2_sol-ichart-processing.Rmd

---
title: "SOL Data Processing"
author: "Kyle MacDonald"
date: "February 5, 2016"
output: html_document
---

This script takes in raw eye movement data and processes it. The outputs are:

1. Data frame with Accuracy and RT measures at the participant-level.
4. Data frame with Accuracy and RT measures at the trial-level.
2. Data frame with cleaned eye movement data for plotting.
3. Data frame with exclusionary information.

# Data processing 

```{r chunk_options}
rm(list=ls())
knitr::opts_chunk$set(warning=FALSE, message=FALSE, sanitize = T, 
                      fig.height=8, fig.width=10, echo=F, cache = T)
```

```{r}
source("../helper_functions//libraries_v_3.6.R")
source("../helper_functions//useful.R")
library(magrittr); library(kmr); library(tidyverse)
```

## Read in data

```{r}
# Demographic data
sol_demo <- read.csv("../../data/demographics/sol_demo_all.csv", stringsAsFactors = F) %>% 
  mutate(Sub.Num = as.character(Sub.Num))

sol_demo$Sub.Num <- as.character(sol_demo$Sub.Num)

# eye movement data
iChart <- read.csv("../../data/processed_data/processed_em_data/sol-ichart-merged-post-gating.csv", 
                   check.names=F, stringsAsFactors=F) %>%  
  mutate(Sub.Num =  as.character(Sub.Num))

# Stimuli information
sol_target_signs_df <- read.csv("../../data/demographics/target_sign_lengths_all.csv")
```

## Filter dataset

Filter out trials/participants that should not go into analyses based on preset exclusionary criteria: a) didn’t know signs in the tas and b) not enough ASL exposur.

```{r}
include <- sol_demo %>%
  filter(include == "yes") %>% 
  select(Sub.Num, include,
         reason_excluded, stimuli,
         age_code, signs_produced, 
         hearing_status_participant) %>% 
  mutate(Sub.Num = as.character(Sub.Num))

iChart <- left_join(iChart, include, by = c("Sub.Num", "stimuli")) %>% 
  filter(include == "yes")
```

## Remove unknown signs

Make a clean target signs variable for filtering later on.

```{r}
targets <- c("juice", "cookie", "cup", "ball", "shoe", "kitty",
             "doll", "teddy", "book", "birdy", "car", "sock")

make_clean_target <- function (target_image, targets) {
  target_img_clean <- targets[str_detect(target_image, targets)]
  return(target_img_clean)
}

iChart$clean_target_img <- unlist(sapply(iChart$Target.Image, 
                                         function (x) make_clean_target(x, targets)))
```

Add unknown signs variable. Taken from the sol_demo data frame.

```{r}
ss_unknown_signs <- sol_demo %>% 
  filter(include == "yes") %>% 
  select(Sub.Num, parent_report_unknown_signs) %>% 
  mutate(parent_report_unknown_signs = ifelse(is.na(parent_report_unknown_signs) == T,
                                              "none", parent_report_unknown_signs))

iChart <- left_join(iChart, ss_unknown_signs, by = "Sub.Num")
```

Now we can filter the iChart, removing the "unknown" signs.

```{r}
# get number of trial for each participant prior to filtering
ss_trials_df <- iChart %>% 
  group_by(Sub.Num, parent_report_unknown_signs) %>% 
  dplyr::summarise(Trials = n()) 
```

Flag and remove unknown trials

```{r}
iChart %<>% 
  mutate(unknown_sign = str_detect(string = parent_report_unknown_signs, 
                                   pattern = clean_target_img)) %>% 
  filter(unknown_sign == FALSE)
```

Check to make sure our filtering worked correctly:

```{r}
post_filter_n <- iChart %>% 
  group_by(Sub.Num) %>% 
  dplyr::summarise(Trials = n()) %>% 
  select(Trials)

ss_trials_df$post_filter <- post_filter_n$Trials

ss_trials_df %<>% mutate(trials_removed = as.integer(Trials) - as.integer(post_filter))

# Get total number of trials removed because of unknown signs
sum(ss_trials_df$trials_removed)
```

