thesis_final_analysis
2024-06-10
##Packages
library(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.3 ✔ tidyr 1.3.1
## ✔ purrr 1.0.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorslibrary(readxl)
library(dplyr)
library(purrr)
library(lme4)## Loading required package: Matrix
##
## Attaching package: 'Matrix'
##
## The following objects are masked from 'package:tidyr':
##
## expand, pack, unpacklibrary(sjPlot)## Learn more about sjPlot with 'browseVignettes("sjPlot")'.library(lmerTest)##
## Attaching package: 'lmerTest'
##
## The following object is masked from 'package:lme4':
##
## lmer
##
## The following object is masked from 'package:stats':
##
## steplibrary(broom)
library(ggeffects)
library(emmeans)
library(ggsignif)
library(nlme)##
## Attaching package: 'nlme'
##
## The following object is masked from 'package:lme4':
##
## lmList
##
## The following object is masked from 'package:dplyr':
##
## collapselibrary(sjPlot)
library(emmeans)
library(effects)## Loading required package: carData
## lattice theme set by effectsTheme()
## See ?effectsTheme for details.library(car)##
## Attaching package: 'car'
##
## The following object is masked from 'package:dplyr':
##
## recode
##
## The following object is masked from 'package:purrr':
##
## somelibrary(corrplot)## corrplot 0.92 loadedlibrary(plotly)##
## Attaching package: 'plotly'
##
## The following object is masked from 'package:ggplot2':
##
## last_plot
##
## The following object is masked from 'package:stats':
##
## filter
##
## The following object is masked from 'package:graphics':
##
## layout#Load datasets
setwd("C:/Users/Marcel/Desktop/Bachelor Thesis/Data Analysis/Relevant R Scripts/Movement Period Data (27 - Last-Step +65)/Hurst Exponents")
# Load and filter the trial_data
execution_data <- read_excel("trial_data.xlsx") %>%
filter(Subject != 18) # Exclude Participant 18
# Function to filter outliers
filter_outliers <- function(data) {
data %>%
group_by(Subject, Block) %>%
mutate(mean_rt = mean(trial_total_rt, na.rm = TRUE),
sd_rt = sd(trial_total_rt, na.rm = TRUE)) %>%
filter(abs(trial_total_rt - mean_rt) <= 2.5 * sd_rt) %>%
dplyr::select(-mean_rt, -sd_rt)
}
# Apply the outlier filter function to the data
execution_data_filtered <- filter_outliers(execution_data)
execution_data$Block <- factor(execution_data$Block, ordered = TRUE, levels = c('1','2','3','4','5','6','7','8','9','10'))
execution_data$Trial <- as.numeric(execution_data$Trial)
# Filter data to only include accurate trials
execution_data_acc <- execution_data %>% filter(trial_acc == 1)#Load datasets
setwd("C:/Users/Marcel/Desktop/Bachelor Thesis/Data Analysis/Relevant R Scripts/Still Period Data/Hurst Exponents")
# Load and filter the trial_data
preparation_data <- read_excel("trial_data.xlsx") %>%
filter(Subject != 18) # Exclude Participant 18
# Function to filter outliers
filter_outliers <- function(data) {
data %>%
group_by(Subject, Block) %>%
mutate(mean_rt = mean(trial_total_rt, na.rm = TRUE),
sd_rt = sd(trial_total_rt, na.rm = TRUE)) %>%
filter(abs(trial_total_rt - mean_rt) <= 2.5 * sd_rt) %>%
dplyr::select(-mean_rt, -sd_rt)
}
# Apply the outlier filter function to the data
preparation_data_filtered <- filter_outliers(preparation_data)
preparation_data$Block <- factor(preparation_data$Block, ordered = TRUE, levels = c('1','2','3','4','5','6','7','8','9','10'))
preparation_data$Trial <- as.numeric(preparation_data$Trial)
# Filter data to only include accurate trials
preparation_data_acc <- preparation_data %>% filter(trial_acc == 1)Analysis Across All Participants: Movement Preparation Data
base_size <- 14
# Acceleration X Hurst Exponents over Blocks
model_hurst_acceleration_x <- lmer(hurst_acceleration_x ~ Block + (1|Subject) + (1|Trial), data = preparation_data)
Anova(model_hurst_acceleration_x)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: hurst_acceleration_x
## Chisq Df Pr(>Chisq)
## Block 85.157 9 1.518e-14 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Obtain effects of the model
model_effects <- allEffects(model_hurst_acceleration_x)
model_effects_df <- as.data.frame(model_effects[[1]])
# Change conf interval to 83% to match p = .05
model_effects_df$l83 <- model_effects_df$fit - 1.3722 * model_effects_df$se
model_effects_df$u83 <- model_effects_df$fit + 1.3722 * model_effects_df$se
ggplot(model_effects_df, aes(x = Block, y = fit)) +
geom_errorbar(aes(ymin = l83, ymax = u83), width = 0.2, size = 1, color = "darkorchid4") +
geom_point(color = "darkorchid2", size = 4) +
geom_vline(xintercept = 8.5, color = "red", size = 1) +
xlab("Block") + ylab("Hurst Exponent") +
ggtitle("Movement Preperation Period") +
labs(subtitle = "Hurst Acceleration X ~ Block + (1|Subject) + (1|Trial)") +
scale_y_continuous(limits = c(0.75, 1)) +
theme_minimal() +
theme(text = element_text(size = base_size),
axis.title = element_text(size = base_size + 2),
plot.title = element_text(size = base_size + 4, face = "bold"),
plot.subtitle = element_text(size = base_size + 2),
axis.text = element_text(size = base_size - 2),
axis.text.y = element_text(face = "bold"),
axis.text.x = element_text(face = "bold"))## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
# Post hocs
emm_blocks <- emmeans(model_hurst_acceleration_x, specs = ~ Block)## Note: D.f. calculations have been disabled because the number of observations exceeds 3000.
## To enable adjustments, add the argument 'pbkrtest.limit = 10560' (or larger)
## [or, globally, 'set emm_options(pbkrtest.limit = 10560)' or larger];
## but be warned that this may result in large computation time and memory use.## Note: D.f. calculations have been disabled because the number of observations exceeds 3000.
## To enable adjustments, add the argument 'lmerTest.limit = 10560' (or larger)
## [or, globally, 'set emm_options(lmerTest.limit = 10560)' or larger];
## but be warned that this may result in large computation time and memory use.comparisons <- pairs(emm_blocks)
print(comparisons)## contrast estimate SE df z.ratio p.value
## Block1 - Block2 0.00158 0.00483 Inf 0.326 1.0000
## Block1 - Block3 -0.00550 0.00483 Inf -1.139 0.9808
## Block1 - Block4 -0.01077 0.00483 Inf -2.229 0.4365
## Block1 - Block5 -0.00573 0.00483 Inf -1.186 0.9746
## Block1 - Block6 -0.01363 0.00483 Inf -2.820 0.1295
## Block1 - Block7 -0.02241 0.00483 Inf -4.636 0.0002
## Block1 - Block8 -0.02833 0.00483 Inf -5.862 <.0001
## Block1 - Block9 -0.02013 0.00483 Inf -4.166 0.0013
## Block1 - Block10 -0.02415 0.00483 Inf -4.997 <.0001
## Block2 - Block3 -0.00708 0.00483 Inf -1.465 0.9061
## Block2 - Block4 -0.01235 0.00483 Inf -2.555 0.2393
## Block2 - Block5 -0.00731 0.00483 Inf -1.512 0.8878
## Block2 - Block6 -0.01520 0.00483 Inf -3.146 0.0527
## Block2 - Block7 -0.02398 0.00483 Inf -4.962 <.0001
## Block2 - Block8 -0.02990 0.00483 Inf -6.188 <.0001
## Block2 - Block9 -0.02171 0.00483 Inf -4.492 0.0003
## Block2 - Block10 -0.02573 0.00483 Inf -5.323 <.0001
## Block3 - Block4 -0.00527 0.00483 Inf -1.091 0.9857
## Block3 - Block5 -0.00023 0.00483 Inf -0.048 1.0000
## Block3 - Block6 -0.00813 0.00483 Inf -1.682 0.8061
## Block3 - Block7 -0.01690 0.00483 Inf -3.498 0.0169
## Block3 - Block8 -0.02283 0.00483 Inf -4.723 0.0001
## Block3 - Block9 -0.01463 0.00483 Inf -3.027 0.0744
## Block3 - Block10 -0.01865 0.00483 Inf -3.859 0.0045
## Block4 - Block5 0.00504 0.00483 Inf 1.043 0.9896
## Block4 - Block6 -0.00286 0.00483 Inf -0.591 0.9999
## Block4 - Block7 -0.01163 0.00483 Inf -2.407 0.3212
## Block4 - Block8 -0.01756 0.00483 Inf -3.633 0.0105
## Block4 - Block9 -0.00936 0.00483 Inf -1.937 0.6436
## Block4 - Block10 -0.01338 0.00483 Inf -2.768 0.1473
## Block5 - Block6 -0.00790 0.00483 Inf -1.634 0.8315
## Block5 - Block7 -0.01667 0.00483 Inf -3.450 0.0199
## Block5 - Block8 -0.02260 0.00483 Inf -4.676 0.0001
## Block5 - Block9 -0.01440 0.00483 Inf -2.980 0.0850
## Block5 - Block10 -0.01842 0.00483 Inf -3.811 0.0054
## Block6 - Block7 -0.00878 0.00483 Inf -1.816 0.7252
## Block6 - Block8 -0.01470 0.00483 Inf -3.042 0.0715
## Block6 - Block9 -0.00650 0.00483 Inf -1.346 0.9429
## Block6 - Block10 -0.01052 0.00483 Inf -2.177 0.4727
## Block7 - Block8 -0.00592 0.00483 Inf -1.226 0.9685
## Block7 - Block9 0.00227 0.00483 Inf 0.470 1.0000
## Block7 - Block10 -0.00174 0.00483 Inf -0.361 1.0000
## Block8 - Block9 0.00820 0.00483 Inf 1.696 0.7982
## Block8 - Block10 0.00418 0.00483 Inf 0.865 0.9974
## Block9 - Block10 -0.00402 0.00483 Inf -0.831 0.9981
##
## Degrees-of-freedom method: asymptotic
## P value adjustment: tukey method for comparing a family of 10 estimates# Acceleration Y Hurst Exponents over Blocks
model_hurst_acceleration_y <- lmer(hurst_acceleration_y ~ Block + (1|Subject) + (1|Trial), data = preparation_data)
Anova(model_hurst_acceleration_y)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: hurst_acceleration_y
## Chisq Df Pr(>Chisq)
## Block 77.683 9 4.658e-13 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Obtain effects of the model
model_effects <- allEffects(model_hurst_acceleration_y)
model_effects_df <- as.data.frame(model_effects[[1]])
# Change conf interval to 83% to match p = .05
model_effects_df$l83 <- model_effects_df$fit - 1.3722 * model_effects_df$se
model_effects_df$u83 <- model_effects_df$fit + 1.3722 * model_effects_df$se
ggplot(model_effects_df, aes(x = Block, y = fit)) +
geom_errorbar(aes(ymin = l83, ymax = u83), width = 0.2, size = 1, color = "darkorchid4") +
geom_point(color = "darkorchid2", size = 4) +
geom_vline(xintercept = 8.5, color = "red", size = 1) +
xlab("Block") + ylab("Hurst Exponent") +
labs(subtitle = "Hurst Acceleration Y ~ Block + (1|Subject) + (1|Trial)") +
scale_y_continuous(limits = c(0.75, 1)) +
theme_minimal() +
theme(text = element_text(size = base_size),
axis.title = element_text(size = base_size + 2),
plot.title = element_text(size = base_size + 4, face = "bold", colour = NA),
plot.subtitle = element_text(size = base_size + 2),
axis.text = element_text(size = base_size - 2),
axis.text.y = element_text(face = "bold"),
axis.text.x = element_text(face = "bold"))
# Post hocs
emm_blocks <- emmeans(model_hurst_acceleration_y, specs = ~ Block)## Note: D.f. calculations have been disabled because the number of observations exceeds 3000.
