Visualization_Pipeline_220629.Rmd

---
title: "5-choice training visualizations and analysis"
output: html_notebook
---

*Current as of June 29th, 2022*

# libraries and functions

read in libraries and functions

```{r}

library(readr)
library(dplyr)
library(stringr)
library(magrittr)
library(tidyr)
library(ggplot2)
library(purrr)

library(ggrepel)

index <- function(x){
  as.integer(row.names(x))
}

```



# read in data, and ID descriptions

```{r}
train = read_csv("PROCESSED DATA FILEPATH GOES HERE") %>% 
  
  mutate(
    
    Litter = case_when(),
    
    Sex = case_when(),
    
    Group = case_when()
    
  ) 

```

```{r}

train$ID <- factor(train$ID, levels = c())

```

# overall training progress by sex

```{r, fig.width=15, fig.height=8}

ggplot(train %>% 
         
         group_by(Group, Sex, Stage, ID, Session_all) %>% 
         
         summarise(n = n()),

       aes(Session_all,
           
           as.integer(Stage),
           
           fill = Group
           
           )) +
  
  facet_wrap(~ Sex) +

  geom_jitter(aes(color = Group),
              height = 0.1,
              width = 0.1) +

  stat_summary(aes(fill = Group),
               fun = "mean", geom = "line",
               size = 1) +

  stat_summary(fun.data = "mean_se", geom = "ribbon",
               alpha = 0.25) +
  
  scale_y_continuous(breaks = seq(0, 6, by = 1)) +
  
  # scale_color_manual(values = c("dodgerblue", "orange", "green2"))+
  #scale_fill_manual(values = c("dodgerblue", "orange", "green"))+

  ggtitle("Progress across training stages") +
  ylab("Stage of Training") +
  xlab("Day of Training") +

  theme_classic(base_size=22) +
  
  theme(axis.text = element_text(size = 14),
        strip.text.x = element_text(size = 16),
        title = element_text(size = 16)
        )

```

# overall training progress

```{r, fig.width=15, fig.height=8}

ggplot(train %>% 
         
         group_by(Group, Sex, Stage, ID, Session_all) %>% 
         
         summarise(n = n()),

       aes(Session_all,
           
           as.integer(Stage),
           
           fill = Group
           
           )) +
  
  # facet_wrap(~ Sex) +

  geom_jitter(aes(color = Group),
              height = 0.1,
              width = 0.1) +

  stat_summary(aes(fill = Group),
               fun = "mean", geom = "line",
               size = 1) +

  stat_summary(fun.data = "mean_se", geom = "ribbon",
               alpha = 0.25) +
  
  scale_y_continuous(breaks = seq(0, 6, by = 1)) +
  
  # scale_color_manual(values = c("dodgerblue", "orange", "green2"))+
  #scale_fill_manual(values = c("dodgerblue", "orange", "green"))+

  ggtitle("Progress across training stages") +
  ylab("Stage of Training") +
  xlab("Day of Training") +

  theme_classic(base_size=22) +
  
  theme(axis.text = element_text(size = 14),
        strip.text.x = element_text(size = 16),
        title = element_text(size = 16)
        )

```


# Stage 0 last day (peak performance measures)

```{r}

stage0 = train %>% 
  
  filter(Stage == 0)

```

```{r}

stage0sum = stage0 %>% 
  
  group_by(Group, Sex, ID, Date) %>% 
  
  summarise(Days = n()) %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(Days = n())

```


```{r}

ggplot(stage0sum,
       
  aes(Group,
      
      Days,
      
     color = Group,
      
      fill = Group,
      
      label = ID)) +
  
  geom_text_repel() + 
  
  geom_point(position = position_jitter(seed = 1,
                                       
                                       width = 0.1,
                                       
                                       height = 0.1)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1) +
  
  # #scale_color_manual(values = c("blue", "red")) +

  theme_classic(base_size=22)

```

```{r}
t.test(Days ~ Group,
       
       stage0sum)
```


```{r}
ggplot(stage0 %>%
         
         group_by(ID) %>% 
         
         filter(Date == last(Date)) %>% 
         
         ungroup() %>% 
         
         group_by(Group, Sex, ID) %>% 
         
         summarise(Correct = sum(Response == "Correct")),
  
  aes(Group,
      
      Correct,
      
     color = Group,
      
      fill = Group,
      
      label = ID)) +
  
  geom_text_repel() + 
  
  geom_point(position = position_jitter(seed = 1,
                                       
                                       width = 0.1,
                                       
                                       height = 0.1)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1) +
  
  #scale_color_manual(values = c("blue", "red")) +

  theme_classic(base_size=22)


```

```{r}

t.test(Correct ~ Group,
       
       stage0 %>%
         
         group_by(ID) %>% 
         
         filter(Date == last(Date)) %>% 
         
         ungroup() %>% 
         
         group_by(Group, Sex, ID) %>% 
         
         summarise(Correct = sum(Response == "Correct")))

```


```{r}

ggplot(stage0 %>%
         
         group_by(ID) %>% 
         
         filter(Date == last(Date)) %>% 
         
         ungroup() %>% 
         
         group_by(Group, Sex, ID) %>% 
         
         summarise(meanRT = mean(RT)),
  
  aes(Group,
      
      meanRT*60,
      
     color = Group,
      
      fill = Group,
      
      label = ID)) +
  
  geom_text_repel() + 
  
  geom_point(position = position_jitter(seed = 1,
                                       
                                       width = 0.1,
                                       
                                       height = 0.1)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1) +
  
  #scale_color_manual(values = c("blue", "red")) +

  theme_classic(base_size=22)


```

```{r}
t.test(meanRT ~ Group,
       
       stage0 %>%
         
         group_by(ID) %>% 
         
         filter(Date == last(Date)) %>% 
         
         ungroup() %>% 
         
         group_by(Group, Sex, ID) %>% 
         
         summarise(meanRT = mean(RT)))
```


```{r}

ggplot(stage0 %>%
         
         group_by(ID) %>% 
         
         filter(Date == last(Date)) %>% 
         
         ungroup() %>% 
         
         group_by(Group, Sex, ID) %>% 
         
         summarise(meanRL = mean(na.omit(RL))),
  
  aes(Group,
      
      meanRL*60,
      
     color = Group,
      
      fill = Group,
      
      label = ID)) +
  
  geom_text_repel() + 
  
  geom_point(position = position_jitter(seed = 1,
                                       
                                       width = 0.1,
                                       
                                       height = 0.1)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1) +
  
  #scale_color_manual(values = c("blue", "red")) +

  theme_classic(base_size=22)