## Remove prescreened out trials

```{r}
ss_prescreened <- iChart %>% 
  group_by(Sub.Num) %>% 
  filter(Prescreen.Notes != "") %>% 
  dplyr::summarise(num_prescreened = n())

ss_trials_df <- left_join(ss_trials_df, ss_prescreened, by = "Sub.Num") 

ss_trials_df %<>% 
  mutate(num_prescreened = as.numeric(num_prescreened),
    num_prescreened = ifelse(is.na(num_prescreened), 0, num_prescreened),
    good_trials = Trials - (trials_removed + num_prescreened))
```

Filter

```{r}
iChart <- filter(iChart, Prescreen.Notes == "")
```

## Remove participants for having too few trials

We define too few trials as less than or equal to 25% of the total number of
trials in the task.

```{r}
total_trials <- 32
trials_cut_point <- total_trials * .25

trials_filter <- ss_trials_df %>% 
  mutate(exclude_few_trials = ifelse(good_trials <= trials_cut_point, 
                                     "exclude", "include")) %>% 
  select(Sub.Num, good_trials, exclude_few_trials)

# get the number of participants removed by filter
trials_filter %>% group_by(exclude_few_trials) %>% dplyr::summarise(n())

# merge filtering information with iChart
iChart <- left_join(iChart, trials_filter, by = "Sub.Num")

# now filter
iChart <- filter(iChart, exclude_few_trials == "include")
```

### Get median split by age

```{r}
kids_age_descriptives <- iChart %>%
  filter(age_code == "child") %>%
  select(Sub.Num, Months) %>% 
  distinct() %>% 
  dplyr::summarise(median(Months), 
                   max(Months),
                   min(Months), 
                   sd(Months),
                   n()) 

# add median split variable to iChart
iChart$age_group <- ifelse(iChart$Months < kids_age_descriptives$`median(Months)`, 
                           "< 27 Months", 
                           ifelse(iChart$Months >= kids_age_descriptives$`median(Months)` & 
                                    iChart$Months <= kids_age_descriptives$`max(Months)`, 
                                  ">= 27 Months",
                                  "Adults"))

iChart %<>% mutate(age_group_collapsed = ifelse(age_group == "Adults", "Adults", "Kids"))

gender_df <- iChart %>%
  filter(age_code == "child") %>%
  select(Sub.Num, Sex, age_group) %>% 
  unique() %>% 
  group_by(Sex, age_group) %>% 
  dplyr::summarise(count = n()) 

iChart %>% group_by(age_group) %>% 
  dplyr::summarise(count = n_distinct(Sub.Num),
                   mean_val = mean(Months), 
                   min_val = min(Months), 
                   max_val = max(Months)) %>% 
  knitr::kable()
```

### Process iChart

First, we need to process the data, keeping only those trials on which the child was looking at the signer at F0.

* C: Center
* D: Distractor
* T: Target
* A: Away

includeOffCenter == FALSE -> only include trials child was looking at center at F0

includeOffCenter == TRUE -> include trials child was looking at center, target, or distractor at F0

```{r}
iChart %>% group_by("0", Response) %>% dplyr::summarise(Trials = n())

# change all trials to "Vanilla" 
iChart$Condition <- "Vanilla"

## define critical onset, change Cs to Ds and everything else to As
iChart <- defineOnsetSOL(iChart, critonset=0, end_critonset=300, 
                         includeOffCenter = FALSE, includeWindow = FALSE)

iChart %>% group_by("0", Response) %>% dplyr::summarise(Trials = n())
```

### Flag C-T and C-D Trials

Datawiz does not tell us which shifts land on a target vs. a disctractor. So we need to use a function that flags each trial as one of the following:

* C_T: center to target
* C_D: center to distractor
* C-C: center to center (child leaves the signer, goes away, and comes back to signer)
* no_shift
* off_signer

```{r}
# apply fun to each row in our dataset to flag trial type
trial_types <- apply(iChart, 1, trial_type_fun, start = "0", end = "5000") 

# merge trial type information with iChart
iChart <- cbind(iChart, trial_types)

iChart %>% group_by(trial_types) %>% dplyr::summarise(Trials = n())
```

Next, we compute statistics over long window 0-5000 ms. This will allow us to see a 
distribution of RTs, which we will use to determine our analysis window.