## To enable adjustments, add the argument 'pbkrtest.limit = 10560' (or larger)
## [or, globally, 'set emm_options(pbkrtest.limit = 10560)' or larger];
## but be warned that this may result in large computation time and memory use.
## Note: D.f. calculations have been disabled because the number of observations exceeds 3000.
## To enable adjustments, add the argument 'lmerTest.limit = 10560' (or larger)
## [or, globally, 'set emm_options(lmerTest.limit = 10560)' or larger];
## but be warned that this may result in large computation time and memory use.comparisons <- pairs(emm_blocks)
print(comparisons)## contrast estimate SE df z.ratio p.value
## Block1 - Block2 0.005256 0.00558 Inf 0.941 0.9951
## Block1 - Block3 0.004658 0.00558 Inf 0.834 0.9981
## Block1 - Block4 0.001261 0.00558 Inf 0.226 1.0000
## Block1 - Block5 0.002059 0.00558 Inf 0.369 1.0000
## Block1 - Block6 -0.015102 0.00558 Inf -2.705 0.1715
## Block1 - Block7 -0.018289 0.00558 Inf -3.276 0.0354
## Block1 - Block8 -0.025410 0.00558 Inf -4.551 0.0002
## Block1 - Block9 -0.013695 0.00558 Inf -2.453 0.2945
## Block1 - Block10 -0.019502 0.00558 Inf -3.493 0.0172
## Block2 - Block3 -0.000598 0.00558 Inf -0.107 1.0000
## Block2 - Block4 -0.003994 0.00558 Inf -0.715 0.9994
## Block2 - Block5 -0.003197 0.00558 Inf -0.573 0.9999
## Block2 - Block6 -0.020358 0.00558 Inf -3.646 0.0100
## Block2 - Block7 -0.023545 0.00558 Inf -4.217 0.0010
## Block2 - Block8 -0.030666 0.00558 Inf -5.492 <.0001
## Block2 - Block9 -0.018951 0.00558 Inf -3.394 0.0241
## Block2 - Block10 -0.024758 0.00558 Inf -4.434 0.0004
## Block3 - Block4 -0.003397 0.00558 Inf -0.608 0.9998
## Block3 - Block5 -0.002599 0.00558 Inf -0.465 1.0000
## Block3 - Block6 -0.019760 0.00558 Inf -3.539 0.0147
## Block3 - Block7 -0.022947 0.00558 Inf -4.110 0.0016
## Block3 - Block8 -0.030068 0.00558 Inf -5.385 <.0001
## Block3 - Block9 -0.018353 0.00558 Inf -3.287 0.0341
## Block3 - Block10 -0.024160 0.00558 Inf -4.327 0.0006
## Block4 - Block5 0.000798 0.00558 Inf 0.143 1.0000
## Block4 - Block6 -0.016363 0.00558 Inf -2.931 0.0972
## Block4 - Block7 -0.019550 0.00558 Inf -3.501 0.0167
## Block4 - Block8 -0.026672 0.00558 Inf -4.777 0.0001
## Block4 - Block9 -0.014956 0.00558 Inf -2.679 0.1822
## Block4 - Block10 -0.020763 0.00558 Inf -3.719 0.0077
## Block5 - Block6 -0.017161 0.00558 Inf -3.074 0.0652
## Block5 - Block7 -0.020348 0.00558 Inf -3.644 0.0101
## Block5 - Block8 -0.027469 0.00558 Inf -4.920 <.0001
## Block5 - Block9 -0.015754 0.00558 Inf -2.822 0.1291
## Block5 - Block10 -0.021561 0.00558 Inf -3.862 0.0044
## Block6 - Block7 -0.003187 0.00558 Inf -0.571 0.9999
## Block6 - Block8 -0.010308 0.00558 Inf -1.846 0.7055
## Block6 - Block9 0.001407 0.00558 Inf 0.252 1.0000
## Block6 - Block10 -0.004400 0.00558 Inf -0.788 0.9988
## Block7 - Block8 -0.007121 0.00558 Inf -1.275 0.9592
## Block7 - Block9 0.004594 0.00558 Inf 0.823 0.9983
## Block7 - Block10 -0.001213 0.00558 Inf -0.217 1.0000
## Block8 - Block9 0.011715 0.00558 Inf 2.098 0.5287
## Block8 - Block10 0.005909 0.00558 Inf 1.058 0.9885
## Block9 - Block10 -0.005807 0.00558 Inf -1.040 0.9898
##
## Degrees-of-freedom method: asymptotic
## P value adjustment: tukey method for comparing a family of 10 estimates# Acceleration Z Hurst Exponents over Blocks
model_hurst_acceleration_z <- lmer(hurst_acceleration_z ~ Block + (1|Subject) + (1|Trial), data = preparation_data)
Anova(model_hurst_acceleration_z)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: hurst_acceleration_z
## Chisq Df Pr(>Chisq)
## Block 54.696 9 1.392e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Obtain effects of the model
model_effects <- allEffects(model_hurst_acceleration_z)
model_effects_df <- as.data.frame(model_effects[[1]])
# Change conf interval to 83% to match p = .05
model_effects_df$l83 <- model_effects_df$fit - 1.3722 * model_effects_df$se
model_effects_df$u83 <- model_effects_df$fit + 1.3722 * model_effects_df$se
ggplot(model_effects_df, aes(x = Block, y = fit)) +
geom_errorbar(aes(ymin = l83, ymax = u83), width = 0.2, size = 1, color = "darkorchid4") +
geom_point(color = "darkorchid2", size = 4) +
geom_vline(xintercept = 8.5, color = "red", size = 1) +
xlab("Block") + ylab("Hurst Exponent") +
labs(subtitle = "Hurst Acceleration Z ~ Block + (1|Subject) + (1|Trial)") +
scale_y_continuous(limits = c(0.75, 1)) +
theme_minimal() +
theme(text = element_text(size = base_size),
axis.title = element_text(size = base_size + 2),
plot.title = element_text(size = base_size + 4, face = "bold", colour = NA),
plot.subtitle = element_text(size = base_size + 2),
axis.text = element_text(size = base_size - 2),
axis.text.y = element_text(face = "bold"),
axis.text.x = element_text(face = "bold"))
# Post hocs
emm_blocks <- emmeans(model_hurst_acceleration_z, specs = ~ Block)## Note: D.f. calculations have been disabled because the number of observations exceeds 3000.