```

```{r}

t.test(meanRL ~ Group, 
       
       stage0 %>%
         
         group_by(ID) %>% 
         
         filter(Date == last(Date)) %>% 
         
         ungroup() %>% 
         
         group_by(Group, Sex, ID) %>% 
         
         summarise(meanRL = mean(na.omit(RL))))

```


# Stage 1 first v last day

```{r}

stage1 = train %>% 
  
  filter(Stage == 1)

```

```{r}

stage1sum = stage1 %>% 
  
    group_by(Group, Sex, ID, Date) %>% 
  
  summarise(Days = n()) %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(Days = n())

```


```{r}

ggplot(stage1sum,
       
  aes(Group,
      
      Days,
      
     color = Group,
      
      fill = Group,
      
      label = ID)) +
  
  geom_text_repel() + 
  
  geom_point(position = position_jitter(seed = 1,
                                       
                                       width = 0.1,
                                       
                                       height = 0.1)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1) +
  
  # #scale_color_manual(values = c("blue", "red")) +

  theme_classic(base_size=22)

```

```{r}

t.test(Days ~ Group,
       
       stage1sum)

```


```{r}

stage1first = stage1 %>% 
  
  group_by(ID) %>% 
  
  filter(Date == first(Date)) %>% 
  
  ungroup() %>% 
  
  select(Group, Sex, ID, Timestamp, Response, Trials, Port, Target, RT, RL)

stage1last = stage1 %>% 
  
  group_by(ID) %>% 
  
  filter(Date == last(Date)) %>% 
  
  ungroup() %>% 
  
  select(Group, Sex, ID, Timestamp, Response, Trials, Port, Target, RT, RL)

```

Let's check 1st day parameters? 

```{r}

stage1first_param = stage1first %>%
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    Correct = sum(Response == "Correct") / max(Trials),
    
    Omissions = sum(Response == "Omission") / max(Trials),
    
    Accuracy = sum(Response == "Correct") / (sum(Response == "Correct") + sum(Response == "Incorrect")),
    
    Prematures = sum(Response == "Premature") / max(Trials),
    
    Perseveratives = sum(Response == "Perseverative") / max(Trials),
    
    Timeouts = sum(Response == "Timeout") / max(Trials)
    
    ) %>% 
  
  ungroup()


head(stage1first_param)

```

Let's plot day 1 params

```{r, fig.width=10}

stage1first_param_piv = stage1first_param %>% 
  
  pivot_longer(cols = Correct : Timeouts)

stage1first_param_piv$name = factor(stage1first_param_piv$name, 
                                     
                                     levels = c("Correct", "Omissions", "Accuracy",
                                                
                                                "Prematures", "Timeouts", "Perseveratives"))

head(stage1first_param_piv)

```

```{r, fig.width=12}

ggplot(stage1first_param_piv,
       
       aes(name,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  # geom_text_repel(position = position_jitterdodge(seed = 1,
  #                                      
  #                                      jitter.width = 0.1,
  #                                      
  #                                      jitter.height = 0.1,
  #                                      
  #                                      dodge.width = 0.5)) + 
  
  geom_point(position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25,
               
               position = position_dodge(width = 0.5)) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +

  #scale_color_manual(values = c("blue", "red")) +
  
  #scale_fill_manual(values = c("blue", "red")) +
  
  ggtitle("Stage 1, First Day") +
  xlab("Parameter") +
  ylab("Proportion") +
  
  theme_classic(base_size=22)

```

Let's check last day parameters? 

```{r}

stage1last_param = stage1last %>%
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    Correct = sum(Response == "Correct") / max(Trials),
    
    Omissions = sum(Response == "Omission") / max(Trials),
    
    Accuracy = sum(Response == "Correct") / (sum(Response == "Correct") + sum(Response == "Incorrect")),
    
    Prematures = sum(Response == "Premature") / max(Trials),
    
    Perseveratives = sum(Response == "Perseverative") / max(Trials),
    
    Timeouts = sum(Response == "Timeout") / max(Trials)
    
    ) %>% 
  
  ungroup()


head(stage1last_param)

```

Let's plot day 1 params

```{r, fig.width=10}

stage1last_param_piv = stage1last_param %>% 
  
  pivot_longer(cols = Correct : Timeouts)

stage1last_param_piv$name = factor(stage1last_param_piv$name, 
                                     
                                     levels = c("Correct", "Omissions", "Accuracy",
                                                
                                                "Prematures", "Timeouts", "Perseveratives"))

head(stage1last_param_piv)

```

```{r, fig.width=12}

ggplot(stage1last_param_piv,
       
       aes(name,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  # geom_text_repel(position = position_jitterdodge(seed = 1,
  #                                      
  #                                      jitter.width = 0.1,
  #                                      
  #                                      jitter.height = 0.1,
  #                                      
  #                                      dodge.width = 0.5)) + 
  
  geom_point(position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25,
               
               position = position_dodge(width = 0.5)) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +

  #scale_color_manual(values = c("blue", "red")) +
  
  #scale_fill_manual(values = c("blue", "red")) +
  
  ggtitle("Stage 1, Last Day") +
  xlab("Parameter") +
  ylab("Proportion") +
  
  theme_classic(base_size=22)

```

# Stage 1: net change

```{r}

stage1.net = rbind(
  
  (stage1first_param_piv %>% mutate(Day = "First")),
  
  (stage1last_param_piv %>% mutate(Day = "Last"))
  
)

```

```{r, fig.width=12}

ggplot(stage1.net,
       
       aes(Day,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  
  facet_wrap( ~ name) +
  
  
  geom_hline(yintercept = 0.5, color = "gray") +
  
  
  geom_point(alpha = 0.5 ,
             
             position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.2,
               
               position = position_dodge(width = 0.5)) +
  
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 0.75,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +
  
  
  ggtitle("Stage 1: Change from first to last day") +
  xlab("Day") +
  ylab("Proportion") +
  
  theme_classic(base_size=20)

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage1.net %>% 
              
              filter(name == "Correct")))

```


```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage1.net %>% 
              
              filter(name == "Omissions")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage1.net %>% 
              
              filter(name == "Accuracy")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage1.net %>% 
              
              filter(name == "Prematures")))

```

```{r}

# summary(aov(value ~ (Group*Day) + Error(ID),
#             
#             stage1.net %>% 
#               
#               filter(name == "Timeouts")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage1.net %>% 
              
              filter(name == "Perseveratives")))

```

# Stage1 RT, CV, RL

```{r}

stage1first.RT = stage1first %>%
  
  filter(Response == "Correct") %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    RT.mean = mean(RT*60),
    
    CV = (sd(RT) / mean(RT))*10,
    
    RL = mean(RL)
    
    ) %>% 
  
  ungroup() %>% 
  
  pivot_longer(cols = RT.mean:RL) %>% 
  
  mutate(Day = "First")

head(stage1first.RT)

