'),4],
clustercountDS = TSW_DS_20[which(TSW_DS_20$type == 'clustercount'),4],
clustercountrel = TSW_relative_DS_20[which(TSW_relative_DS_20$type == 'countbyduration'),4],
clustersize = TSW_DS_20[which(TSW_DS_20$type == 'clustersize'),4],
travelpeakdur = TSW_DS_20[which(TSW_DS_20$type == 'peakduration'),4],
traveldistancemax = TSW_DS_20[which(TSW_DS_20$type == 'distancemax'),4],
travelspeedmax = TSW_DS_20[which(TSW_DS_20$type == 'speedmax'),4],
socount_z = as.numeric(scale(sleep_so_20[which(sleep_so_20$type == 'counttotal'),4])),
FG_z = as.numeric(scale(LGT_categories_stim_20[which(LGT_categories_stim_20$category == 'FG'),7])),
clustersize_z = as.numeric(scale(TSW_DS_20[which(TSW_DS_20$type == 'clustersize'),4])),
travelpeakdur_z = as.numeric(scale(TSW_DS_20[which(TSW_DS_20$type == 'peakduration'),4])),
traveldistancemax_z = as.numeric(scale(TSW_DS_20[which(TSW_DS_20$type == 'distancemax'),4])),
N2_z = as.numeric(scale(rel_sleep_stim_20[which(rel_sleep_stim_20$type == 'N2'),7])),
SWS_z = as.numeric(scale(rel_sleep_stim_20[which(rel_sleep_stim_20$type == 'SWS'),7]))
)
# Install if necessary
# install.packages("lme4")
# install.packages("lmerTest")
library(lme4)
library(lmerTest)
#### end ####
#### Principal component analysis: traveling features ####
library(dplyr) # For easy averaging
# List your raw feature column names (NOT the z-scored ones yet)
feature_cols <- c("travelpeakdur", "traveldistancemax", "clustersize")
# Calculate the mean for each subject across the two conditions
# This creates a dataset with 20 rows (one per person)
pca_input_data <- df.regression %>%
group_by(id) %>%
summarise(across(all_of(feature_cols), mean, na.rm = TRUE))
# Check dimensions: Should be 20 rows
dim(pca_input_data)
# Run PCA
pca_result <- prcomp(pca_input_data[, feature_cols],
center = TRUE,
scale. = TRUE)
# Inspect the results
summary(pca_result)
# Biplot: Arrows show how much each feature contributes to the components
biplot(pca_result, scale = 0)
# 1. Extract the rotation matrix (the "rules" of the PCA)
rotation_matrix <- pca_result$rotation
# 2. Select the feature columns from the ORIGINAL 40-row data
# Ensure they are scaled exactly the same way as the PCA input
# (prcomp scales internally, so we must manually scale our full data to match)
full_data_features <- df.regression[, feature_cols]
scaled_full_data <- scale(full_data_features,
center = pca_result$center,
scale = pca_result$scale)
# 3. Calculate the PC scores for all 40 rows
# Matrix multiplication: Data %*% Rotation
pc_scores <- scaled_full_data %*% rotation_matrix
# 4. Add the first Principal Component (PC1) to your main data frame
df.regression$PC1_AllFeatures <- pc_scores[, 1]
# Optional: Add PC2 if it also explains significant variance
df.regression$PC2_AllFeatures <- pc_scores[, 2]
# Verify it worked
head(df.regression[, c("id", "conditions", "PC1_AllFeatures")])
loadings_pc1 <- pca_result$rotation[, "PC1"]
print(sort(loadings_pc1, decreasing = TRUE))
#### end ####
#### Linear mixed-effects model (LMM) using lme4: Part 2 SWS model ####
# Model 1: The baseline (Control for SWS% and Condition)
model_baseline <- lmer(FG_z ~ SWS_z + conditions + (1 | id), data = df.regression)
# Model 2: The full model (PCA for travelling features)
model_full_SWSPCA <- lmer(FG_z ~ SWS_z + conditions + PC1_AllFeatures + (1 | id), data = df.regression)
#model_full_SWSpeakdur <- lmer(FG_z ~ SWS_z + conditions + travelpeakdur_z + (1 | id), data = df.regression)
#model_full_SWSmaxdist <- lmer(FG_z ~ SWS_z + conditions + traveldistancemax_z + (1 | id), data = df.regression)
#model_full_SWSclustersize <- lmer(FG_z ~ SWS_z + conditions + clustersize_z + (1 | id), data = df.regression)
# Compare the two models
anova(model_baseline, model_full_SWSPCA)
summary(model_full_SWSPCA)
# assumption checks for this model
plot(model_full_SWSPCA) # Look for a random cloud. No "funnel" shape.