First, for adults.

```{r get RT analysis window adults, warning=F, message=F, fig.height = 5}
iChart_adults <- filter(iChart, age_group == "Adults")
iChart_adults <- computeStatistics(iChart_adults, startWindow=0, endWindow=4000)

# get analyisis window where 90% of RTs occur for kids, include all shifts
rts_adults <- filter(iChart_adults, trial_types %in% c("C_T"))

analysis.window_adults <- quantile(rts_adults$RT, probs=c(0.05, 0.95), na.rm=T)
```

```{r}
ggplot(aes(RT), data = rts_adults) +
  geom_histogram() +
  labs(x = "RT (msec)",
       title = "Analysis Window (Adults)") +
  geom_vline(xintercept = analysis.window_adults[1], col="red", 
             lwd=1.5, linetype = "dashed") +
  geom_vline(xintercept = analysis.window_adults[2], col="red", 
             lwd=1.5, linetype = "dashed") +
  annotate("text", x = 2500, y = 30, 
           label = "Analysis Window \n (90% RTs)")
```

Now do the same for kids.

```{r get rt analysis window kids, warning=F, message=F, fig.height = 5}
iChart_kids <- filter(iChart, age_group != "Adults")
iChart_kids <- computeStatistics(iChart_kids, startWindow=0, endWindow=4000)

# get analyisis window where 90% of RTs occur for kids, include all shifts
rts <- filter(iChart_kids, trial_types %in% c("C_T"), age_group != "Adults")

analysis.window <- quantile(rts$RT, probs=c(0.05, 0.95), na.rm=T)
```

```{r}
ggplot(aes(x=RT), data=rts)  +
  geom_histogram() +
  labs(x = "RT (msec)",
       title = "Analysis Window (Kids)") +
  geom_vline(xintercept=analysis.window[1], col="red", lwd=1.5, linetype = "dashed") +
  geom_vline(xintercept=analysis.window[2], col="red", lwd=1.5, linetype = "dashed") +
  annotate("text", x = 3000, y = 30, 
           label = "Analysis Window \n (90% RTs)")
```

Compute statistics over analysis window: 0 to analysis window + 500ms. We add 500ms because it allows us to include trials in which the participant initiates and completes a shift at the very end of the window.

```{r, message=F, warning=F}
iChart <- computeStatistics(iChart, startWindow=0, 
                            endWindow=analysis.window[2]+500)

iChart_kids <- computeStatistics(iChart_kids, startWindow=0,
                                 endWindow=analysis.window[2]+500)

iChart_adults <- computeStatistics(iChart_adults, startWindow=0,
                                   endWindow=analysis.window_adults[2]+500)
```

Reject trials with really long RTs and with long gaps. Gaps are defined as 
a sequence of frames when the child is not looking at either picture or 
at the signer.

```{r RT filters}
# filter for both kids and adults
iChart <- filteriChart(iChart, minRT = analysis.window[1], 
                       maxRT = analysis.window[2], 
                       maxfirstgap=15, maxlonggap=15)  

# filter for kids
iChart_kids <- filteriChart(iChart_kids, minRT = analysis.window[1], 
                            maxRT = analysis.window[2], 
                            maxfirstgap=15, maxlonggap=15) 

# filter for adults
iChart_adults <- filteriChart(iChart_adults, minRT = analysis.window_adults[1], 
                              maxRT = analysis.window_adults[2], 
                              maxfirstgap=15, maxlonggap=15) 
```