```

```{r}

stage1last.RT = stage1last %>%
  
  filter(Response == "Correct") %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    RT.mean = mean(RT*60),
    
    CV = (sd(RT) / mean(RT))*10,
    
    RL = mean(RL)
    
    ) %>% 
  
  ungroup() %>% 
  
  pivot_longer(cols = RT.mean:RL) %>% 

  mutate(Day = "Last")

head(stage1first.RT)

```

```{r}

stage1.net.RT = rbind(
  
  stage1first.RT,
  
  stage1last.RT

  )

stage1.net.RT$name = factor(stage1.net.RT$name, 
                                     
                                     levels = c("RT.mean", "CV", "RL"))

```

```{r, fig.width=12}

ggplot(stage1.net.RT,
       
       aes(Day,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  
  facet_wrap( ~ name) +
  
  
  geom_hline(yintercept = 0.5, color = "gray") +
  
  
  geom_point(alpha = 0.5 ,
             
             position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.2,
               
               position = position_dodge(width = 0.5)) +
  
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 0.75,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +
  
  
  ggtitle("Stage1: Change from first to last day") +
  xlab("Day") +
  ylab("Value") +
  
  theme_classic(base_size=20)

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage1.net.RT %>% 
              
              filter(name == "RT.mean")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage1.net.RT %>% 
              
              filter(name == "CV")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage1.net.RT %>% 
              
              filter(name == "RL")))


```

# Stage 2 first v last day

```{r}

stage2 = train %>% 
  
  filter(Stage == 2)

```

```{r}

stage2sum = stage2 %>% 
  
    group_by(Group, Sex, ID, Date) %>% 
  
  summarise(Days = n()) %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(Days = n())

```


```{r}

ggplot(stage2sum,
       
  aes(Group,
      
      Days,
      
     color = Group,
      
      fill = Group,
      
      label = ID)) +
  
  geom_text_repel() + 
  
  geom_point(position = position_jitter(seed = 1,
                                       
                                       width = 0.1,
                                       
                                       height = 0.1)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1) +
  
  # #scale_color_manual(values = c("blue", "red")) +

  theme_classic(base_size=22)

```

```{r}
t.test(Days ~ Group,
       
       stage2sum)
```


```{r}

stage2first = stage2 %>% 
  
  group_by(ID) %>% 
  
  filter(Date == first(Date)) %>% 
  
  ungroup() %>% 
  
  select(Group, Sex, ID, Timestamp, Response, Trials, Port, Target, RT, RL)

stage2last = stage2 %>% 
  
  group_by(ID) %>% 
  
  filter(Date == last(Date)) %>% 
  
  ungroup() %>% 
  
  select(Group, Sex, ID, Timestamp, Response, Trials, Port, Target, RT, RL)

```

Let's check 1st day parameters? 

```{r}

stage2first_param = stage2first %>%
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    Correct = sum(Response == "Correct") / max(Trials),
    
    Omissions = sum(Response == "Omission") / max(Trials),
    
    Accuracy = sum(Response == "Correct") / (sum(Response == "Correct") + sum(Response == "Incorrect")),
    
    Prematures = sum(Response == "Premature") / max(Trials),
    
    Perseveratives = sum(Response == "Perseverative") / max(Trials),
    
    Timeouts = sum(Response == "Timeout") / max(Trials)
    
    ) %>% 
  
  ungroup()


head(stage2first_param)

```

Let's plot day 1 params

```{r, fig.width=10}

stage2first_param_piv = stage2first_param %>% 
  
  pivot_longer(cols = Correct : Timeouts)

stage2first_param_piv$name = factor(stage2first_param_piv$name, 
                                     
                                     levels = c("Correct", "Omissions", "Accuracy",
                                                
                                                "Prematures", "Perseveratives", "Timeouts"))

head(stage2first_param_piv)

```

```{r, fig.width=12}

ggplot(stage2first_param_piv,
       
       aes(name,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  # geom_text_repel(position = position_jitterdodge(seed = 1,
  #                                      
  #                                      jitter.width = 0.1,
  #                                      
  #                                      jitter.height = 0.1,
  #                                      
  #                                      dodge.width = 0.5)) + 
  
  geom_point(position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25,
               
               position = position_dodge(width = 0.5)) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +

  #scale_color_manual(values = c("blue", "red")) +
  
  #scale_fill_manual(values = c("blue", "red")) +
  
  ggtitle("Stage 2, First Day") +
  xlab("Parameter") +
  ylab("Proportion") +
  
  theme_classic(base_size=22)

```

Let's check last day parameters? 

```{r}

stage2last_param = stage2last %>%
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    Correct = sum(Response == "Correct") / max(Trials),
    
    Omissions = sum(Response == "Omission") / max(Trials),
    
    Accuracy = sum(Response == "Correct") / (sum(Response == "Correct") + sum(Response == "Incorrect")),
    
    Prematures = sum(Response == "Premature") / max(Trials),
    
    Perseveratives = sum(Response == "Perseverative") / max(Trials),
    
    Timeouts = sum(Response == "Timeout") / max(Trials)
    
    ) %>% 
  
  ungroup()


head(stage2last_param)

```

Let's plot day 2 params

```{r, fig.width=10}

stage2last_param_piv = stage2last_param %>% 
  
  pivot_longer(cols = Correct : Timeouts)

stage2last_param_piv$name = factor(stage2last_param_piv$name, 
                                     
                                     levels = c("Correct", "Omissions", "Accuracy",
                                                
                                                "Prematures", "Perseveratives", "Timeouts"))

head(stage2last_param_piv)

```

```{r, fig.width=12}

ggplot(stage2last_param_piv,
       
       aes(name,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  # geom_text_repel(position = position_jitterdodge(seed = 1,
  #                                      
  #                                      jitter.width = 0.1,
  #                                      
  #                                      jitter.height = 0.1,
  #                                      
  #                                      dodge.width = 0.5)) + 
  
  geom_point(position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25,
               
               position = position_dodge(width = 0.5)) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +

  #scale_color_manual(values = c("blue", "red")) +
  
  #scale_fill_manual(values = c("blue", "red")) +
  
  ggtitle("Stage 2, Last Day") +
  xlab("Parameter") +
  ylab("Proportion") +
  
  theme_classic(base_size=22)

```

# Stage 2: net change

```{r}

stage2.net = rbind(
  
  (stage2first_param_piv %>% mutate(Day = "First")),
  
  (stage2last_param_piv %>% mutate(Day = "Last"))
  
)