qqnorm(resid(model_full_SWSPCA)) # Points should hug the diagonal line.
qqline(resid(model_full_SWSPCA))
# Quick visual check
library(ggplot2)
ggplot(df.regression, aes(x = SWS_z, y = FG_z, color = conditions)) +
# Jitter points to avoid overlap
geom_point(position = position_jitterdodge(jitter.width = 0.5, dodge.width = 0), size = 2.5, alpha = 0.8) +
# Add regression lines separately for each condition
geom_smooth(method = "lm", se = FALSE, linewidth = 1) +
# Improve aesthetics
theme_bw() +
scale_color_manual(values = c("sham" = color_sham, "stim" = color_stim)) +
labs(x = "Percentage of Slow Wave Sleep", y = "Memory Outcome", color = "Condition")
# Check correlation between predictors (excluding the categorical Condition)
cor(df.regression[, c("SWS_z", "PC1_AllFeatures")], use = "complete.obs")
# Fit your full model (the one you want to check)
# Note: VIF is calculated on the fixed effects structure
lme_full <- lmer(FG_z ~ SWS_z + PC1_AllFeatures + conditions + (1 | id), data = df.regression)
# Calculate VIF
vif(lme_full)
#### end ####
#### Side: SWS & individual TSW features ####
# Model 2: The full model (PCA for travelling features)
model_full_SWSpeakdur <- lmer(FG_z ~ SWS_z + conditions + travelpeakdur_z + (1 | id), data = df.regression)
model_full_SWSmaxdist <- lmer(FG_z ~ SWS_z + conditions + traveldistancemax_z + (1 | id), data = df.regression)
model_full_SWSclustersize <- lmer(FG_z ~ SWS_z + conditions + clustersize_z + (1 | id), data = df.regression)
# Compare the two models SWSpeakdur
anova(model_baseline, model_full_SWSpeakdur)
summary(model_full_SWSpeakdur)
# assumption checks for this model
# Check correlation between predictors (excluding the categorical Condition)
cor(df.regression[, c("SWS_z", "travelpeakdur_z")], use = "complete.obs")
# Fit your full model (the one you want to check)
# Note: VIF is calculated on the fixed effects structure
lme_full <- lmer(FG_z ~ SWS_z + travelpeakdur_z + conditions + (1 | id), data = df.regression)
# Calculate VIF
vif(lme_full)
# Compare the two models SWSmax dist
anova(model_baseline, model_full_SWSmaxdist)
summary(model_full_SWSmaxdist)
# assumption checks for this model
# Check correlation between predictors (excluding the categorical Condition)
cor(df.regression[, c("SWS_z", "traveldistancemax_z")], use = "complete.obs")
# Fit your full model (the one you want to check)
# Note: VIF is calculated on the fixed effects structure
lme_full <- lmer(FG_z ~ SWS_z + traveldistancemax_z + conditions + (1 | id), data = df.regression)
# Calculate VIF
vif(lme_full)
# Compare the two models SWSclustersize
anova(model_baseline, model_full_SWSclustersize)
summary(model_full_SWSclustersize)
# assumption checks for this model
# Check correlation between predictors (excluding the categorical Condition)
cor(df.regression[, c("SWS_z", "clustersize_z")], use = "complete.obs")
# Fit your full model (the one you want to check)
# Note: VIF is calculated on the fixed effects structure
lme_full <- lmer(FG_z ~ SWS_z + clustersize_z + conditions + (1 | id), data = df.regression)
# Calculate VIF
vif(lme_full)
#### end ####
#### Linear mixed-effects model (LMM) using lme4: Part 2 N2 model ####
# Model 1: The baseline (Control for N2% and Condition)
model_baseline <- lmer(FG_z ~ N2_z + conditions + (1 | id), data = df.regression)
# Model 2: The full model (PCA for travelling features)
model_full_N2PCA <- lmer(FG_z ~ N2_z + conditions + PC1_AllFeatures + (1 | id), data = df.regression)
# Compare the two models
anova(model_baseline, model_full_N2PCA)
summary(model_full_N2PCA)
# assumption checks for this model
plot(model_full_N2PCA) # Look for a random cloud. No "funnel" shape.
qqnorm(resid(model_full_N2PCA)) # Points should hug the diagonal line.
qqline(resid(model_full_N2PCA))
# Quick visual check
library(ggplot2)