## To enable adjustments, add the argument 'pbkrtest.limit = 10560' (or larger)
## [or, globally, 'set emm_options(pbkrtest.limit = 10560)' or larger];
## but be warned that this may result in large computation time and memory use.
## Note: D.f. calculations have been disabled because the number of observations exceeds 3000.
## To enable adjustments, add the argument 'lmerTest.limit = 10560' (or larger)
## [or, globally, 'set emm_options(lmerTest.limit = 10560)' or larger];
## but be warned that this may result in large computation time and memory use.comparisons <- pairs(emm_blocks)
print(comparisons)## contrast estimate SE df z.ratio p.value
## Block1 - Block2 0.00904 0.0064 Inf 1.412 0.9238
## Block1 - Block3 0.00505 0.0064 Inf 0.789 0.9987
## Block1 - Block4 0.00642 0.0064 Inf 1.003 0.9922
## Block1 - Block5 0.01571 0.0064 Inf 2.454 0.2937
## Block1 - Block6 -0.00191 0.0064 Inf -0.298 1.0000
## Block1 - Block7 -0.00611 0.0064 Inf -0.955 0.9946
## Block1 - Block8 -0.01878 0.0064 Inf -2.935 0.0961
## Block1 - Block9 -0.00794 0.0064 Inf -1.241 0.9657
## Block1 - Block10 -0.01711 0.0064 Inf -2.673 0.1847
## Block2 - Block3 -0.00399 0.0064 Inf -0.623 0.9998
## Block2 - Block4 -0.00262 0.0064 Inf -0.410 1.0000
## Block2 - Block5 0.00667 0.0064 Inf 1.042 0.9897
## Block2 - Block6 -0.01095 0.0064 Inf -1.711 0.7898
## Block2 - Block7 -0.01515 0.0064 Inf -2.367 0.3457
## Block2 - Block8 -0.02782 0.0064 Inf -4.347 0.0006
## Block2 - Block9 -0.01698 0.0064 Inf -2.654 0.1930
## Block2 - Block10 -0.02614 0.0064 Inf -4.085 0.0018
## Block3 - Block4 0.00137 0.0064 Inf 0.214 1.0000
## Block3 - Block5 0.01066 0.0064 Inf 1.665 0.8153
## Block3 - Block6 -0.00696 0.0064 Inf -1.087 0.9860
## Block3 - Block7 -0.01116 0.0064 Inf -1.744 0.7701
## Block3 - Block8 -0.02383 0.0064 Inf -3.724 0.0075
## Block3 - Block9 -0.01300 0.0064 Inf -2.031 0.5772
## Block3 - Block10 -0.02216 0.0064 Inf -3.462 0.0192
## Block4 - Block5 0.00929 0.0064 Inf 1.451 0.9108
## Block4 - Block6 -0.00833 0.0064 Inf -1.301 0.9537
## Block4 - Block7 -0.01253 0.0064 Inf -1.958 0.6290
## Block4 - Block8 -0.02520 0.0064 Inf -3.938 0.0033
## Block4 - Block9 -0.01436 0.0064 Inf -2.244 0.4263
## Block4 - Block10 -0.02352 0.0064 Inf -3.676 0.0090
## Block5 - Block6 -0.01762 0.0064 Inf -2.752 0.1531
## Block5 - Block7 -0.02182 0.0064 Inf -3.409 0.0229
## Block5 - Block8 -0.03449 0.0064 Inf -5.389 <.0001
## Block5 - Block9 -0.02365 0.0064 Inf -3.696 0.0083
## Block5 - Block10 -0.03281 0.0064 Inf -5.127 <.0001
## Block6 - Block7 -0.00420 0.0064 Inf -0.657 0.9997
## Block6 - Block8 -0.01687 0.0064 Inf -2.637 0.2005
## Block6 - Block9 -0.00604 0.0064 Inf -0.943 0.9950
## Block6 - Block10 -0.01520 0.0064 Inf -2.375 0.3412
## Block7 - Block8 -0.01267 0.0064 Inf -1.980 0.6133
## Block7 - Block9 -0.00183 0.0064 Inf -0.286 1.0000
## Block7 - Block10 -0.01099 0.0064 Inf -1.718 0.7856
## Block8 - Block9 0.01084 0.0064 Inf 1.694 0.7995
## Block8 - Block10 0.00168 0.0064 Inf 0.262 1.0000
## Block9 - Block10 -0.00916 0.0064 Inf -1.432 0.9176
##
## Degrees-of-freedom method: asymptotic
## P value adjustment: tukey method for comparing a family of 10 estimatesAnalysis Across All Participants: Sequence Execution Data
base_size <- 14
# Acceleration X Hurst Exponents over Blocks
model_hurst_acceleration_x <- lmer(hurst_acceleration_x ~ Block + (1|Subject) + (1|Trial), data = execution_data)
Anova(model_hurst_acceleration_x)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: hurst_acceleration_x
## Chisq Df Pr(>Chisq)
## Block 316.73 9 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Obtain effects of the model
model_effects <- allEffects(model_hurst_acceleration_x)
model_effects_df <- as.data.frame(model_effects[[1]])
# Change conf interval to 83% to match p = .05
model_effects_df$l83 <- model_effects_df$fit - 1.3722 * model_effects_df$se
model_effects_df$u83 <- model_effects_df$fit + 1.3722 * model_effects_df$se
ggplot(model_effects_df, aes(x = Block, y = fit)) +
geom_errorbar(aes(ymin = l83, ymax = u83), width = 0.2, size = 1, color = "aquamarine4") +
geom_point(color = "aquamarine3", size = 4) +
geom_vline(xintercept = 8.5, color = "red", size = 1) +
xlab("Block") + ylab("Hurst Exponent") +
labs(subtitle = "Hurst Acceleration X ~ Block + (1|Subject) + (1|Trial)") +
scale_y_continuous(limits = c(0.75, 1)) +
theme_minimal() +
theme(text = element_text(size = base_size),
axis.title = element_text(size = base_size + 2),
plot.title = element_text(size = base_size + 4, face = "bold"),
plot.subtitle = element_text(size = base_size + 2),
axis.text = element_text(size = base_size - 2),
axis.text.y = element_text(face = "bold"),
axis.text.x = element_text(face = "bold"))
# Post hocs
emm_blocks <- emmeans(model_hurst_acceleration_x, specs = ~ Block)## Note: D.f. calculations have been disabled because the number of observations exceeds 3000.
## To enable adjustments, add the argument 'pbkrtest.limit = 10560' (or larger)
## [or, globally, 'set emm_options(pbkrtest.limit = 10560)' or larger];
## but be warned that this may result in large computation time and memory use.## Note: D.f. calculations have been disabled because the number of observations exceeds 3000.
## To enable adjustments, add the argument 'lmerTest.limit = 10560' (or larger)
## [or, globally, 'set emm_options(lmerTest.limit = 10560)' or larger];
## but be warned that this may result in large computation time and memory use.comparisons <- pairs(emm_blocks)
print(comparisons)## contrast estimate SE df z.ratio p.value
## Block1 - Block2 8.24e-03 0.00307 Inf 2.684 0.1798
## Block1 - Block3 1.56e-02 0.00307 Inf 5.094 <.0001
## Block1 - Block4 2.77e-02 0.00307 Inf 9.032 <.0001
## Block1 - Block5 2.70e-02 0.00307 Inf 8.784 <.0001
## Block1 - Block6 2.77e-02 0.00307 Inf 9.037 <.0001
## Block1 - Block7 3.51e-02 0.00307 Inf 11.429 <.0001
## Block1 - Block8 3.14e-02 0.00307 Inf 10.214 <.0001
## Block1 - Block9 3.59e-02 0.00307 Inf 11.681 <.0001
## Block1 - Block10 3.98e-02 0.00307 Inf 12.978 <.0001
## Block2 - Block3 7.40e-03 0.00307 Inf 2.409 0.3200
## Block2 - Block4 1.95e-02 0.00307 Inf 6.348 <.0001
## Block2 - Block5 1.87e-02 0.00307 Inf 6.100 <.0001
## Block2 - Block6 1.95e-02 0.00307 Inf 6.353 <.0001
## Block2 - Block7 2.68e-02 0.00307 Inf 8.745 <.0001
## Block2 - Block8 2.31e-02 0.00307 Inf 7.529 <.0001
## Block2 - Block9 2.76e-02 0.00307 Inf 8.997 <.0001
## Block2 - Block10 3.16e-02 0.00307 Inf 10.294 <.0001
## Block3 - Block4 1.21e-02 0.00307 Inf 3.938 0.0033
## Block3 - Block5 1.13e-02 0.00307 Inf 3.691 0.0085
## Block3 - Block6 1.21e-02 0.00307 Inf 3.943 0.0032
## Block3 - Block7 1.94e-02 0.00307 Inf 6.336 <.0001
## Block3 - Block8 1.57e-02 0.00307 Inf 5.120 <.0001
## Block3 - Block9 2.02e-02 0.00307 Inf 6.587 <.0001
## Block3 - Block10 2.42e-02 0.00307 Inf 7.884 <.0001
## Block4 - Block5 -7.60e-04 0.00307 Inf -0.248 1.0000
## Block4 - Block6 1.59e-05 0.00307 Inf 0.005 1.0000
## Block4 - Block7 7.36e-03 0.00307 Inf 2.397 0.3271
## Block4 - Block8 3.63e-03 0.00307 Inf 1.182 0.9752
## Block4 - Block9 8.13e-03 0.00307 Inf 2.649 0.1949
## Block4 - Block10 1.21e-02 0.00307 Inf 3.946 0.0032
## Block5 - Block6 7.76e-04 0.00307 Inf 0.253 1.0000
## Block5 - Block7 8.12e-03 0.00307 Inf 2.645 0.1967
## Block5 - Block8 4.39e-03 0.00307 Inf 1.429 0.9183
## Block5 - Block9 8.89e-03 0.00307 Inf 2.897 0.1063
## Block5 - Block10 1.29e-02 0.00307 Inf 4.194 0.0011
## Block6 - Block7 7.34e-03 0.00307 Inf 2.392 0.3303
## Block6 - Block8 3.61e-03 0.00307 Inf 1.176 0.9760
## Block6 - Block9 8.12e-03 0.00307 Inf 2.644 0.1972
## Block6 - Block10 1.21e-02 0.00307 Inf 3.941 0.0032
## Block7 - Block8 -3.73e-03 0.00307 Inf -1.216 0.9701
## Block7 - Block9 7.73e-04 0.00307 Inf 0.252 1.0000
## Block7 - Block10 4.75e-03 0.00307 Inf 1.549 0.8724
## Block8 - Block9 4.50e-03 0.00307 Inf 1.468 0.9050
## Block8 - Block10 8.49e-03 0.00307 Inf 2.764 0.1487
## Block9 - Block10 3.98e-03 0.00307 Inf 1.297 0.9546
##
## Degrees-of-freedom method: asymptotic
## P value adjustment: tukey method for comparing a family of 10 estimates# Acceleration Y Hurst Exponents over Blocks
model_hurst_acceleration_y <- lmer(hurst_acceleration_y ~ Block + (1|Subject) + (1|Trial), data = execution_data)
Anova(model_hurst_acceleration_y)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: hurst_acceleration_y
## Chisq Df Pr(>Chisq)
## Block 170.86 9 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Obtain effects of the model
model_effects <- allEffects(model_hurst_acceleration_y)
model_effects_df <- as.data.frame(model_effects[[1]])
# Change conf interval to 83% to match p = .05
model_effects_df$l83 <- model_effects_df$fit - 1.3722 * model_effects_df$se
model_effects_df$u83 <- model_effects_df$fit + 1.3722 * model_effects_df$se
ggplot(model_effects_df, aes(x = Block, y = fit)) +
geom_errorbar(aes(ymin = l83, ymax = u83), width = 0.2, size = 1, color = "aquamarine4") +
geom_point(color = "aquamarine3", size = 4) +
geom_vline(xintercept = 8.5, color = "red", size = 1) +
xlab("Block") + ylab("Hurst Exponent") +
labs(subtitle = "Hurst Acceleration Y ~ Block + (1|Subject) + (1|Trial)") +
scale_y_continuous(limits = c(0.75, 1)) +
theme_minimal() +
theme(text = element_text(size = base_size),
axis.title = element_text(size = base_size + 2),
plot.title = element_text(size = base_size + 4, face = "bold", colour = NA),
plot.subtitle = element_text(size = base_size + 2),
axis.text = element_text(size = base_size - 2),
axis.text.y = element_text(face = "bold"),
axis.text.x = element_text(face = "bold")) 
# Post hocs
emm_blocks <- emmeans(model_hurst_acceleration_y, specs = ~ Block)## Note: D.f. calculations have been disabled because the number of observations exceeds 3000.