```

```{r, fig.width=12}

ggplot(stage2.net,
       
       aes(Day,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  
  facet_wrap( ~ name) +
  
  
  geom_hline(yintercept = 0.5, color = "gray") +
  
  
  geom_point(alpha = 0.5 ,
             
             position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.2,
               
               position = position_dodge(width = 0.5)) +
  
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 0.75,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +
  
  
  ggtitle("Stage 2: Change from first to last day") +
  xlab("Day") +
  ylab("Proportion") +
  
  theme_classic(base_size=20)

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage2.net %>% 
              
              filter(name == "Correct")))

```


```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage2.net %>% 
              
              filter(name == "Omissions")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage2.net %>% 
              
              filter(name == "Accuracy")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage2.net %>% 
              
              filter(name == "Prematures")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),

            stage2.net %>%

              filter(name == "Timeouts")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage2.net %>% 
              
              filter(name == "Perseveratives")))

```

# stage2 RT, CV, RL

```{r}

stage2first.RT = stage2first %>%
  
  filter(Response == "Correct") %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    RT.mean = mean(RT*60),
    
    CV = (sd(RT) / mean(RT))*10,
    
    RL = mean(RL)
    
    ) %>% 
  
  ungroup() %>% 
  
  pivot_longer(cols = RT.mean:RL) %>% 
  
  mutate(Day = "First")

head(stage2first.RT)

```

```{r}

stage2last.RT = stage2last %>%
  
  filter(Response == "Correct") %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    RT.mean = mean(RT*60),
    
    CV = (sd(RT) / mean(RT))*10,
    
    RL = mean(RL)
    
    ) %>% 
  
  ungroup() %>% 
  
  pivot_longer(cols = RT.mean:RL) %>% 

  mutate(Day = "Last")

head(stage2first.RT)

```

```{r}

stage2.net.RT = rbind(
  
  stage2first.RT,
  
  stage2last.RT

  )

stage2.net.RT$name = factor(stage2.net.RT$name, 
                                     
                                     levels = c("RT.mean", "CV", "RL"))

```

```{r, fig.width=12}

ggplot(stage2.net.RT,
       
       aes(Day,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  
  facet_wrap( ~ name) +
  
  
  geom_hline(yintercept = 0.5, color = "gray") +
  
  
  geom_point(alpha = 0.5 ,
             
             position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.2,
               
               position = position_dodge(width = 0.5)) +
  
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 0.75,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +
  
  
  ggtitle("stage2: Change from first to last day") +
  xlab("Day") +
  ylab("Value") +
  
  theme_classic(base_size=20)

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage2.net.RT %>% 
              
              filter(name == "RT.mean")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage2.net.RT %>% 
              
              filter(name == "CV")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage2.net.RT %>% 
              
              filter(name == "RL")))


```


# Stage 3 first v last day

```{r}

stage3 = train %>% 
  
  filter(Stage == 3)

```

```{r}

stage3sum = stage3 %>% 
  
    group_by(Group, Sex, ID, Date) %>% 
  
  summarise(Days = n()) %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(Days = n())

```


```{r}

ggplot(stage3sum,
       
  aes(Group,
      
      Days,
      
     color = Group,
      
      fill = Group,
      
      label = ID)) +
  
  geom_text_repel() + 
  
  geom_point(position = position_jitter(seed = 1,
                                       
                                       width = 0.1,
                                       
                                       height = 0.1)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1) +
  
  # #scale_color_manual(values = c("blue", "red")) +

  theme_classic(base_size=22)

```

```{r}

t.test(Days ~ Group,
       
       stage3sum)

```


```{r}

stage3first = stage3 %>% 
  
  group_by(ID) %>% 
  
  filter(Date == first(Date)) %>% 
  
  ungroup() %>% 
  
  select(Group, Sex, ID, Timestamp, Response, Trials, Port, Target, RT, RL)

stage3last = stage3 %>% 
  
  group_by(ID) %>% 
  
  filter(Date == last(Date)) %>% 
  
  ungroup() %>% 
  
  select(Group, Sex, ID, Timestamp, Response, Trials, Port, Target, RT, RL)

```

Let's check 1st day parameters? 

```{r}

stage3first_param = stage3first %>%
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    Correct = sum(Response == "Correct") / max(Trials),
    
    Omissions = sum(Response == "Omission") / max(Trials),
    
    Accuracy = sum(Response == "Correct") / (sum(Response == "Correct") + sum(Response == "Incorrect")),
    
    Prematures = sum(Response == "Premature") / max(Trials),
    
    Perseveratives = sum(Response == "Perseverative") / max(Trials),
    
    Timeouts = sum(Response == "Timeout") / max(Trials)
    
    ) %>% 
  
  ungroup()


head(stage3first_param)

```

Let's plot day 1 params

```{r, fig.width=10}

stage3first_param_piv = stage3first_param %>% 
  
  pivot_longer(cols = Correct : Timeouts)

stage3first_param_piv$name = factor(stage3first_param_piv$name, 
                                     
                                     levels = c("Correct", "Omissions", "Accuracy",
                                                
                                                "Prematures", "Perseveratives", "Timeouts"))

head(stage3first_param_piv)

```

```{r, fig.width=12}

ggplot(stage3first_param_piv,
       
       aes(name,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  # geom_text_repel(position = position_jitterdodge(seed = 1,
  #                                      
  #                                      jitter.width = 0.1,
  #                                      
  #                                      jitter.height = 0.1,
  #                                      
  #                                      dodge.width = 0.5)) + 
  
  geom_point(position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25,
               
               position = position_dodge(width = 0.5)) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +

  #scale_color_manual(values = c("blue", "red")) +
  
  #scale_fill_manual(values = c("blue", "red")) +
  
  ggtitle("Stage 3, First Day") +
  xlab("Parameter") +
  ylab("Proportion") +
  
  theme_classic(base_size=22)

```

Let's check last day parameters? 

```{r}

stage3last_param = stage3last %>%
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    Correct = sum(Response == "Correct") / max(Trials),
    
    Omissions = sum(Response == "Omission") / max(Trials),
    
    Accuracy = sum(Response == "Correct") / (sum(Response == "Correct") + sum(Response == "Incorrect")),
    
    Prematures = sum(Response == "Premature") / max(Trials),
    
    Perseveratives = sum(Response == "Perseverative") / max(Trials),
    
    Timeouts = sum(Response == "Timeout") / max(Trials)
    
    ) %>% 
  
  ungroup()


head(stage3last_param)

```

Let's plot day 3 params

```{r, fig.width=10}

stage3last_param_piv = stage3last_param %>% 
  
  pivot_longer(cols = Correct : Timeouts)

stage3last_param_piv$name = factor(stage3last_param_piv$name, 
                                     
                                     levels = c("Correct", "Omissions", "Accuracy",
                                                
                                                "Prematures", "Perseveratives", "Timeouts"))

head(stage3last_param_piv)