ggplot(df.regression, aes(x = N2_z, y = FG_z, color = conditions)) +
# Jitter points to avoid overlap
geom_point(position = position_jitterdodge(jitter.width = 0.5, dodge.width = 0), size = 2.5, alpha = 0.8) +
# Add regression lines separately for each condition
geom_smooth(method = "lm", se = FALSE, linewidth = 1) +
# Improve aesthetics
theme_bw() +
scale_color_manual(values = c("sham" = color_sham, "stim" = color_stim)) +
labs(x = "Percentage of N2 Sleep", y = "Memory Outcome", color = "Condition")
# Check correlation between predictors (excluding the categorical Condition)
cor(df.regression[, c("N2_z", "PC1_AllFeatures")], use = "complete.obs")
# Fit your full model (the one you want to check)
# Note: VIF is calculated on the fixed effects structure
lme_full <- lmer(FG_z ~ N2_z + PC1_AllFeatures + conditions + (1 | id), data = df.regression)
# Calculate VIF
vif(lme_full)
#### end ####
#### socount LMM model ####
# Model 1: The baseline (Control for socount% and Condition)
model_baseline <- lmer(FG_z ~ socount_z + conditions + (1 | id), data = df.regression)
# Model 2: The full model (PCA for travelling features)
#model_full_socountpeakdur <- lmer(FG_z ~ socount_z + conditions + travelpeakdur_z + (1 | id), data = df.regression)
#model_full_socountmaxdist <- lmer(FG_z ~ socount_z + conditions + traveldistancemax_z + (1 | id), data = df.regression)
#model_full_socountclustersize <- lmer(FG_z ~ socount_z + conditions + clustersize_z + (1 | id), data = df.regression)
model_full_socountPCA <- lmer(FG_z ~ socount_z + conditions + PC1_AllFeatures + (1 | id), data = df.regression)
# Compare the two models
anova(model_baseline, model_full_socountPCA)
summary(model_full_socountPCA)
# assumption checks for this model
plot(model_full_socountPCA) # Look for a random cloud. No "funnel" shape.
qqnorm(resid(model_full_socountPCA)) # Points should hug the diagonal line.
qqline(resid(model_full_socountPCA))
# Quick visual check
library(ggplot2)
ggplot(df.regression, aes(x = socount_z, y = FG_z, color = conditions)) +
# Jitter points to avoid overlap
geom_point(position = position_jitterdodge(jitter.width = 0.5, dodge.width = 0), size = 2.5, alpha = 0.8) +
# Add regression lines separately for each condition
geom_smooth(method = "lm", se = FALSE, linewidth = 1) +
# Improve aesthetics
theme_bw() +
scale_color_manual(values = c("sham" = color_sham, "stim" = color_stim)) +
labs(x = "Percentage of socount Sleep", y = "Memory Outcome", color = "Condition")
# Check correlation between predictors (excluding the categorical Condition)
cor(df.regression[, c("socount_z", "PC1_AllFeatures")], use = "complete.obs")
# Fit your full model (the one you want to check)
# Note: VIF is calculated on the fixed effects structure
lme_full <- lmer(FG_z ~ socount_z + PC1_AllFeatures + conditions + (1 | id), data = df.regression)
# Calculate VIF
vif(lme_full)
#### end ####
##### mediation #####
# create data frame and mutate conditions
df.mediation <- df.regression %>%
mutate(
conditions = ifelse(conditions == "stim", 1, 0)
)
# --- Install/load packages ---
if (!require(glmnet)) install.packages("glmnet")
if (!require(lavaan)) install.packages("lavaan")
# initiate libraries
library(glmnet)
library(lavaan)
# Define the model: SWS -> PC1 -> Memory
model_sws_med <- '
# Path A: SWS % predicts Features (PC1)
PC1_AllFeatures ~ a * SWS_z + conditions
# Path B & C: Features and SWS predict Memory
# b = effect of PC1 on Memory (controlling for SWS)
# c_prime = direct effect of SWS on Memory (not via PC1)
FG_z ~ c_prime * SWS_z + b * PC1_AllFeatures + conditions
# Indirect Effect: SWS -> PC1 -> Memory
indirect := a * b
# Total Effect of SWS on Memory
total := c_prime + (a * b)
'
# Fit the model
# Cluster by ParticipantID to handle repeated measures
fit_sws <- sem(model_sws_med, data = df.mediation, cluster = "id", se = "bootstrap", bootstrap = 5000)