## To enable adjustments, add the argument 'pbkrtest.limit = 10560' (or larger)
## [or, globally, 'set emm_options(pbkrtest.limit = 10560)' or larger];
## but be warned that this may result in large computation time and memory use.
## Note: D.f. calculations have been disabled because the number of observations exceeds 3000.
## To enable adjustments, add the argument 'lmerTest.limit = 10560' (or larger)
## [or, globally, 'set emm_options(lmerTest.limit = 10560)' or larger];
## but be warned that this may result in large computation time and memory use.comparisons <- pairs(emm_blocks)
print(comparisons)## contrast estimate SE df z.ratio p.value
## Block1 - Block2 -0.001647 0.00295 Inf -0.558 0.9999
## Block1 - Block3 0.007527 0.00295 Inf 2.551 0.2414
## Block1 - Block4 0.005733 0.00295 Inf 1.943 0.6393
## Block1 - Block5 0.010603 0.00295 Inf 3.594 0.0121
## Block1 - Block6 0.010351 0.00295 Inf 3.508 0.0163
## Block1 - Block7 0.021067 0.00295 Inf 7.140 <.0001
## Block1 - Block8 0.021634 0.00295 Inf 7.332 <.0001
## Block1 - Block9 0.024060 0.00295 Inf 8.154 <.0001
## Block1 - Block10 0.017490 0.00295 Inf 5.928 <.0001
## Block2 - Block3 0.009175 0.00295 Inf 3.109 0.0588
## Block2 - Block4 0.007380 0.00295 Inf 2.501 0.2675
## Block2 - Block5 0.012251 0.00295 Inf 4.152 0.0014
## Block2 - Block6 0.011998 0.00295 Inf 4.066 0.0019
## Block2 - Block7 0.022714 0.00295 Inf 7.698 <.0001
## Block2 - Block8 0.023281 0.00295 Inf 7.890 <.0001
## Block2 - Block9 0.025707 0.00295 Inf 8.712 <.0001
## Block2 - Block10 0.019137 0.00295 Inf 6.486 <.0001
## Block3 - Block4 -0.001794 0.00295 Inf -0.608 0.9999
## Block3 - Block5 0.003076 0.00295 Inf 1.042 0.9896
## Block3 - Block6 0.002824 0.00295 Inf 0.957 0.9945
## Block3 - Block7 0.013540 0.00295 Inf 4.589 0.0002
## Block3 - Block8 0.014107 0.00295 Inf 4.781 0.0001
## Block3 - Block9 0.016532 0.00295 Inf 5.603 <.0001
## Block3 - Block10 0.009963 0.00295 Inf 3.377 0.0255
## Block4 - Block5 0.004870 0.00295 Inf 1.651 0.8230
## Block4 - Block6 0.004618 0.00295 Inf 1.565 0.8651
## Block4 - Block7 0.015334 0.00295 Inf 5.197 <.0001
## Block4 - Block8 0.015901 0.00295 Inf 5.389 <.0001
## Block4 - Block9 0.018326 0.00295 Inf 6.211 <.0001
## Block4 - Block10 0.011757 0.00295 Inf 3.985 0.0027
## Block5 - Block6 -0.000252 0.00295 Inf -0.085 1.0000
## Block5 - Block7 0.010464 0.00295 Inf 3.546 0.0143
## Block5 - Block8 0.011031 0.00295 Inf 3.738 0.0071
## Block5 - Block9 0.013456 0.00295 Inf 4.560 0.0002
## Block5 - Block10 0.006887 0.00295 Inf 2.334 0.3667
## Block6 - Block7 0.010716 0.00295 Inf 3.632 0.0105
## Block6 - Block8 0.011283 0.00295 Inf 3.824 0.0051
## Block6 - Block9 0.013708 0.00295 Inf 4.646 0.0001
## Block6 - Block10 0.007139 0.00295 Inf 2.420 0.3139
## Block7 - Block8 0.000567 0.00295 Inf 0.192 1.0000
## Block7 - Block9 0.002992 0.00295 Inf 1.014 0.9915
## Block7 - Block10 -0.003577 0.00295 Inf -1.212 0.9706
## Block8 - Block9 0.002426 0.00295 Inf 0.822 0.9983
## Block8 - Block10 -0.004144 0.00295 Inf -1.404 0.9263
## Block9 - Block10 -0.006569 0.00295 Inf -2.226 0.4384
##
## Degrees-of-freedom method: asymptotic
## P value adjustment: tukey method for comparing a family of 10 estimates# Acceleration Z Hurst Exponents over Blocks
model_hurst_acceleration_z <- lmer(hurst_acceleration_z ~ Block + (1|Subject) + (1|Trial), data = execution_data)## boundary (singular) fit: see help('isSingular')Anova(model_hurst_acceleration_z)## Analysis of Deviance Table (Type II Wald chisquare tests)
##
## Response: hurst_acceleration_z
## Chisq Df Pr(>Chisq)
## Block 26.039 9 0.002013 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Obtain effects of the model
model_effects <- allEffects(model_hurst_acceleration_z)
model_effects_df <- as.data.frame(model_effects[[1]])
# Change conf interval to 83% to match p = .05
model_effects_df$l83 <- model_effects_df$fit - 1.3722 * model_effects_df$se
model_effects_df$u83 <- model_effects_df$fit + 1.3722 * model_effects_df$se
ggplot(model_effects_df, aes(x = Block, y = fit)) +
geom_errorbar(aes(ymin = l83, ymax = u83), width = 0.2, size = 1, color = "aquamarine4") +
geom_point(color = "aquamarine3", size = 4) +
geom_vline(xintercept = 8.5, color = "red", size = 1) +
xlab("Block") + ylab("Hurst Exponent") +
labs(subtitle = "Hurst Acceleration Z ~ Block + (1|Subject) + (1|Trial)") +
scale_y_continuous(limits = c(0.98, 1)) +
theme_minimal() +
theme(text = element_text(size = base_size),
axis.title = element_text(size = base_size + 2),
plot.title = element_text(size = base_size + 4, face = "bold", colour = NA),
plot.subtitle = element_text(size = base_size + 2),
axis.text = element_text(size = base_size - 2),
axis.text.y = element_text(face = "bold", color = "red"),
axis.text.x = element_text(face = "bold"))
# Post hocs
emm_blocks <- emmeans(model_hurst_acceleration_z, specs = ~ Block)## Note: D.f. calculations have been disabled because the number of observations exceeds 3000.
## To enable adjustments, add the argument 'pbkrtest.limit = 10560' (or larger)
## [or, globally, 'set emm_options(pbkrtest.limit = 10560)' or larger];
## but be warned that this may result in large computation time and memory use.
## Note: D.f. calculations have been disabled because the number of observations exceeds 3000.
## To enable adjustments, add the argument 'lmerTest.limit = 10560' (or larger)
## [or, globally, 'set emm_options(lmerTest.limit = 10560)' or larger];
## but be warned that this may result in large computation time and memory use.comparisons <- pairs(emm_blocks)
print(comparisons)## contrast estimate SE df z.ratio p.value
## Block1 - Block2 -0.003981 0.00126 Inf -3.157 0.0510
## Block1 - Block3 -0.003472 0.00126 Inf -2.753 0.1528
## Block1 - Block4 -0.001417 0.00126 Inf -1.123 0.9824
## Block1 - Block5 0.000530 0.00126 Inf 0.421 1.0000
## Block1 - Block6 -0.000695 0.00126 Inf -0.551 0.9999
## Block1 - Block7 -0.000852 0.00126 Inf -0.676 0.9996
## Block1 - Block8 -0.001033 0.00126 Inf -0.819 0.9983
## Block1 - Block9 -0.000135 0.00126 Inf -0.107 1.0000
## Block1 - Block10 0.000185 0.00126 Inf 0.147 1.0000
## Block2 - Block3 0.000509 0.00126 Inf 0.404 1.0000
## Block2 - Block4 0.002564 0.00126 Inf 2.033 0.5752
## Block2 - Block5 0.004512 0.00126 Inf 3.577 0.0128
## Block2 - Block6 0.003286 0.00126 Inf 2.606 0.2147
## Block2 - Block7 0.003129 0.00126 Inf 2.481 0.2785
## Block2 - Block8 0.002948 0.00126 Inf 2.338 0.3644
## Block2 - Block9 0.003846 0.00126 Inf 3.050 0.0698
## Block2 - Block10 0.004166 0.00126 Inf 3.303 0.0324
## Block3 - Block4 0.002055 0.00126 Inf 1.630 0.8338
## Block3 - Block5 0.004002 0.00126 Inf 3.174 0.0485
## Block3 - Block6 0.002777 0.00126 Inf 2.202 0.4553
## Block3 - Block7 0.002620 0.00126 Inf 2.077 0.5436
## Block3 - Block8 0.002439 0.00126 Inf 1.934 0.6457
## Block3 - Block9 0.003337 0.00126 Inf 2.646 0.1962
## Block3 - Block10 0.003657 0.00126 Inf 2.900 0.1055
## Block4 - Block5 0.001947 0.00126 Inf 1.544 0.8744
## Block4 - Block6 0.000722 0.00126 Inf 0.572 0.9999
## Block4 - Block7 0.000565 0.00126 Inf 0.448 1.0000
## Block4 - Block8 0.000384 0.00126 Inf 0.304 1.0000
## Block4 - Block9 0.001282 0.00126 Inf 1.017 0.9914
## Block4 - Block10 0.001602 0.00126 Inf 1.270 0.9603
## Block5 - Block6 -0.001226 0.00126 Inf -0.972 0.9938
## Block5 - Block7 -0.001383 0.00126 Inf -1.096 0.9852
## Block5 - Block8 -0.001564 0.00126 Inf -1.240 0.9660
## Block5 - Block9 -0.000665 0.00126 Inf -0.528 1.0000
## Block5 - Block10 -0.000346 0.00126 Inf -0.274 1.0000
## Block6 - Block7 -0.000157 0.00126 Inf -0.125 1.0000
## Block6 - Block8 -0.000338 0.00126 Inf -0.268 1.0000
## Block6 - Block9 0.000560 0.00126 Inf 0.444 1.0000
## Block6 - Block10 0.000880 0.00126 Inf 0.698 0.9995
## Block7 - Block8 -0.000181 0.00126 Inf -0.143 1.0000
## Block7 - Block9 0.000717 0.00126 Inf 0.569 0.9999
## Block7 - Block10 0.001037 0.00126 Inf 0.822 0.9983
## Block8 - Block9 0.000898 0.00126 Inf 0.712 0.9994
## Block8 - Block10 0.001218 0.00126 Inf 0.966 0.9941
## Block9 - Block10 0.000320 0.00126 Inf 0.254 1.0000
##
## Degrees-of-freedom method: asymptotic
## P value adjustment: tukey method for comparing a family of 10 estimatesFastest: 6, 15, 28 Slowest: 26, 23, 35
block_10_data <- execution_data_acc %>% filter(Block ==10)
# Calculate mean trial_total_rt and mean trial_acc for each participant
summary_stats <- block_10_data %>%
group_by(Subject) %>%
summarise(mean_total_rt = mean(trial_total_rt, na.rm = TRUE),
mean_trial_acc = mean(trial_acc, na.rm = TRUE))
# Rank participants by mean_total_rt
ranked_by_rt <- summary_stats %>%
arrange(mean_total_rt) %>%
mutate(rank_by_rt = row_number())
# Print the ranked list by response time
print(ranked_by_rt)## # A tibble: 22 × 4
## Subject mean_total_rt mean_trial_acc rank_by_rt
## <chr> <dbl> <dbl> <int>
## 1 6 913. 1 1
## 2 15 1343. 1 2
## 3 28 1352. 1 3
## 4 16 1512. 1 4
## 5 27 1535. 1 5
## 6 11 1636 1 6
## 7 24 1683. 1 7
## 8 19 1709. 1 8
## 9 34 2116. 1 9
## 10 22 2219. 1 10
## # ℹ 12 more rows# Rank participants by mean_trial_acc
ranked_by_acc <- summary_stats %>%
arrange(desc(mean_trial_acc)) %>%
mutate(rank_by_acc = row_number())
# Print the ranked list by accuracy
print(ranked_by_acc)## # A tibble: 22 × 4