```

```{r, fig.width=12}

ggplot(stage3last_param_piv,
       
       aes(name,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  # geom_text_repel(position = position_jitterdodge(seed = 1,
  #                                      
  #                                      jitter.width = 0.1,
  #                                      
  #                                      jitter.height = 0.1,
  #                                      
  #                                      dodge.width = 0.5)) + 
  
  geom_point(position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25,
               
               position = position_dodge(width = 0.5)) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +

  #scale_color_manual(values = c("blue", "red")) +
  
  #scale_fill_manual(values = c("blue", "red")) +
  
  ggtitle("Stage 3, Last Day") +
  xlab("Parameter") +
  ylab("Proportion") +
  
  theme_classic(base_size=22)

```

# Stage 3: net change

```{r}

stage3.net = rbind(
  
  (stage3first_param_piv %>% mutate(Day = "First")),
  
  (stage3last_param_piv %>% mutate(Day = "Last"))
  
)

```

```{r, fig.width=12}

ggplot(stage3.net,
       
       aes(Day,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  
  facet_wrap( ~ name) +
  
  
  geom_hline(yintercept = 0.5, color = "gray") +
  
  
  geom_point(alpha = 0.5 ,
             
             position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.2,
               
               position = position_dodge(width = 0.5)) +
  
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 0.75,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +
  
  
  ggtitle("Stage 3: Change from first to last day") +
  xlab("Day") +
  ylab("Proportion") +
  
  theme_classic(base_size=20)

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage3.net %>% 
              
              filter(name == "Correct")))

```


```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage3.net %>% 
              
              filter(name == "Omissions")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage3.net %>% 
              
              filter(name == "Accuracy")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage3.net %>% 
              
              filter(name == "Prematures")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),

            stage3.net %>%

              filter(name == "Timeouts")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage3.net %>% 
              
              filter(name == "Perseveratives")))

```

# stage3 RT, CV, RL

```{r}

stage3first.RT = stage3first %>%
  
  filter(Response == "Correct") %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    RT.mean = mean(RT*60),
    
    CV = (sd(RT) / mean(RT))*10,
    
    RL = mean(RL)
    
    ) %>% 
  
  ungroup() %>% 
  
  pivot_longer(cols = RT.mean:RL) %>% 
  
  mutate(Day = "First")

head(stage3first.RT)

```

```{r}

stage3last.RT = stage3last %>%
  
  filter(Response == "Correct") %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    RT.mean = mean(RT*60),
    
    CV = (sd(RT) / mean(RT))*10,
    
    RL = mean(RL)
    
    ) %>% 
  
  ungroup() %>% 
  
  pivot_longer(cols = RT.mean:RL) %>% 

  mutate(Day = "Last")

head(stage3first.RT)

```

```{r}

stage3.net.RT = rbind(
  
  stage3first.RT,
  
  stage3last.RT

  )

stage3.net.RT$name = factor(stage3.net.RT$name, 
                                     
                                     levels = c("RT.mean", "CV", "RL"))

```

```{r, fig.width=12}

ggplot(stage3.net.RT,
       
       aes(Day,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  
  facet_wrap( ~ name) +
  
  
  geom_hline(yintercept = 0.5, color = "gray") +
  
  
  geom_point(alpha = 0.5 ,
             
             position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.2,
               
               position = position_dodge(width = 0.5)) +
  
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 0.75,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +
  
  
  ggtitle("stage3: Change from first to last day") +
  xlab("Day") +
  ylab("Value") +
  
  theme_classic(base_size=20)

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage3.net.RT %>% 
              
              filter(name == "RT.mean")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage3.net.RT %>% 
              
              filter(name == "CV")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage3.net.RT %>% 
              
              filter(name == "RL")))


```


# Stage 4 first v last day

```{r}

stage4 = train %>% 
  
  filter(Stage == 4)

```

```{r}

stage4sum = stage4 %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(Days = n())

```


```{r}

ggplot(stage4sum,
       
  aes(Group,
      
      Days,
      
     color = Group,
      
      fill = Group,
      
      label = ID)) +
  
  geom_text_repel() + 
  
  geom_point(position = position_jitter(seed = 1,
                                       
                                       width = 0.1,
                                       
                                       height = 0.1)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1) +
  
  # #scale_color_manual(values = c("blue", "red")) +

  theme_classic(base_size=22)

```

```{r}

Stage4 = train %>% 
  
  filter(Stage == 4)

```


```{r}

stage4first = stage4 %>% 
  
  group_by(ID) %>% 
  
  filter(Date == first(Date)) %>% 
  
  ungroup() %>% 
  
  select(Group, Sex, ID, Timestamp, Response, Trials, Port, Target, RT, RL)

stage4last = stage4 %>% 
  
  group_by(ID) %>% 
  
  filter(Date == last(Date)) %>% 
  
  ungroup() %>% 
  
  select(Group, Sex, ID, Timestamp, Response, Trials, Port, Target, RT, RL)

```

Let's check 1st day parameters? 

```{r}

stage4first_param = stage4first %>%
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    Correct = sum(Response == "Correct") / max(Trials),
    
    Omissions = sum(Response == "Omission") / max(Trials),
    
    Accuracy = sum(Response == "Correct") / (sum(Response == "Correct") + sum(Response == "Incorrect")),
    
    Prematures = sum(Response == "Premature") / max(Trials),
    
    Perseveratives = sum(Response == "Perseverative") / max(Trials),
    
    Timeouts = sum(Response == "Timeout") / max(Trials)
    
    ) %>% 
  
  ungroup()


head(stage4first_param)

```

Let's plot day 1 params

```{r, fig.width=10}

stage4first_param_piv = stage4first_param %>% 
  
  pivot_longer(cols = Correct : Timeouts)

stage4first_param_piv$name = factor(stage4first_param_piv$name, 
                                     
                                     levels = c("Correct", "Omissions", "Accuracy",
                                                
                                                "Prematures", "Perseveratives", "Timeouts"))

head(stage4first_param_piv)

```

```{r, fig.width=12}

ggplot(stage4first_param_piv,
       
       aes(name,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  # geom_text_repel(position = position_jitterdodge(seed = 1,
  #                                      
  #                                      jitter.width = 0.1,
  #                                      
  #                                      jitter.height = 0.1,
  #                                      
  #                                      dodge.width = 0.5)) + 
  
  geom_point(position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25,
               
               position = position_dodge(width = 0.5)) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +

  #scale_color_manual(values = c("blue", "red")) +
  
  #scale_fill_manual(values = c("blue", "red")) +
  
  ggtitle("Stage 4, First Day") +
  xlab("Parameter") +
  ylab("Proportion") +
  
  theme_classic(base_size=22)