### Compute mean accuracy and RT for each participant 

```{r, warning=F, message=F, eval = F}
acc_ss <- poolData(meanAccuracy(iChart_kids, startWindowAcc=analysis.window[1], 
                                endWindowAcc=analysis.window[2]), 
                   RejectFirstGap=TRUE,RejectLongestGap=TRUE, 
                   RejectRT=FALSE, color=TRUE, dependent="Accuracy", 
                   group="", facet="", dodge="", 
                   xlab="", ylab= "Proportion\n  Looking\n  to target", 
                   paired=TRUE, miny = 0.2, maxy = 0.80, 
                   size=13, legend.direction="horizontal", 
                   legend.position="bottom", 
                   breaks=c(0.25, 0.50, 0.75))

rt_ss <- poolData(iChart_kids, 
                  RejectFirstGap=TRUE, RejectLongestGap=TRUE,
                  RejectRT=TRUE, color=FALSE, dependent="RT", group="trial_types", 
                  facet="", dodge="Response",
                  xlab="", ylab="mean RT (ms)", 
                  paired=TRUE, 
                  miny = 400, maxy=1300, 
                  size=13, 
                  legend.direction = "horizontal", 
                  legend.position="bottom", 
                  breaks=c(400, 800, 1200))

acc_ss_adults <- poolData(meanAccuracy(iChart_adults, startWindowAcc=300, 
                                       endWindowAcc=analysis.window_adults[2]), 
                          RejectFirstGap=TRUE,RejectLongestGap=TRUE, 
                          RejectRT=FALSE, color=TRUE, dependent="Accuracy", 
                          group="", facet="", dodge="", 
                          xlab="", ylab= "Proportion\n  Looking\n  to target", 
                          paired=TRUE, miny = 0.2, maxy = 0.80, 
                          size=13, legend.direction="horizontal", 
                          legend.position="bottom", 
                          breaks=c(0.25, 0.50, 0.75))

rt_ss_adults <- poolData(iChart_adults, 
                         RejectFirstGap=TRUE, RejectLongestGap=TRUE,
                         RejectRT=TRUE, color=FALSE, dependent="RT", group="trial_types", 
                         facet="", dodge="Response",
                         xlab="", ylab="mean RT (ms)", 
                         paired=TRUE, 
                         miny = 400, maxy=1300, 
                         size=13, 
                         legend.direction = "horizontal", 
                         legend.position="bottom", 
                         breaks=c(400, 800, 1200))
```

### Compute Acc and RT for each participant by age group 

```{r, warning=F, message=F, eval = F}
acc_hearing_status <- poolData(meanAccuracy(iChart_kids, startWindowAcc=analysis.window[1],
                                            endWindowAcc=analysis.window[2]), 
                               RejectFirstGap=TRUE,RejectLongestGap=TRUE, 
                               RejectRT=FALSE, color=TRUE, dependent="Accuracy", 
                               group="hearing_status_participant", facet="", dodge="", 
                               xlab="", ylab= "Proportion\n  Looking\n  to target", 
                               paired=TRUE, miny = 0.2, maxy = 0.80, 
                               size=13, legend.direction="horizontal", 
                               legend.position="bottom", 
                               breaks=c(0.25, 0.50, 0.75))

rt_hearing_status <- poolData(iChart_kids, 
                              RejectFirstGap=TRUE, RejectLongestGap=TRUE,
                              RejectRT=TRUE, color=FALSE, dependent="RT",
                              group="hearing_status_participant", 
                              facet="", dodge="Response",
                              xlab="", ylab="mean RT (ms)", 
                              paired=TRUE, 
                              miny = 400, maxy=1300, 
                              size=13, 
                              legend.direction = "horizontal", 
                              legend.position="bottom", 
                              breaks=c(400, 800, 1200))

acc <- poolData(meanAccuracy(iChart_kids, startWindowAcc = analysis.window[1], 
                             endWindowAcc = analysis.window[2]), 
                RejectFirstGap=TRUE,RejectLongestGap=TRUE, 
                RejectRT=FALSE, color=TRUE, dependent="Accuracy", 
                group="age_group", facet="", dodge="", 
                xlab="", ylab= "Proportion\n  Looking\n  to target", 
                paired=TRUE, miny = 0.2, maxy = 0.80, 
                size=13, legend.direction="horizontal", 
                legend.position="bottom", 
                breaks=c(0.25, 0.50, 0.75))

rt <- poolData(iChart_kids, 
               RejectFirstGap=TRUE, RejectLongestGap=TRUE,
               RejectRT=TRUE, color=FALSE, dependent="RT", group="age_group", 
               facet="", dodge="Response",
               xlab="", ylab="mean RT (ms)", 
               paired=TRUE, 
               miny = 400, maxy=1300, 
               size=13, 
               legend.direction = "horizontal", 
               legend.position="bottom", 
               breaks=c(400, 800, 1200))


acc_adults <- poolData(meanAccuracy(iChart_adults, startWindowAcc=300,
                                    endWindowAcc=analysis.window_adults[2]), 
                       RejectFirstGap=TRUE,RejectLongestGap=TRUE, 
                       RejectRT=FALSE, color=TRUE, dependent="Accuracy", 
                       group="age_group", facet="", dodge="", 
                       xlab="", ylab= "Proportion\n  Looking\n  to target", 
                       paired=TRUE, miny = 0.2, maxy = 0.80, 
                       size=13, legend.direction="horizontal", 
                       legend.position="bottom", 
                       breaks=c(0.25, 0.50, 0.75))


rt_adults <- poolData(iChart_adults, 
                      RejectFirstGap=TRUE, RejectLongestGap=TRUE,
                      RejectRT=TRUE, color=FALSE, dependent="RT", group="age_group", 
                      facet="", dodge="Response",
                      xlab="", ylab="mean RT (ms)", 
                      paired=TRUE, 
                      miny = 400, maxy=1300, 
                      size=13, 
                      legend.direction = "horizontal", 
                      legend.position="bottom", 
                      breaks=c(400, 800, 1200))
```