summary(fit_sws, standardized = TRUE, ci = TRUE)
# N2 mediation
# Define the model: N2 -> PC1 -> Memory
model_N2_med <- '
# Path A: N2 % predicts Features (PC1)
PC1_AllFeatures ~ a * N2_z + conditions
# Path B & C: Features and N2 predict Memory
# b = effect of PC1 on Memory (controlling for N2)
# c_prime = direct effect of N2 on Memory (not via PC1)
FG_z ~ c_prime * N2_z + b * PC1_AllFeatures + conditions
# Indirect Effect: N2 -> PC1 -> Memory
indirect := a * b
# Total Effect of N2 on Memory
total := c_prime + (a * b)
'
# Fit the model
# Cluster by ParticipantID to handle repeated measures
fit_N2 <- sem(model_N2_med, data = df.mediation, cluster = "id", se = "bootstrap", bootstrap = 5000)
summary(fit_N2, standardized = TRUE, ci = TRUE)
# so count mediation
# Define the model: socount -> PC1 -> Memory
model_socount_med <- '
# Path A: socount % predicts Features (PC1)
PC1_AllFeatures ~ a * socount_z + conditions
# Path B & C: Features and socount predict Memory
# b = effect of PC1 on Memory (controlling for socount)
# c_prime = direct effect of socount on Memory (not via PC1)
FG_z ~ c_prime * socount_z + b * PC1_AllFeatures + conditions
# Indirect Effect: socount -> PC1 -> Memory
indirect := a * b
# Total Effect of socount on Memory
total := c_prime + (a * b)
'
# Fit the model
# Cluster by ParticipantID to handle repeated measures
fit_socount <- sem(model_socount_med, data = df.mediation, cluster = "id", se = "bootstrap", bootstrap = 5000)
summary(fit_socount, standardized = TRUE, ci = TRUE)
#### end ####
#### SWS and individual features mediation ####
# Define the model: SWS -> peak duration -> Memory
model_sws_peakdur_med <- '
# Path A: SWS % predicts Features
travelpeakdur_z ~ a * SWS_z + conditions
# Path B & C: Features and SWS predict Memory
# b = effect of features on Memory (controlling for SWS)
# c_prime = direct effect of SWS on Memory (not via features)
FG_z ~ c_prime * SWS_z + b * travelpeakdur_z + conditions
# Indirect Effect: SWS -> features -> Memory
indirect := a * b
# Total Effect of SWS on Memory
total := c_prime + (a * b)
'
# Fit the model
# Cluster by ParticipantID to handle repeated measures
fit_sws_peakdur <- sem(model_sws_peakdur_med, data = df.mediation, cluster = "id", se = "bootstrap", bootstrap = 5000)
summary(fit_sws_peakdur, standardized = TRUE, ci = TRUE)
# Define the model: SWS -> distmax -> Memory
model_sws_distmax_med <- '
# Path A: SWS % predicts Features
traveldistancemax_z ~ a * SWS_z + conditions
# Path B & C: Features and SWS predict Memory
# b = effect of features on Memory (controlling for SWS)
# c_prime = direct effect of SWS on Memory (not via features)
FG_z ~ c_prime * SWS_z + b * traveldistancemax_z + conditions
# Indirect Effect: SWS -> features -> Memory
indirect := a * b
# Total Effect of SWS on Memory
total := c_prime + (a * b)
'
# Fit the model
# Cluster by ParticipantID to handle repeated measures
fit_sws_distmax <- sem(model_sws_distmax_med, data = df.mediation, cluster = "id", se = "bootstrap", bootstrap = 5000)
summary(fit_sws_distmax, standardized = TRUE, ci = TRUE)
# Define the model: SWS -> clustersize -> Memory
model_sws_clustersize_med <- '
# Path A: SWS % predicts Features
clustersize_z ~ a * SWS_z + conditions
# Path B & C: Features and SWS predict Memory
# b = effect of features on Memory (controlling for SWS)
# c_prime = direct effect of SWS on Memory
FG_z ~ c_prime * SWS_z + b * clustersize_z + conditions
# Indirect Effect: SWS -> features -> Memory
indirect := a * b
# Total Effect of SWS on Memory
total := c_prime + (a * b)
'
# Fit the model
# Cluster by ParticipantID to handle repeated measures
fit_sws_clustersize <- sem(model_sws_clustersize_med, data = df.mediation, cluster = "id", se = "bootstrap", bootstrap = 5000)
summary(fit_sws_clustersize, standardized = TRUE, ci = TRUE)
#### end ####