```

Let's check last day parameters? 

```{r}

stage4last_param = stage4last %>%
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    Correct = sum(Response == "Correct") / max(Trials),
    
    Omissions = sum(Response == "Omission") / max(Trials),
    
    Accuracy = sum(Response == "Correct") / (sum(Response == "Correct") + sum(Response == "Incorrect")),
    
    Prematures = sum(Response == "Premature") / max(Trials),
    
    Perseveratives = sum(Response == "Perseverative") / max(Trials),
    
    Timeouts = sum(Response == "Timeout") / max(Trials)
    
    ) %>% 
  
  ungroup()


head(stage4last_param)

```

Let's plot day 4 params

```{r, fig.width=10}

stage4last_param_piv = stage4last_param %>% 
  
  pivot_longer(cols = Correct : Timeouts)

stage4last_param_piv$name = factor(stage4last_param_piv$name, 
                                     
                                     levels = c("Correct", "Omissions", "Accuracy",
                                                
                                                "Prematures", "Perseveratives", "Timeouts"))

head(stage4last_param_piv)

```

```{r, fig.width=12}

ggplot(stage4last_param_piv,
       
       aes(name,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  # geom_text_repel(position = position_jitterdodge(seed = 1,
  #                                      
  #                                      jitter.width = 0.1,
  #                                      
  #                                      jitter.height = 0.1,
  #                                      
  #                                      dodge.width = 0.5)) + 
  
  geom_point(position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25,
               
               position = position_dodge(width = 0.5)) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +

  #scale_color_manual(values = c("blue", "red")) +
  
  #scale_fill_manual(values = c("blue", "red")) +
  
  ggtitle("Stage 4, Last Day") +
  xlab("Parameter") +
  ylab("Proportion") +
  
  theme_classic(base_size=22)

```

# Stage 4: net change

```{r}

stage4.net = rbind(
  
  (stage4first_param_piv %>% mutate(Day = "First")),
  
  (stage4last_param_piv %>% mutate(Day = "Last"))
  
)

```

```{r, fig.width=12}

ggplot(stage4.net,
       
       aes(Day,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  
  facet_wrap( ~ name) +
  
  
  geom_hline(yintercept = 0.5, color = "gray") +
  
  
  geom_point(alpha = 0.5 ,
             
             position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.2,
               
               position = position_dodge(width = 0.5)) +
  
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 0.75,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +
  
  
  ggtitle("Stage 4: Change from first to last day") +
  xlab("Day") +
  ylab("Proportion") +
  
  theme_classic(base_size=20)

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage4.net %>% 
              
              filter(name == "Correct")))

```


```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage4.net %>% 
              
              filter(name == "Omissions")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage4.net %>% 
              
              filter(name == "Accuracy")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage4.net %>% 
              
              filter(name == "Prematures")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),

            stage4.net %>%

              filter(name == "Timeouts")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage4.net %>% 
              
              filter(name == "Perseveratives")))

```

# stage4 RT, CV, RL

```{r}

stage4first.RT = stage4first %>%
  
  filter(Response == "Correct") %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    RT.mean = mean(RT*60),
    
    CV = (sd(RT) / mean(RT))*10,
    
    RL = mean(RL)
    
    ) %>% 
  
  ungroup() %>% 
  
  pivot_longer(cols = RT.mean:RL) %>% 
  
  mutate(Day = "First")

head(stage4first.RT)

```

```{r}

stage4last.RT = stage4last %>%
  
  filter(Response == "Correct") %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    RT.mean = mean(RT*60),
    
    CV = (sd(RT) / mean(RT))*10,
    
    RL = mean(RL)
    
    ) %>% 
  
  ungroup() %>% 
  
  pivot_longer(cols = RT.mean:RL) %>% 

  mutate(Day = "Last")

head(stage4first.RT)

```

```{r}

stage4.net.RT = rbind(
  
  stage4first.RT,
  
  stage4last.RT

  )

stage4.net.RT$name = factor(stage4.net.RT$name, 
                                     
                                     levels = c("RT.mean", "CV", "RL"))

```

```{r, fig.width=12}

ggplot(stage4.net.RT,
       
       aes(Day,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  
  facet_wrap( ~ name) +
  
  
  geom_hline(yintercept = 0.5, color = "gray") +
  
  
  geom_point(alpha = 0.5 ,
             
             position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.2,
               
               position = position_dodge(width = 0.5)) +
  
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 0.75,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +
  
  
  ggtitle("stage4: Change from first to last day") +
  xlab("Day") +
  ylab("Value") +
  
  theme_classic(base_size=20)

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage4.net.RT %>% 
              
              filter(name == "RT.mean")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage4.net.RT %>% 
              
              filter(name == "CV")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage4.net.RT %>% 
              
              filter(name == "RL")))


```


# stage 5: programmatic graduation


```{r}

grad5 = train %>% 
  
  filter(Stage == 5) %>% 
  
  arrange(Date) %>% 
  
  group_by(ID, Date) %>% 

  mutate(perc.corr = sum(Response == "Correct") / max(Trials)) %>% 
  
  ungroup() %>% 
  
  group_by(ID) %>% 
  
  mutate(
    
    Session = Session_all - (min(Session_all) - 1)
         
         )  

head(grad5)

```

```{r, fig.height=10}

check5 = grad5 %>%
         
         group_by(ID, Date) %>% 
         
         slice(n()) %>% 
         
         ungroup() %>% 
         
         group_by(ID) %>% 
         
         mutate(
           
           grad = case_when(
             
             perc.corr >= 0.4 &
               dplyr::lag(perc.corr >= 0.4, n = 1) ~ 1,
             
             TRUE ~ 0),
           
           prog = case_when(
             
             lag(grad > 0, n = 1) ~ 1,
             
             TRUE ~ 0
           ),
           
           prog.sum = cumsum(prog)
           
           ) %>% 
  
  ungroup()

```


```{r, fig.height=10}
ggplot(check5 %>% 
         
         filter(prog.sum < 1),
       
       aes(Session,
           
           perc.corr,
           
           color = grad > 0)) +
  
  facet_wrap(~ ID) +
  
  geom_hline(yintercept = 0.40, color = "red") +
  
  geom_col(alpha = 0.1) +
  
  theme_classic()

```

```{r}

passed5 = check5 %>% 
         
         filter(prog.sum < 1,
                
                ID != "1-10-1",
                
                ID != "1-8-2",
                
                ID != "2-16-4",
                
                ID != "2-28-2") %>% 
  
  mutate(combo = paste(ID, Date, sep = "-"))