## Subject mean_total_rt mean_trial_acc rank_by_acc
## <chr> <dbl> <dbl> <int>
## 1 11 1636 1 1
## 2 12 2660. 1 2
## 3 13 3048. 1 3
## 4 14 3314. 1 4
## 5 15 1343. 1 5
## 6 16 1512. 1 6
## 7 19 1709. 1 7
## 8 22 2219. 1 8
## 9 23 3897. 1 9
## 10 24 1683. 1 10
## # ℹ 12 more rows# Create the data frame
data <- data.frame(
Participant = factor(c("Subject 6", "Subject 15", "Subject 28", "Subject 26", "Subject 23", "Subject 35"),
levels = c("Subject 6", "Subject 15", "Subject 28", "Subject 26", "Subject 23", "Subject 35")),
Mean_Total_RT = c(913.07, 1342.83, 1352.37, 3708.68, 3896.67, 4534.33),
Group = c("Fastest", "Fastest", "Fastest", "Slowest", "Slowest", "Slowest")
)
# Create the plot
plot <- ggplot(data, aes(x = Participant, y = Mean_Total_RT, fill = Group)) +
geom_bar(stat = "identity", position = position_dodge()) +
geom_text(aes(label = paste(Mean_Total_RT, "ms"), fontface="bold"), vjust = -0.3, size = 3.5) +
labs(x = "Participant",
y = "Mean Total Trial RT (ms)") +
scale_fill_manual(values = c("Fastest" = "aquamarine4", "Slowest" = "darkorchid4")) +
theme_minimal() +
theme(
plot.title = element_blank(),
axis.title.x = element_text(face = "bold"),
axis.title.y = element_text(face = "bold")
)
# Print the plot
print(plot)
Accuracy of those Participants
# Filter data for subjects
filtered_data <- execution_data %>%
filter(Subject %in% c(6, 15, 28, 26, 23, 35))
# Compute accuracy as the proportion of successful trials (trial_acc == 1) by Subject and Block
accuracy_data <- filtered_data %>%
group_by(Subject, Block) %>%
summarise(
Total_Trials = n(),
Success_Trials = sum(trial_acc == 1),
Accuracy = Success_Trials / Total_Trials * 100
)## `summarise()` has grouped output by 'Subject'. You can override using the
## `.groups` argument.# View the computed accuracies
print(accuracy_data)## # A tibble: 60 × 5
## # Groups: Subject [6]
## Subject Block Total_Trials Success_Trials Accuracy
## <chr> <ord> <int> <int> <dbl>
## 1 15 1 48 37 77.1
## 2 15 2 48 43 89.6
## 3 15 3 48 43 89.6
## 4 15 4 48 46 95.8
## 5 15 5 48 40 83.3
## 6 15 6 48 41 85.4
## 7 15 7 48 39 81.2
## 8 15 8 48 41 85.4
## 9 15 9 48 40 83.3
## 10 15 10 48 46 95.8
## # ℹ 50 more rows# Compute average accuracy across all blocks for each participant
average_accuracy <- accuracy_data %>%
group_by(Subject) %>%
summarise(
Average_Accuracy = mean(Accuracy)
)
# View the average accuracies
print(average_accuracy)## # A tibble: 6 × 2
## Subject Average_Accuracy
## <chr> <dbl>
## 1 15 86.7
## 2 23 95.6
## 3 26 92.9
## 4 28 96.0
## 5 35 87.1
## 6 6 84.2filtered_data <- execution_data %>%
filter(Subject %in% c(6, 15, 28, 35, 26, 23)) %>%
filter(Block == 1) %>%
mutate(Group = ifelse(Subject %in% c(6, 15, 28), "Fast", "Slow"))
filtered_data2 <- execution_data %>%
filter(Subject %in% c(6, 15, 28, 35, 26, 23)) %>%
filter(Block == 9) %>%
mutate(Group = ifelse(Subject %in% c(6, 15, 28), "Fast", "Slow"))
filtered_data3 <- execution_data %>%
filter(Subject %in% c(6, 15, 28, 35, 26, 23)) %>%
filter(Block == 10) %>%
mutate(Group = ifelse(Subject %in% c(6, 15, 28), "Fast", "Slow"))
filtered_data4 <- preparation_data %>%
filter(Subject %in% c(6, 15, 28, 35, 26, 23)) %>%
filter(Block == 1) %>%
mutate(Group = ifelse(Subject %in% c(6, 15, 28), "Fast", "Slow"))
filtered_data5 <- preparation_data %>%
filter(Subject %in% c(6, 15, 28, 35, 26, 23)) %>%
filter(Block == 9) %>%
mutate(Group = ifelse(Subject %in% c(6, 15, 28), "Fast", "Slow"))
filtered_data6 <- preparation_data %>%
filter(Subject %in% c(6, 15, 28, 35, 26, 23)) %>%
filter(Block == 10) %>%
mutate(Group = ifelse(Subject %in% c(6, 15, 28), "Fast", "Slow"))# Sequence Execution Period - Block 1
model <- lmer(hurst_acceleration_x ~ Group + (1 | Subject), data = filtered_data)
anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.0037696 0.0037696 1 4 0.8172 0.4171model <- lmer(hurst_acceleration_y ~ Group + (1 | Subject), data = filtered_data)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_y ~ Group + (1 | Subject)
## Data: filtered_data
##
## REML criterion at convergence: -589.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.7015 0.0055 0.0178 0.0297 3.9292
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.017834 0.13354
## Residual 0.006719 0.08197
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.8863 0.0774 4.0000 11.451 0.000332 ***
## GroupSlow 0.1047 0.1095 4.0000 0.957 0.392865
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.0061508 0.0061508 1 4 0.9154 0.3929model <- lmer(hurst_acceleration_z ~ Group + (1 | Subject), data = filtered_data)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_z ~ Group + (1 | Subject)
## Data: filtered_data
##
## REML criterion at convergence: -1031.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -13.2161 0.0126 0.0474 0.0835 0.4660
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 5.726e-05 0.007567
## Residual 1.512e-03 0.038881
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.994768 0.005439 4.000000 182.887 5.36e-09 ***
## GroupSlow -0.005979 0.007692 4.000000 -0.777 0.48
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.00091338 0.00091338 1 4 0.6042 0.4804# Sequence Execution Period - Block 9
model <- lmer(hurst_acceleration_x ~ Group + (1 | Subject), data = filtered_data2)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_x ~ Group + (1 | Subject)
## Data: filtered_data2
##
## REML criterion at convergence: -659.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -5.1809 -0.0048 0.0049 0.0840 3.3631
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.012518 0.1119
## Residual 0.005271 0.0726
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.81572 0.06488 4.00000 12.573 0.00023 ***
## GroupSlow 0.17847 0.09175 4.00000 1.945 0.12364
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.019945 0.019945 1 4 3.7836 0.1236model <- lmer(hurst_acceleration_y ~ Group + (1 | Subject), data = filtered_data2)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_y ~ Group + (1 | Subject)
## Data: filtered_data2
##
## REML criterion at convergence: -612.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.8535 -0.0038 0.0106 0.0288 3.1364
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.013895 0.11788
## Residual 0.006216 0.07884
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.80293 0.06837 4.00000 11.744 0.000301 ***
## GroupSlow 0.19285 0.09669 4.00000 1.994 0.116851
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.024727 0.024727 1 4 3.978 0.1169model <- lmer(hurst_acceleration_z ~ Group + (1 | Subject), data = filtered_data2)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_z ~ Group + (1 | Subject)
## Data: filtered_data2
##
## REML criterion at convergence: -1252.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -9.5784 -0.0018 0.0287 0.0559 2.0123
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.0004699 0.02168
## Residual 0.0006739 0.02596
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.96842 0.01270 4.00000 76.250 1.77e-07 ***
## GroupSlow 0.02764 0.01796 4.00000 1.539 0.199
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.0015961 0.0015961 1 4 2.3686 0.1986# Sequence Execution Period - Block 10
model <- lmer(hurst_acceleration_x ~ Group + (1 | Subject), data = filtered_data3)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_x ~ Group + (1 | Subject)
## Data: filtered_data3
##
## REML criterion at convergence: -684.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -5.0970 -0.0134 0.0207 0.1979 3.2837
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.006801 0.08247
## Residual 0.004865 0.06975
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.85909 0.04797 4.00000 17.910 5.71e-05 ***
## GroupSlow 0.13255 0.06784 4.00000 1.954 0.122
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.018574 0.018574 1 4 3.8179 0.1224model <- lmer(hurst_acceleration_y ~ Group + (1 | Subject), data = filtered_data3)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_y ~ Group + (1 | Subject)
## Data: filtered_data3
##
## REML criterion at convergence: -704.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.5938 -0.0034 0.0075 0.1555 4.7329
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.01343 0.11588
## Residual 0.00449 0.06701
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.85650 0.06714 4.00000 12.757 0.000218 ***
## GroupSlow 0.13645 0.09495 4.00000 1.437 0.224043
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.009274 0.009274 1 4 2.0654 0.224model <- lmer(hurst_acceleration_z ~ Group + (1 | Subject), data = filtered_data3)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_z ~ Group + (1 | Subject)
## Data: filtered_data3
##
## REML criterion at convergence: -1154.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -9.5483 0.0054 0.0603 0.4014 1.1018
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.0001969 0.01403
## Residual 0.0009648 0.03106
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.975087 0.008505 4.000000 114.648 3.47e-08 ***
## GroupSlow 0.016889 0.012028 4.000000 1.404 0.233
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.0019021 0.0019021 1 4 1.9716 0.233# Movement Preparation Period - Block 1
model <- lmer(hurst_acceleration_x ~ Group + (1 | Subject), data = filtered_data4)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_x ~ Group + (1 | Subject)
## Data: filtered_data4
##
## REML criterion at convergence: -375.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.8495 -0.6289 0.3521 0.6333 2.1973
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.007256 0.08518