ana25_correlations_behaviour_PCA_LMM_mediation.R, no license · at the source

Overview

Authors: Nora M Roüast¹, Deniz Kumral^1,2, Steffen Gais³, Monika Schönauer^1,2,4

ORCID iDs: Nora M Roüast

Institute of Psychology, Neuropsychology, University of Freiburg, Freiburg im Breisgau, Germany
BrainLinks-BrainTools, University of Freiburg, Freiburg im Breisgau, Germany
Institute of Medical Psychology and Behavioral Neurobiology, University of Tübingen, Tübingen, Germany
Bernstein Center Freiburg, University of Freiburg, Freiburg im Breisgau, Germany

Institutions: University of Freiburg (Germany); University of Tübingen (Germany); Bernstein Center for Computational Neuroscience Freiburg (Germany)

Journal: iScience, volume 29, issue 7, article 116601

Dates: received 5 December 2025; accepted 11 June 2026; published online 25 June 2026

Type: Research article · Language: English

License: CC BY

Identifiers: DOI 10.1016/j.isci.2026.116601 · PMID 42491684 · PMCID PMC13378391 · OpenAlex W7165893356

Open access: gold, a free copy (OpenAlex)

Status: code verified

Categories: cognitive (subfield)

Methods: Spectral & time-frequency, Preprocessing, Connectivity, Statistics, Smoothing, state filtering, decompositions, Evoked potentials, fMRI & imaging, Physiology & signal measures

Keywords: sleep, memory consolidation, auditory stimulation, traveling waves, SWS, declarative memory

Topic: Sleep and Wakefulness Research (Cognitive Neuroscience, Neuroscience), according to OpenAlex

Funding: Bundesministerium für Forschung, Technologie und Raumfahrt; German Research Foundation (RO6828/1-1, RO6828/1\u20132, GA730/3-1, SCHO1820/2\u20131)

Citations: not cited yet (Europe PMC); 77 references in the paper

Research resources: RStudio 2023.12.1 + 420 RRID:SCR_000432, MATLAB R2020b RRID:SCR_001622, R 4.5.3 (2026-03-11) RRID:SCR_001905, FieldTrip Version: 20230418 RRID:SCR_004849, EEGLAB version 2023.0 RRID:SCR_007292, Brain Products RRID:SCR_009443, lme4 RRID:SCR_015654

Abstract

Sleep plays a crucial role in memory consolidation, and various methods have attempted to enhance this process by using auditory stimulation. However, the broader impact of auditory stimulation on sleep physiology and memory retention is not fully understood. Here, we apply random or sham auditory stimulation during an afternoon nap to investigate its effects on neural activity and memory consolidation. Stimulation led to a specific reduction in slow-wave sleep, a decreased slow-wave count, and impaired declarative recall that correlated with the diminished sleep depth. Furthermore, we observed altered slow-wave traveling dynamics, with stimulation resulting in shorter traveling trajectories with less reach and reduced spatial spread, particularly impacting frontal regions. Disruptions in these features further predicted memory deficits. Our findings highlight the role of dynamic slow-wave properties in declarative memory consolidation and reveal some methodological limitations for using auditory cues during sleep, since they may disrupt complex processing patterns.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 20 matches between paragraphs and lines of code.