```


```{r}

prefilter.train.5 = train %>% 
  
  filter(Stage == 5) %>% 
  
  group_by(ID, Date) %>%
  
  mutate(perc.corr = sum(Response == "Correct") / max(Trials)) %>% 
  
  ungroup() %>% 
  
  group_by(ID) %>% 
  
  mutate(Session = Session_all - (min(Session_all) - 1),
         
         combo = paste(ID, Date, sep = "-")) %>% 
  
  ungroup()

```

```{r}

train.pass.5 = prefilter.train.5 %>% 
  
  filter(combo %in% passed5$combo)

  
```

```{r, fig.height=10}

ggplot(train.pass.5 %>% 
         
         group_by(ID, Date) %>% 
         
         slice(n()),
       
       aes(Session,
           
           perc.corr,
           
           color = Group)) +
  
  facet_wrap(~ ID) +
  
  geom_hline(yintercept = 0.40, color = "red") +
  
  geom_col(alpha = 0.1) +
  
  theme_classic()

```

# Stage 5 first v last day

```{r}

stage5 = train.pass.5 %>% 
  
  filter(ID != "1-3-3",
         
         ID != "2-1-2")

```


```{r}

stage5sum = stage5 %>% 

  group_by(Group, Sex, ID, Date) %>% 
  
  summarise(Days = n()) %>% 
    
  group_by(Group, Sex, ID) %>% 
  
  summarise(Days = n())

```


```{r}

ggplot(stage5sum,
       
  aes(Group,
      
      Days,
      
     color = Group,
      
      fill = Group,
      
      label = ID)) +
  
  geom_text_repel() + 
  
  geom_point(position = position_jitter(seed = 1,
                                       
                                       width = 0.1,
                                       
                                       height = 0.1)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1) +
  
  # #scale_color_manual(values = c("blue", "red")) +

  theme_classic(base_size=22)

```

```{r}

t.test(Days ~ Group,
       
       stage5sum)

```


```{r}

stage5first = stage5 %>% 
  
  group_by(ID) %>% 
  
  filter(Date == first(Date)) %>% 
  
  ungroup() %>% 
  
  select(Group, Sex, ID, Timestamp, Response, Trials, Port, Target, RT, RL)

stage5last = stage5 %>% 
  
  group_by(ID) %>% 
  
  filter(Date == last(Date)) %>% 
  
  ungroup() %>% 
  
  select(Group, Sex, ID, Timestamp, Response, Trials, Port, Target, RT, RL)

```

Let's check 1st day parameters? 

```{r}

stage5first_param = stage5first %>%
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    Correct = sum(Response == "Correct") / max(Trials),
    
    Omissions = sum(Response == "Omission") / max(Trials),
    
    Accuracy = sum(Response == "Correct") / (sum(Response == "Correct") + sum(Response == "Incorrect")),
    
    Prematures = sum(Response == "Premature") / max(Trials),
    
    Perseveratives = sum(Response == "Perseverative") / max(Trials),
    
    Timeouts = sum(Response == "Timeout") / max(Trials)
    
    ) %>% 
  
  ungroup()


head(stage5first_param)

```

Let's plot day 1 params

```{r, fig.width=10}

stage5first_param_piv = stage5first_param %>% 
  
  pivot_longer(cols = Correct : Timeouts)

stage4first_param_piv$name = factor(stage4first_param_piv$name, 
                                     
                                     levels = c("Correct", "Omissions", "Accuracy",
                                                
                                                "Prematures", "Perseveratives", "Timeouts"))

head(stage5first_param_piv)

```

```{r, fig.width=12}

ggplot(stage5first_param_piv,
       
       aes(name,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  # geom_text_repel(position = position_jitterdodge(seed = 1,
  #                                      
  #                                      jitter.width = 0.1,
  #                                      
  #                                      jitter.height = 0.1,
  #                                      
  #                                      dodge.width = 0.5)) + 
  
  geom_point(position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25,
               
               position = position_dodge(width = 0.5)) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +

  #scale_color_manual(values = c("blue", "red")) +
  
  #scale_fill_manual(values = c("blue", "red")) +
  
  ggtitle("Stage 5, First Day") +
  xlab("Parameter") +
  ylab("Proportion") +
  
  theme_classic(base_size=22)

```

Let's check last day parameters? 

```{r}

stage5last_param = stage5last %>%
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    Correct = sum(Response == "Correct") / max(Trials),
    
    Omissions = sum(Response == "Omission") / max(Trials),
    
    Accuracy = sum(Response == "Correct") / (sum(Response == "Correct") + sum(Response == "Incorrect")),
    
    Prematures = sum(Response == "Premature") / max(Trials),
    
    Perseveratives = sum(Response == "Perseverative") / max(Trials),
    
    Timeouts = sum(Response == "Timeout") / max(Trials)
    
    ) %>% 
  
  ungroup()


head(stage5last_param)

```

Let's plot day 5 params

```{r, fig.width=10}

stage5last_param_piv = stage5last_param %>% 
  
  pivot_longer(cols = Correct : Timeouts)

stage5last_param_piv$name = factor(stage5last_param_piv$name, 
                                     
                                     levels = c("Correct", "Omissions", "Accuracy",
                                                
                                                "Prematures", "Perseveratives", "Timeouts"))

head(stage5last_param_piv)

```

```{r, fig.width=12}

ggplot(stage5last_param_piv,
       
       aes(name,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  # geom_text_repel(position = position_jitterdodge(seed = 1,
  #                                      
  #                                      jitter.width = 0.1,
  #                                      
  #                                      jitter.height = 0.1,
  #                                      
  #                                      dodge.width = 0.5)) + 
  
  geom_point(position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.25,
               
               position = position_dodge(width = 0.5)) +
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 1,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +

  #scale_color_manual(values = c("blue", "red")) +
  
  #scale_fill_manual(values = c("blue", "red")) +
  
  ggtitle("Stage 5, Last Day") +
  xlab("Parameter") +
  ylab("Proportion") +
  
  theme_classic(base_size=22)

```

# Stage 5: net change

```{r}

stage5.net = rbind(
  
  (stage5first_param_piv %>% mutate(Day = "First")),
  
  (stage5last_param_piv %>% mutate(Day = "Last"))
  
)

```

```{r, fig.width=12}

ggplot(stage5.net,
       
       aes(Day,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  
  facet_wrap( ~ name) +
  
  
  geom_hline(yintercept = 0.5, color = "gray") +
  
  
  geom_point(alpha = 0.5 ,
             
             position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.2,
               
               position = position_dodge(width = 0.5)) +
  
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 0.75,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +
  
  
  ggtitle("Stage 5: Change from first to last day") +
  xlab("Day") +
  ylab("Proportion") +
  
  theme_classic(base_size=20)

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage5.net %>% 
              
              filter(name == "Correct")))