## Residual 0.014535 0.12056
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.87354 0.05019 4.00000 17.403 6.4e-05 ***
## GroupSlow -0.03789 0.07099 4.00000 -0.534 0.622
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.004141 0.004141 1 4 0.2849 0.6218model <- lmer(hurst_acceleration_y ~ Group + (1 | Subject), data = filtered_data4)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_y ~ Group + (1 | Subject)
## Data: filtered_data4
##
## REML criterion at convergence: -291.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.5275 -0.6136 0.2329 0.5544 2.2016
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.01078 0.1038
## Residual 0.01948 0.1396
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.83383 0.06105 4.00000 13.658 0.000166 ***
## GroupSlow -0.02882 0.08634 4.00000 -0.334 0.755293
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.0021707 0.0021707 1 4 0.1114 0.7553model <- lmer(hurst_acceleration_z ~ Group + (1 | Subject), data = filtered_data4)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_z ~ Group + (1 | Subject)
## Data: filtered_data4
##
## REML criterion at convergence: -222.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9767 -0.6007 0.3586 0.5930 1.7903
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.01060 0.1030
## Residual 0.02486 0.1577
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.83587 0.06089 4.00000 13.728 0.000163 ***
## GroupSlow -0.04014 0.08611 4.00000 -0.466 0.665368
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.0054024 0.0054024 1 4 0.2173 0.6654# Movement Preparation Period - Block 9
model <- lmer(hurst_acceleration_x ~ Group + (1 | Subject), data = filtered_data5)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_x ~ Group + (1 | Subject)
## Data: filtered_data5
##
## REML criterion at convergence: -457.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2435 -0.6456 0.2152 0.6517 2.4547
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.002552 0.05052
## Residual 0.011038 0.10506
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.91648 0.03045 4.00000 30.095 7.26e-06 ***
## GroupSlow -0.12646 0.04307 4.00000 -2.936 0.0425 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.095171 0.095171 1 4 8.6224 0.04254 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1model <- lmer(hurst_acceleration_y ~ Group + (1 | Subject), data = filtered_data5)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_y ~ Group + (1 | Subject)
## Data: filtered_data5
##
## REML criterion at convergence: -383.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.9054 -0.6378 0.1432 0.6196 2.2454
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.002463 0.04963
## Residual 0.014345 0.11977
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.90182 0.03034 4.00000 29.724 7.63e-06 ***
## GroupSlow -0.17326 0.04291 4.00000 -4.038 0.0156 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.2339 0.2339 1 4 16.305 0.01563 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1model <- lmer(hurst_acceleration_z ~ Group + (1 | Subject), data = filtered_data5)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_z ~ Group + (1 | Subject)
## Data: filtered_data5
##
## REML criterion at convergence: -270.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2150 -0.6143 0.1575 0.4678 2.0743
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.005819 0.07628
## Residual 0.021172 0.14551
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.89652 0.04568 4.00000 19.626 3.98e-05 ***
## GroupSlow -0.19257 0.06460 4.00000 -2.981 0.0407 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.18812 0.18812 1 4 8.8856 0.0407 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Movement Preparation Period - Block 10
model <- lmer(hurst_acceleration_x ~ Group + (1 | Subject), data = filtered_data6)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_x ~ Group + (1 | Subject)
## Data: filtered_data6
##
## REML criterion at convergence: -436.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.9092 -0.5828 0.1531 0.5359 2.4489
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.002738 0.05232
## Residual 0.011876 0.10898
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.93090 0.03154 4.00000 29.512 7.85e-06 ***
## GroupSlow -0.14290 0.04461 4.00000 -3.203 0.0328 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.12186 0.12186 1 4 10.261 0.0328 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1model <- lmer(hurst_acceleration_y ~ Group + (1 | Subject), data = filtered_data6)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_y ~ Group + (1 | Subject)
## Data: filtered_data6
##
## REML criterion at convergence: -342.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.6091 -0.6421 0.1276 0.5117 2.3976
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.002039 0.04515
## Residual 0.016615 0.12890
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.90962 0.02819 4.00000 32.263 5.5e-06 ***
## GroupSlow -0.18697 0.03987 4.00000 -4.689 0.00938 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.36534 0.36534 1 4 21.989 0.009383 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1model <- lmer(hurst_acceleration_z ~ Group + (1 | Subject), data = filtered_data6)
summary(model) ## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: hurst_acceleration_z ~ Group + (1 | Subject)
## Data: filtered_data6
##
## REML criterion at convergence: -234.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.93995 -0.65482 0.09468 0.35049 2.17322
##
## Random effects:
## Groups Name Variance Std.Dev.
## Subject (Intercept) 0.006358 0.07974
## Residual 0.023990 0.15489
## Number of obs: 288, groups: Subject, 6
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.90593 0.04781 4.00000 18.948 4.57e-05 ***
## GroupSlow -0.22457 0.06762 4.00000 -3.321 0.0293 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## GroupSlow -0.707anova(model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## Group 0.26463 0.26463 1 4 11.031 0.02934 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1‘Fast’ vs. ‘Slow’ Group Differences Plots
# Add a block identifier to each dataset
filtered_data <- filtered_data %>% mutate(Block = "Block 1")
filtered_data2 <- filtered_data2 %>% mutate(Block = "Block 9")
filtered_data3 <- filtered_data3 %>% mutate(Block = "Block 10")
# Combine the data into one data frame
combined_data <- bind_rows(filtered_data, filtered_data2, filtered_data3)
# Adjust the factor levels to control plot order
combined_data$Block <- factor(combined_data$Block, levels = c("Block 1", "Block 9", "Block 10"))
# Adjust the plot with larger and bold fonts
plot1 <- ggplot(combined_data, aes(x = Group, y = hurst_acceleration_x, fill = Group)) +
geom_boxplot() +
labs(subtitle = "Hurst Acceleration X",
x = "Group",
y = "Hurst Exponent") +
theme_minimal() +
scale_fill_manual(values = c("Fast" = "aquamarine4", "Slow" = "darkorchid4")) +
facet_wrap(~Block) +
theme(
axis.text = element_text(size = 14, face = "bold"),
axis.text.x = element_text(angle = 45, hjust = 1, size = 14, face = "bold"),
axis.text.y = element_text(size = 14, face = "bold"),
axis.title = element_text(size = 16, face = "bold"),
plot.title = element_text(size = 24, face = "bold", hjust = 0.5),
plot.subtitle = element_text(size = 22, face = "bold", hjust = 0.5),
strip.text = element_text(size = 20, face = "bold"),
legend.text = element_text(size = 18, face = "bold"),
legend.title = element_text(size = 20, face = "bold"),
legend.key.size = unit(3.5, "lines")
)
# Print the updated plot
print(plot1)
# Add a block identifier to each dataset
filtered_data <- filtered_data %>% mutate(Block = "Block 1")
filtered_data2 <- filtered_data2 %>% mutate(Block = "Block 9")
filtered_data3 <- filtered_data3 %>% mutate(Block = "Block 10")
# Combine the data into one data frame
combined_data <- bind_rows(filtered_data, filtered_data2, filtered_data3)
# Adjust the factor levels to control plot order
combined_data$Block <- factor(combined_data$Block, levels = c("Block 1", "Block 9", "Block 10"))
# Create the combined boxplot with the corrected order and larger font sizes
plot2 <- ggplot(combined_data, aes(x = Group, y = hurst_acceleration_y, fill = Group)) +
geom_boxplot() +
labs(subtitle = "Hurst Acceleration Y",
x = "Group",
y = "Hurst Exponent") +
theme_minimal() +
scale_fill_manual(values = c("Fast" = "aquamarine4", "Slow" = "darkorchid4")) +
facet_wrap(~Block) +
theme(
axis.text = element_text(size = 14, face = "bold"),
axis.text.x = element_text(angle = 45, hjust = 1, size = 14, face = "bold"),
axis.text.y = element_text(size = 14, face = "bold"),
axis.title = element_text(size = 16, face = "bold"),
plot.title = element_text(size = 24, face = "bold", hjust = 0.5),
plot.subtitle = element_text(size = 22, face = "bold", hjust = 0.5),
strip.text = element_text(size = 20, face = "bold"),
legend.text = element_text(size = 18, face = "bold"),
legend.title = element_text(size = 20, face = "bold"),
legend.key.size = unit(3.5, "lines")
)
print(plot2)
# Add a block identifier to each dataset
filtered_data <- filtered_data %>% mutate(Block = "Block 1")
filtered_data2 <- filtered_data2 %>% mutate(Block = "Block 9")
filtered_data3 <- filtered_data3 %>% mutate(Block = "Block 10")