OSF wjzs9

License: none: the authors keep all their rights

State: the link answers, verified on 27 September 2026

Evidence: files inventoried

Languages: MATLAB (60), R (7)

Size: 110 files, 67 scripts

Software Heritage: not checked

Found in: “Data and code availability”

Not found: README, license file, CITATION.cff, environment file, tests, continuous integration, documentation

Tools: FieldTrip (40 files), Statistics and Machine Learning Toolbox (9 files), rstatix (7 files), tidyverse (7 files), CircStat (4 files), Signal Processing Toolbox (3 files), ggplot2 (2 files), ggpubr (2 files), car (1 file), glmnet (1 file), lavaan (1 file), lme4 (1 file), lmerTest (1 file)

Availability: 1 check, the latest on 27 September 2026: the link answers (HTTP 200)

27 September 2026: the link answers (HTTP 200)

67 files

analysis/SchlafEin.m, MATLAB, 1,259 lines
analysis/ana03_artefact.m, MATLAB, 78 lines
analysis/ana04_timefreq_analysis.m, MATLAB, 44 lines
analysis/ana05_timefrequency_conditions.m, MATLAB, 202 lines
analysis/ana06_timefrequency_statistics.m, MATLAB, 96 lines
analysis/ana07_timefrequency_plotting.m, MATLAB, 198 lines
analysis/ana08_frequency_conditions.m, MATLAB, 295 lines
analysis/ana09_frequency_statistics.m, MATLAB, 94 lines, 1 match
analysis/ana10_sleep_architecture.m, MATLAB, 231 lines, 1 match
analysis/ana10_sleep_architecture_stats.R, R, 248 lines
analysis/ana11_spindle_SO_detection.m, MATLAB, 307 lines, 1 match
analysis/ana12_spindle_SO_condition.m, MATLAB, 104 lines
analysis/ana13_spindle_SO_count.m, MATLAB, 278 lines
analysis/ana13_spindle_SO_count_stats.R, R, 190 lines
analysis/ana14_clusterSOs.m, MATLAB, 100 lines
analysis/ana15_travelSOs.m, MATLAB, 300 lines, 1 match
analysis/ana15_travelSOs_plottingexample.m, MATLAB, 481 lines
analysis/ana16_TSW_analysis.m, MATLAB, 474 lines
analysis/ana16_TSW_analysis_stats.R, R, 125 lines
analysis/ana16_TSW_analysis_stats_SWS.R, R, 134 lines, 1 match
analysis/ana17_TSW_electrodes.m, MATLAB, 608 lines, 1 match
analysis/ana18_spindle_frequency.m, MATLAB, 308 lines, 2 matches
analysis/ana19_spindle_frequency_mean.m, MATLAB, 125 lines
analysis/ana20_spindle_features.m, MATLAB, 125 lines
analysis/ana20_spindle_features_stats.R, R, 118 lines
analysis/ana21_ERP_analysis.m, MATLAB, 244 lines
analysis/ana22_ERP_stats.m, MATLAB, 758 lines, 1 match
analysis/ana23_behavioural_readin.m, MATLAB, 348 lines, 1 match
analysis/ana24_behavioural_mean.m, MATLAB, 340 lines, 1 match
analysis/ana25_correlations_behaviour_PCA_LMM_mediation.R, R, 1,327 lines, 3 matches
analysis/ana26_behaviouralbykcomplex.m, MATLAB, 135 lines
analysis/ana26_behaviouralbykcomplex_stats.R, R, 215 lines
analysis/ana27_hearingthreshold_calculation.m, MATLAB, 108 lines, 2 matches
analysis/ana28_triggertiming.m, MATLAB, 44 lines
analysis/append_participants_derivatives_fieldtrip.m, MATLAB, 74 lines
analysis/artifact_rejection.m, MATLAB, 278 lines
analysis/average_reference.m, MATLAB, 7 lines
analysis/cluster_SOs.m, MATLAB, 87 lines
analysis/config_artreject_trial.m, MATLAB, 9 lines
analysis/config_file_paths.m, MATLAB, 41 lines
analysis/coupling_analysis/average_across_channels.m, MATLAB, 102 lines
analysis/coupling_analysis/prep1.m, MATLAB, 176 lines, 1 match
analysis/coupling_analysis/prep2.m, MATLAB, 216 lines
analysis/coupling_analysis/statistics_v2.m, MATLAB, 56 lines
analysis/coupling_analysis/statistics_v3.m, MATLAB, 121 lines
analysis/coupling_analysis/statistics_v4.m, MATLAB, 421 lines, 1 match
analysis/detectSOs.m, MATLAB, 203 lines
analysis/detectSpindles.m, MATLAB, 160 lines, 1 match
analysis/gather_EOG_artifacts.m, MATLAB, 47 lines
analysis/gather_jump_artifacts.m, MATLAB, 27 lines
analysis/gather_muscle_artifacts.m, MATLAB, 28 lines
analysis/get_layout.m, MATLAB, 46 lines
analysis/get_neighbour_structure.m, MATLAB, 111 lines
analysis/get_stage_data.m, MATLAB, 12 lines
analysis/inspect_current_stage.m, MATLAB, 22 lines
analysis/interpolate.m, MATLAB, 37 lines
analysis/interpolate_and_rereference.m, MATLAB, 105 lines
analysis/interpolate_and_rereference_mastoid.m, MATLAB, 131 lines
analysis/plot_interpolation.m, MATLAB, 26 lines
analysis/plot_timefrequency.m, MATLAB, 61 lines
analysis/raincloudplots.m, MATLAB, 164 lines
analysis/rawdata_2_bids.m, MATLAB, 107 lines
analysis/signal_processing.m, MATLAB, 218 lines, 1 match
analysis/trialbased_frequencyanalysis_whole.m, MATLAB, 188 lines
analysis/trialbased_timefrequencyanalysis.m, MATLAB, 70 lines
analysis/trialfun_5sec_trials.m, MATLAB, 41 lines
analysis/trialfun_nap.m, MATLAB, 14 lines