```


```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage5.net %>% 
              
              filter(name == "Omissions")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage5.net %>% 
              
              filter(name == "Accuracy")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage5.net %>% 
              
              filter(name == "Prematures")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),

            stage5.net %>%

              filter(name == "Timeouts")))

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage5.net %>% 
              
              filter(name == "Perseveratives")))

```

# stage5 RT, CV, RL

```{r}

stage5first.RT = stage5first %>%
  
  filter(Response == "Correct") %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    RT.mean = mean(RT*60),
    
    CV = (sd(RT) / mean(RT)),
    
    RL = mean(RL)
    
    ) %>% 
  
  ungroup() %>% 
  
  pivot_longer(cols = RT.mean:RL) %>% 
  
  mutate(Day = "First")

head(stage5first.RT)

```

```{r}

stage5last.RT = stage5last %>%
  
  filter(Response == "Correct") %>% 
  
  group_by(Group, Sex, ID) %>% 
  
  summarise(
    
    RT.mean = mean(RT*60),
    
    CV = (sd(RT) / mean(RT)),
    
    RL = mean(RL)
    
    ) %>% 
  
  ungroup() %>% 
  
  pivot_longer(cols = RT.mean:RL) %>% 

  mutate(Day = "Last")

head(stage5first.RT)

```

```{r}

stage5.net.RT = rbind(
  
  stage5first.RT,
  
  stage5last.RT

  )

stage5.net.RT$name = factor(stage5.net.RT$name, 
                                     
                                     levels = c("RT.mean", "CV", "RL"))

```

```{r, fig.width=12}

ggplot(stage5.net.RT,
       
       aes(Day,
           
           value,
           
           label = ID,
           
           color = Group,
           
           fill = Group)) +
  
  
  facet_wrap( ~ name) +
  
  
  geom_hline(yintercept = 0.5, color = "gray") +
  
  
  geom_point(alpha = 0.5 ,
             
             position = position_jitterdodge(seed = 1,
                                       
                                       jitter.width = 0.1,
                                       
                                       jitter.height = 0.1,
                                       
                                       dodge.width = 0.5)) +
  
  
  stat_summary(fun = "mean",
               
               geom = "col",
               
               width = 0.33,
               
               alpha = 0.2,
               
               position = position_dodge(width = 0.5)) +
  
  
  stat_summary(fun.data = "mean_se",
               
               geom = "errorbar",
               
               size = 0.75,
               
               width = 0.1,
               
               position = position_dodge(width = 0.5),
               
               color = "black") +
  
  
  ggtitle("stage5: Change from first to last day") +
  xlab("Day") +
  ylab("Value") +
  
  theme_classic(base_size=20)

```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage5.net.RT %>% 
              
              filter(name == "RT.mean")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage5.net.RT %>% 
              
              filter(name == "CV")))


```

```{r}

summary(aov(value ~ (Group*Day) + Error(ID),
            
            stage5.net.RT %>% 
              
              filter(name == "RL")))


```


# graduation timeline, by sex

```{r}

stage5 = train.pass.5 %>% 
  
  filter(ID != "1-3-3",
         
         ID != "2-1-2")

```

```{r}

stage5.to6.pre = train %>% 
  
  mutate(
    
    combo = paste(ID, Date, sep = "-"),
    
    grad.check = case_when(
      
      combo %in% stage5$combo ~ FALSE,
      
      TRUE ~ TRUE
      
    ))
    
stage5.to6 = stage5.to6.pre %>% 
  
  mutate(Stage = case_when(
    
    ID %in% stage5$ID & Stage == 5 & grad.check == FALSE ~ 6,
    
    TRUE ~ Stage
    
  ))

```


```{r, fig.width=12, fig.height=8}

ggplot(stage5.to6 %>% 
         
         filter(Session_all < 100) %>% 
         
         group_by(Group, Sex, Stage, ID, Session_all) %>% 
         
         summarise(n = n()),

       aes(Session_all,
           
           as.integer(Stage),
           
           fill = Group
           
           )) +
  
  facet_wrap(~ Sex) +

  geom_jitter(aes(color = Group),
              height = 0.1,
              width = 0.1) +

  stat_summary(aes(fill = Group),
               fun = "mean", geom = "line",
               size = 1) +

  stat_summary(fun.data = "mean_se", geom = "ribbon",
               alpha = 0.25) +
  
  scale_y_continuous(breaks = seq(0, 6, by = 1)) +
  
  # scale_color_manual(values = c("dodgerblue", "orange", "green2"))+
  #scale_fill_manual(values = c("dodgerblue", "orange", "green"))+

  ggtitle("Progress across training stages") +
  ylab("Stage of Training") +
  xlab("Day of Training") +

  theme_classic(base_size=22) +
  
  theme(axis.text = element_text(size = 14),
        strip.text.x = element_text(size = 16),
        title = element_text(size = 16)
        )

```

# graduation timeline

```{r, fig.width=10, fig.height=8}

ggplot(stage5.to6 %>% 
         
         filter(Session_all < 100) %>% 
         
         group_by(Group, Sex, Stage, ID, Session_all) %>% 
         
         summarise(n = n()),

       aes(Session_all,
           
           as.integer(Stage),
           
           fill = Group
           
           )) +
  
  # facet_wrap(~ Sex) +

  geom_jitter(aes(color = Group),
              height = 0.1,
              width = 0.1) +

  stat_summary(aes(fill = Group),
               fun = "mean", geom = "line",
               size = 1) +

  stat_summary(fun.data = "mean_se", geom = "ribbon",
               alpha = 0.25) +
  
  scale_y_continuous(breaks = seq(0, 6, by = 1)) +
  
  # scale_color_manual(values = c("dodgerblue", "orange", "green2"))+
  #scale_fill_manual(values = c("dodgerblue", "orange", "green"))+

  ggtitle("Progress across training stages") +
  ylab("Stage of Training") +
  xlab("Day of Training") +

  theme_classic(base_size=22) +
  
  theme(axis.text = element_text(size = 14),
        strip.text.x = element_text(size = 16),
        title = element_text(size = 16)
        )

```

```{r}

summary(aov(Stage ~ (Group * Session_all) + Error(ID),
            
            data = stage5.to6 %>%
              
              filter(Session_all < 20)))

```

```{r}

train.lmer40 = lme4::lmer(Stage ~ Group * Session_all + (1 | ID),
     
     data = stage5.to6 %>%
       
              group_by(ID, Session_all) %>% 
       
       slice(n()) %>% 
       
       ungroup() %>% 
       
       filter(Session_all < 41))

```


```{r}

summary(train.lmer40)

```


```{r}

car::Anova(train.lmer40)

```