# Combine the data into one data frame
combined_data <- bind_rows(filtered_data, filtered_data2, filtered_data3)
# Adjust the factor levels to control plot order
combined_data$Block <- factor(combined_data$Block, levels = c("Block 1", "Block 9", "Block 10"))
# Create the combined boxplot with the corrected order and larger font sizes
plot3 <- ggplot(combined_data, aes(x = Group, y = hurst_acceleration_z, fill = Group)) +
geom_boxplot() +
labs(subtitle = "Hurst Acceleration Z",
x = "Group",
y = "Hurst Exponent") +
theme_minimal() +
scale_fill_manual(values = c("Fast" = "aquamarine4", "Slow" = "darkorchid4")) +
facet_wrap(~Block) +
theme(
axis.text = element_text(size = 14, face = "bold"),
axis.text.x = element_text(angle = 45, hjust = 1, size = 14, face = "bold"),
axis.text.y = element_text(size = 14, face = "bold"),
axis.title = element_text(size = 16, face = "bold"),
plot.title = element_text(size = 24, face = "bold", hjust = 0.5),
plot.subtitle = element_text(size = 22, face = "bold", hjust = 0.5),
strip.text = element_text(size = 20, face = "bold"),
legend.text = element_text(size = 18, face = "bold"),
legend.title = element_text(size = 20, face = "bold"),
legend.key.size = unit(3.5, "lines")
)
print(plot3)
# Add a block identifier to each dataset
filtered_data4 <- filtered_data4 %>% mutate(Block = "Block 1")
filtered_data5 <- filtered_data5 %>% mutate(Block = "Block 9**")
filtered_data6 <- filtered_data6 %>% mutate(Block = "Block 10**")
# Combine the data into one data frame
combined_data <- bind_rows(filtered_data4, filtered_data5, filtered_data6)
# Adjust the factor levels to control plot order
combined_data$Block <- factor(combined_data$Block, levels = c("Block 1", "Block 9**", "Block 10**"))
# Create the combined boxplot with the corrected order and larger font sizes
plot4 <- ggplot(combined_data, aes(x = Group, y = hurst_acceleration_x, fill = Group)) +
geom_boxplot() +
labs(subtitle = "Hurst Acceleration X",
x = "Group",
y = "Hurst Exponent") +
theme_minimal() +
scale_fill_manual(values = c("Fast" = "aquamarine4", "Slow" = "darkorchid4")) +
facet_wrap(~Block) +
theme(
axis.text = element_text(size = 14, face = "bold"),
axis.text.x = element_text(angle = 45, hjust = 1, size = 14, face = "bold"),
axis.text.y = element_text(size = 14, face = "bold"),
axis.title = element_text(size = 16, face = "bold"),
plot.title = element_text(size = 24, face = "bold", hjust = 0.5),
plot.subtitle = element_text(size = 22, face = "bold", hjust = 0.5),
strip.text = element_text(size = 20, face = "bold"),
legend.text = element_text(size = 18, face = "bold"),
legend.title = element_text(size = 20, face = "bold"),
legend.key.size = unit(3.5, "lines")
)
print(plot4)
# Add a block identifier to each dataset
filtered_data4 <- filtered_data4 %>% mutate(Block = "Block 1")
filtered_data5 <- filtered_data5 %>% mutate(Block = "Block 9**")
filtered_data6 <- filtered_data6 %>% mutate(Block = "Block 10**")
# Combine the data into one data frame
combined_data <- bind_rows(filtered_data4, filtered_data5, filtered_data6)
# Adjust the factor levels to control plot order
combined_data$Block <- factor(combined_data$Block, levels = c("Block 1", "Block 9**", "Block 10**"))
# Create the combined boxplot with the corrected order and larger font sizes
plot5 <- ggplot(combined_data, aes(x = Group, y = hurst_acceleration_y, fill = Group)) +
geom_boxplot() +
labs(subtitle = "Hurst Acceleration Y",
x = "Group",
y = "Hurst Exponent") +
theme_minimal() +
scale_fill_manual(values = c("Fast" = "aquamarine4", "Slow" = "darkorchid4")) +
facet_wrap(~Block) +
theme(
axis.text = element_text(size = 14, face = "bold"),
axis.text.x = element_text(angle = 45, hjust = 1, size = 14, face = "bold"),
axis.text.y = element_text(size = 14, face = "bold"),
axis.title = element_text(size = 16, face = "bold"),
plot.title = element_text(size = 24, face = "bold", hjust = 0.5),
plot.subtitle = element_text(size = 22, face = "bold", hjust = 0.5),
strip.text = element_text(size = 20, face = "bold"),
legend.text = element_text(size = 18, face = "bold"),
legend.title = element_text(size = 20, face = "bold"),
legend.key.size = unit(3.5, "lines")
)
print(plot5)
# Add a block identifier to each dataset
filtered_data4 <- filtered_data4 %>% mutate(Block = "Block 1")
filtered_data5 <- filtered_data5 %>% mutate(Block = "Block 9**")
filtered_data6 <- filtered_data6 %>% mutate(Block = "Block 10**")
# Combine the data into one data frame
combined_data <- bind_rows(filtered_data4, filtered_data5, filtered_data6)
# Adjust the factor levels to control plot order
combined_data$Block <- factor(combined_data$Block, levels = c("Block 1", "Block 9**", "Block 10**"))
# Create the combined boxplot with the corrected order and larger font sizes
plot6 <- ggplot(combined_data, aes(x = Group, y = hurst_acceleration_z, fill = Group)) +
geom_boxplot() +
labs(subtitle = "Hurst Acceleration Z",
x = "Group",
y = "Hurst Exponent") +
theme_minimal() +
scale_fill_manual(values = c("Fast" = "aquamarine4", "Slow" = "darkorchid4")) +
facet_wrap(~Block) +
theme(
axis.text = element_text(size = 14, face = "bold"),
axis.text.x = element_text(angle = 45, hjust = 1, size = 14, face = "bold"),
axis.text.y = element_text(size = 14, face = "bold"),
axis.title = element_text(size = 16, face = "bold"),
plot.title = element_text(size = 24, face = "bold", hjust = 0.5),
plot.subtitle = element_text(size = 22, face = "bold", hjust = 0.5),
strip.text = element_text(size = 20, face = "bold"),
legend.text = element_text(size = 18, face = "bold"),
legend.title = element_text(size = 20, face = "bold"),
legend.key.size = unit(3.5, "lines")
)
print(plot6)
RT Trajectory Plot
# Filter data for subjects 6, 15, 28, 26, 23, and 35
filtered_data <- execution_data %>%
filter(Subject %in% c(6, 15, 28, 26, 23, 35))
# Calculate the mean of trial_total_rt for each subject within each block
mean_rt_data <- filtered_data %>%
group_by(Subject, Block) %>%
summarise(
Mean_RT = mean(trial_total_rt, na.rm = TRUE),
.groups = 'drop'
)
# Plotting the trajectory of response times across blocks
rt_plot <- ggplot(mean_rt_data, aes(x = Block, y = Mean_RT, group = Subject, color = as.factor(Subject))) +
geom_smooth(se = FALSE, size = 2) +
labs(title = "Trajectory of Response Times Across Blocks",
x = "Block",
y = "Mean Response Time (ms)",
color = "Participant") +
theme_minimal() +
scale_color_manual(values = c("brown3", "cornflowerblue", "chartreuse3", "darkmagenta", "darkorange", "aquamarine2"))
# Print the plot
print(rt_plot)## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
# Specify the participants
participants <- c(6, 15, 28, 26, 23, 35)
# Filter trial_data for sequence execution period
filtered_execution <- execution_data %>%
filter(Subject %in% participants) %>%
group_by(Subject, Block) %>%
summarise(
Mean_Hurst_X = mean(hurst_acceleration_x, na.rm = TRUE),
Mean_Hurst_Y = mean(hurst_acceleration_y, na.rm = TRUE),
Mean_Hurst_Z = mean(hurst_acceleration_z, na.rm = TRUE),
.groups = 'drop'
)
# Filter trial_data2 for movement preparation period
filtered_preparation <- preparation_data %>%
filter(Subject %in% participants) %>%
group_by(Subject, Block) %>%
summarise(
Mean_Hurst_X = mean(hurst_acceleration_x, na.rm = TRUE),
Mean_Hurst_Y = mean(hurst_acceleration_y, na.rm = TRUE),
Mean_Hurst_Z = mean(hurst_acceleration_z, na.rm = TRUE),
.groups = 'drop'
)
# Plotting trajectory for each axis in the execution period
for(axis in c("X", "Y", "Z")) {
plot <- ggplot(filtered_execution, aes(x = Block, y = get(paste0("Mean_Hurst_", axis)), group = Subject, color = as.factor(Subject))) +
geom_smooth(se = TRUE, method = "loess", size = 2) +
labs(title = paste("Execution Period -", axis, "Axis"),
x = "Block",
y = "Mean Hurst Exponent",
color = "Participant") +
theme_minimal() +
scale_color_manual(values = c("brown3", "cornflowerblue", "chartreuse3", "darkmagenta", "darkorange", "aquamarine2")) +
theme(
text = element_text(size = 14, face = "bold"),
axis.title = element_text(size = 16, face = "bold"),
plot.title = element_text(size = 20, face = "bold"),
legend.text = element_text(size = 14, face = "bold"),
legend.title = element_text(size = 16, face = "bold"),
strip.text = element_text(size = 16, face = "bold")
)
print(plot)
}## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'
# Plotting trajectory for each axis in the preparation period
for(axis in c("X", "Y", "Z")) {
plot <- ggplot(filtered_preparation, aes(x = Block, y = get(paste0("Mean_Hurst_", axis)), group = Subject, color = as.factor(Subject))) +
geom_smooth(se = TRUE, method = "loess", size = 2) +
labs(title = paste("Preparation Period -", axis, "Axis"),
x = "Block",
y = "Mean Hurst Exponent",
color = "Participant") +
theme_minimal() +
scale_color_manual(values = c("brown3", "cornflowerblue", "chartreuse3", "darkmagenta", "darkorange", "aquamarine2")) +
theme(
text = element_text(size = 14, face = "bold"),
axis.title = element_text(size = 16, face = "bold"),
plot.title = element_text(size = 20, face = "bold"),
legend.text = element_text(size = 14, face = "bold"),
legend.title = element_text(size = 16, face = "bold"),
strip.text = element_text(size = 16, face = "bold")
)
print(plot)
}## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'
Comparison between Mean Hurst Exponents in Sequence Execution Phase for Top 50% of Accurate Performers and Bottom 50% of Accurate Performers