At the source: doi.org/10.17605/OSF.IO/WJZS9

The paper's code and data availability statement is in the Data section.

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
67 scripts, each with its path and the digest of its content;
20 matches between paragraphs of the paper and lines of the code (method lexical-v1);
neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data and code availability

All data necessary to replicate the findings reported in this paper have been deposited at the Open Science Framework: https://doi.org/10.17605/OSF.IO/WJZS9 and FreiData: https://doi.org/10.60493/8t6p9-xfq49.

Transcribed behavioral data reported in this paper have been deposited at the Open Science Framework: https://doi.org/10.17605/OSF.IO/WJZS9.

All original code has been deposited at the Open Science Framework: https://doi.org/10.17605/OSF.IO/WJZS9.

Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 2, 28 September 2026

Authors: added Nora M Roüast (0000-0002-4112-358X); removed Nora M Roüast

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 4 authors, 6 keywords, 2 funders, 70 references, 7 RRIDs.

Cite

This paper

Roüast, N. M., Kumral, D., Gais, S., & Schönauer, M. (2026). Random auditory stimulation during sleep disturbs traveling slow waves and declarative memory. iScience, 29(7), 116601. https://doi.org/10.1016/j.isci.2026.116601

BibTeX

@article{rouast2026random, author = {Roüast, Nora M and Kumral, Deniz and Gais, Steffen and Schönauer, Monika}, title = {{Random auditory stimulation during sleep disturbs traveling slow waves and declarative memory}}, journal = {iScience}, year = {2026}, month = jun, volume = {29}, number = {7}, pages = {116601}, publisher = {Elsevier}, issn = {2589-0042}, doi = {10.1016/j.isci.2026.116601}, url = {https://doi.org/10.1016/j.isci.2026.116601}, pmid = {42491684}, pmcid = {PMC13378391} }

RIS

TY - JOUR AU - Roüast, Nora M AU - Kumral, Deniz AU - Gais, Steffen AU - Schönauer, Monika TI - Random auditory stimulation during sleep disturbs traveling slow waves and declarative memory T2 - iScience J2 - iScience PY - 2026 DA - 2026/06/25 VL - 29 IS - 7 SP - 116601 SN - 2589-0042 PB - Elsevier DO - 10.1016/j.isci.2026.116601 UR - https://doi.org/10.1016/j.isci.2026.116601 LA - en ER -

CSL-JSON

{ "id": "10.1016/j.isci.2026.116601", "type": "article-journal", "title": "Random auditory stimulation during sleep disturbs traveling slow waves and declarative memory", "container-title": "iScience", "author": [ { "family": "Roüast", "given": "Nora M" }, { "family": "Kumral", "given": "Deniz" }, { "family": "Gais", "given": "Steffen" }, { "family": "Schönauer", "given": "Monika" } ], "container-title-short": "iScience", "volume": "29", "issue": "7", "page": "116601", "DOI": "10.1016/j.isci.2026.116601", "PMID": "42491684", "PMCID": "PMC13378391", "ISSN": "2589-0042", "publisher": "Elsevier", "URL": "https://doi.org/10.1016/j.isci.2026.116601", "language": "en", "issued": { "date-parts": [ [ 2026, 6, 25 ] ] } }

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.