# Calculate overall accuracy
overall_accuracy <- execution_data %>%
group_by(Subject) %>%
summarize(Accuracy = mean(trial_acc, na.rm = TRUE))
mean_hurst <- execution_data %>%
group_by(Subject) %>%
summarize(
Mean_Hurst_X = mean(hurst_acceleration_x, na.rm = TRUE),
Mean_Hurst_Y = mean(hurst_acceleration_y, na.rm = TRUE),
Mean_Hurst_Z = mean(hurst_acceleration_z, na.rm = TRUE)
)
# Merge the two data frames
participant_data <- merge(overall_accuracy, mean_hurst, by = "Subject")
# Sort and find the median accuracy
sorted_data <- participant_data %>%
arrange(desc(Accuracy))
# Split into top 50% and bottom 50%
top_50 <- sorted_data[1:11, ]
bottom_50 <- sorted_data[12:22, ]
# Prepare data for plotting
plot_data <- bind_rows(
mutate(top_50, Group = "Top 50%"),
mutate(bottom_50, Group = "Bottom 50%")
)# Plot for Hurst Exponent X
ggplot(plot_data, aes(x = Group, y = Mean_Hurst_X, fill = Group)) +
geom_boxplot() +
labs(title = "Comparison of Hurst Exponent X during Sequence Execution - Accuracy", y = "Hurst Exponent X", x = "") +
scale_fill_manual(values = c("Top 50%" = "aquamarine4", "Bottom 50%" = "darkorchid4")) +
theme_minimal()
# Plot for Hurst Exponent Y
ggplot(plot_data, aes(x = Group, y = Mean_Hurst_Y, fill = Group)) +
geom_boxplot() +
labs(title = "Comparison of Hurst Exponent Y during Sequence Execution - Accuracy", y = "Hurst Exponent Y", x = "") +
scale_fill_manual(values = c("Top 50%" = "aquamarine4", "Bottom 50%" = "darkorchid4")) +
theme_minimal()
# Plot for Hurst Exponent Z
ggplot(plot_data, aes(x = Group, y = Mean_Hurst_Z, fill = Group)) +
geom_boxplot() +
labs(title = "Comparison of Hurst Exponent Z during Sequence Execution - Accuracy", y = "Hurst Exponent Z", x = "") +
scale_fill_manual(values = c("Top 50%" = "aquamarine4", "Bottom 50%" = "darkorchid4")) +
theme_minimal()
Comparison between Mean Hurst Exponents in Sequence Execution Phase for Top 50% of the fastest Performers and Bottom 50%
# Calculate overall speed
overall_speed <- execution_data_filtered %>%
group_by(Subject) %>%
summarize(Speed = mean(trial_total_rt, na.rm = TRUE))
mean_hurst <- preparation_data %>%
group_by(Subject) %>%
summarize(
Mean_Hurst_X = mean(hurst_acceleration_x, na.rm = TRUE),
Mean_Hurst_Y = mean(hurst_acceleration_y, na.rm = TRUE),
Mean_Hurst_Z = mean(hurst_acceleration_z, na.rm = TRUE)
)
# Merge the two data frames
participant_data <- merge(overall_speed, mean_hurst, by = "Subject")
# Sort and find the median accuracy
sorted_data <- participant_data %>%
arrange(desc(Speed))
# Split into top 50% and bottom 50%
top_50 <- sorted_data[1:11, ]
bottom_50 <- sorted_data[12:22, ]
# Prepare data for plotting
plot_data <- bind_rows(
mutate(top_50, Group = "Top 50%"),
mutate(bottom_50, Group = "Bottom 50%")
)
# Plot for Hurst Exponent X
ggplot(plot_data, aes(x = Group, y = Mean_Hurst_X, fill = Group)) +
geom_boxplot() +
labs(title = "Comparison of Hurst Exponent X during Sequence Execution - Speed", y = "Hurst Exponent X", x = "") +
scale_fill_manual(values = c("Top 50%" = "aquamarine4", "Bottom 50%" = "darkorchid4")) +
theme_minimal()
# Plot for Hurst Exponent Y
ggplot(plot_data, aes(x = Group, y = Mean_Hurst_Y, fill = Group)) +
geom_boxplot() +
labs(title = "Comparison of Hurst Exponent Y during Sequence Execution - Speed", y = "Hurst Exponent Y", x = "") +
scale_fill_manual(values = c("Top 50%" = "aquamarine4", "Bottom 50%" = "darkorchid4")) +
theme_minimal()
# Plot for Hurst Exponent Z
ggplot(plot_data, aes(x = Group, y = Mean_Hurst_Z, fill = Group)) +
geom_boxplot() +
labs(title = "Comparison of Hurst Exponent Z during Sequence Execution - Speed", y = "Hurst Exponent Z", x = "") +
scale_fill_manual(values = c("Top 50%" = "aquamarine4", "Bottom 50%" = "darkorchid4")) +
theme_minimal()
shapiro.test(top_50$Mean_Hurst_X)##
## Shapiro-Wilk normality test
##
## data: top_50$Mean_Hurst_X
## W = 0.96794, p-value = 0.865shapiro.test(bottom_50$Mean_Hurst_X)##
## Shapiro-Wilk normality test
##
## data: bottom_50$Mean_Hurst_X
## W = 0.87221, p-value = 0.08275shapiro.test(top_50$Mean_Hurst_Y)##
## Shapiro-Wilk normality test
##
## data: top_50$Mean_Hurst_Y
## W = 0.96188, p-value = 0.7949shapiro.test(bottom_50$Mean_Hurst_Y)##
## Shapiro-Wilk normality test
##
## data: bottom_50$Mean_Hurst_Y
## W = 0.96222, p-value = 0.7991shapiro.test(top_50$Mean_Hurst_Z)##
## Shapiro-Wilk normality test
##
## data: top_50$Mean_Hurst_Z
## W = 0.9308, p-value = 0.419shapiro.test(bottom_50$Mean_Hurst_Z)##
## Shapiro-Wilk normality test
##
## data: bottom_50$Mean_Hurst_Z
## W = 0.91863, p-value = 0.3074# Mann-Whitney U Test for Hurst Exponent X
wilcox.test(top_50$Mean_Hurst_X, bottom_50$Mean_Hurst_X)##
## Wilcoxon rank sum exact test
##
## data: top_50$Mean_Hurst_X and bottom_50$Mean_Hurst_X
## W = 31, p-value = 0.05567
## alternative hypothesis: true location shift is not equal to 0# Mann-Whitney U Test for Hurst Exponent Y
wilcox.test(top_50$Mean_Hurst_Y, bottom_50$Mean_Hurst_Y)##
## Wilcoxon rank sum exact test
##
## data: top_50$Mean_Hurst_Y and bottom_50$Mean_Hurst_Y
## W = 31, p-value = 0.05567
## alternative hypothesis: true location shift is not equal to 0# Mann-Whitney U Test for Hurst Exponent Z
wilcox.test(top_50$Mean_Hurst_Z, bottom_50$Mean_Hurst_Z)##
## Wilcoxon rank sum exact test
##
## data: top_50$Mean_Hurst_Z and bottom_50$Mean_Hurst_Z
## W = 25, p-value = 0.01923
## alternative hypothesis: true location shift is not equal to 0Comparison between Mean Hurst Exponents in Movement Preparation Phase for Top 50% of Accurate Performers and Bottom 50% of Accurate Performers
# Calculate overall accuracy
overall_accuracy <- preparation_data %>%
group_by(Subject) %>%
summarize(Accuracy = mean(trial_acc, na.rm = TRUE))
mean_hurst <- preparation_data %>%
group_by(Subject) %>%
summarize(
Mean_Hurst_X = mean(hurst_acceleration_x, na.rm = TRUE),
Mean_Hurst_Y = mean(hurst_acceleration_y, na.rm = TRUE),
Mean_Hurst_Z = mean(hurst_acceleration_z, na.rm = TRUE)
)
# Merge the two data frames
participant_data <- merge(overall_accuracy, mean_hurst, by = "Subject")
# Sort and find the median accuracy
sorted_data <- participant_data %>%
arrange(desc(Accuracy))
# Split into top 50% and bottom 50%
top_50 <- sorted_data[1:11, ]
bottom_50 <- sorted_data[12:22, ]
# Prepare data for plotting
plot_data <- bind_rows(
mutate(top_50, Group = "Top 50%"),
mutate(bottom_50, Group = "Bottom 50%")
)
# Plot for Hurst Exponent X
ggplot(plot_data, aes(x = Group, y = Mean_Hurst_X, fill = Group)) +
geom_boxplot() +
labs(title = "Comparison of Hurst Exponent X during Movement Preparation - Accuracy", y = "Hurst Exponent X", x = "") +
scale_fill_manual(values = c("Top 50%" = "aquamarine4", "Bottom 50%" = "darkorchid4")) +
theme_minimal()
# Plot for Hurst Exponent Y
ggplot(plot_data, aes(x = Group, y = Mean_Hurst_Y, fill = Group)) +
geom_boxplot() +
labs(title = "Comparison of Hurst Exponent Y during Movement Preparation - Accuracy", y = "Hurst Exponent Y", x = "") +
scale_fill_manual(values = c("Top 50%" = "aquamarine4", "Bottom 50%" = "darkorchid4")) +
theme_minimal()
# Plot for Hurst Exponent Z
ggplot(plot_data, aes(x = Group, y = Mean_Hurst_Z, fill = Group)) +
geom_boxplot() +
labs(title = "Comparison of Hurst Exponent Z during Movement Preparation - Accuracy", y = "Hurst Exponent Z", x = "") +
scale_fill_manual(values = c("Top 50%" = "aquamarine4", "Bottom 50%" = "darkorchid4")) +
theme_minimal()
Comparison between Mean Hurst Exponents in Movement Preparation Phase for Top 50% of the fastest Performers and Bottom 50%
# Calculate overall Speed
overall_speed <- preparation_data_filtered %>%
group_by(Subject) %>%
summarize(Speed = mean(trial_total_rt, na.rm = TRUE))
mean_hurst <- preparation_data %>%
group_by(Subject) %>%
summarize(
Mean_Hurst_X = mean(hurst_acceleration_x, na.rm = TRUE),
Mean_Hurst_Y = mean(hurst_acceleration_y, na.rm = TRUE),
Mean_Hurst_Z = mean(hurst_acceleration_z, na.rm = TRUE)
)
# Merge the two data frames
participant_data <- merge(overall_speed, mean_hurst, by = "Subject")
# Sort and find the median accuracy
sorted_data <- participant_data %>%
arrange(desc(Speed))
# Split into top 50% and bottom 50%
top_50 <- sorted_data[1:11, ]
bottom_50 <- sorted_data[12:22, ]
# Prepare data for plotting
plot_data <- bind_rows(
mutate(top_50, Group = "Top 50%"),
mutate(bottom_50, Group = "Bottom 50%")
)
# Plot for Hurst Exponent X
ggplot(plot_data, aes(x = Group, y = Mean_Hurst_X, fill = Group)) +
geom_boxplot() +
labs(title = "Comparison of Hurst Exponent X during Movement Preparation - Speed", y = "Hurst Exponent X", x = "") +
scale_fill_manual(values = c("Top 50%" = "aquamarine4", "Bottom 50%" = "darkorchid4")) +
theme_minimal()
# Plot for Hurst Exponent Y
ggplot(plot_data, aes(x = Group, y = Mean_Hurst_Y, fill = Group)) +
geom_boxplot() +
labs(title = "Comparison of Hurst Exponent Y during Movement Preparation - Speed", y = "Hurst Exponent Y", x = "") +
scale_fill_manual(values = c("Top 50%" = "aquamarine4", "Bottom 50%" = "darkorchid4")) +
theme_minimal()
# Plot for Hurst Exponent Z
ggplot(plot_data, aes(x = Group, y = Mean_Hurst_Z, fill = Group)) +
geom_boxplot() +
labs(title = "Comparison of Hurst Exponent Z during Movement Preparation - Speed", y = "Hurst Exponent Z", x = "") +
scale_fill_manual(values = c("Top 50%" = "aquamarine4", "Bottom 50%" = "darkorchid4")) +
theme_minimal()