### Tanenhaus computation

Here we analyze window accuracy based on proportion looking to either target, distractor, or signer. Since there are three fixations points, chance looking behavior is 0.33.

First, we grab just the eye movement data and age group information.

```{r}
ss_iChart <- iChart %>% 
  filter(GoodFirstGap == T, GoodLongestGap == T) %>% 
  select(Sub.Num, Tr.Num, age_group, Response, `0`:`4000`) 
```

Convert the data to long format.

```{r}
ss_iChart_long <- ss_iChart %>% 
  gather(key = Time.ms, value = value, `0`:`4000`) %>% 
  filter(value %in% c("0", "1", "0.5", ".", "-")) %>%
  mutate(value = ifelse(value == "-", ".", value),
         value_cat = factor(value, labels = c("Away", "Distractor", "Signer", "Target")),
         Time.ms_numeric = to.n(Time.ms)) 
```    

Get proportion looking to Target, Distractor, Away, and Signer at each time slice.

```{r}
## Include all types of looking
ms_iChart_count <- ss_iChart_long %>% 
  filter(value %in% c("0", "1", ".", "0.5")) %>%  ## specify acc computation
  group_by(Sub.Num, Time.ms, age_group, value_cat) %>% 
  dplyr::summarise(count = ifelse(n() == 0, 0, n())) %>% 
  dplyr::summarise(sum_count = sum(count))

ms_iChart <- as.data.frame(xtabs(~ Sub.Num + value_cat + Time.ms, 
                                 data = filter(ss_iChart_long, 
                                               value %in% c("0", "1", ".", "0.5")))) %>% 
  left_join(x = ms_iChart_count, by = c("Sub.Num", "Time.ms")) %>% 
  mutate(proportion_looking = Freq / sum_count,
         age_group_factor = as.factor(age_group))

## Only include Target and Distractor looking
ms_iChart_count_TD <- ss_iChart_long %>% 
  filter(value %in% c("0", "1", ".")) %>%  ## specify acc computation
  group_by(Sub.Num, Time.ms, age_group, value_cat) %>% 
  dplyr::summarise(count = ifelse(n() == 0, 0, n())) %>% 
  dplyr::summarise(sum_count = sum(count))


ms_iChart_TD <- as.data.frame(xtabs(~ Sub.Num + value_cat + Time.ms, 
                                    data = filter(ss_iChart_long, 
                                                  value %in% c("0", "1", ".")))) %>% 
  left_join(x = ms_iChart_count_TD, by = c("Sub.Num", "Time.ms")) %>% 
  mutate(proportion_looking = Freq / sum_count,
         age_group_factor = as.factor(age_group))
```

Now we can compute mean proportion looking to signer, target, and distractor for each participant and for each age group.

```{r}
ss_tanenhaus_comp <- ms_iChart %>% 
  filter(to.n(Time.ms) >= analysis.window[1], to.n(Time.ms) <= analysis.window[2]) %>% 
  group_by(Sub.Num, value_cat, age_group) %>% 
  dplyr::summarise(mean_prop_looking_signer = mean(proportion_looking, na.rm = T))

ss_tanenhaus_comp_TD <- ms_iChart_TD %>% 
  filter(to.n(Time.ms) >= analysis.window[1], to.n(Time.ms) <= analysis.window[2]) %>% 
  group_by(Sub.Num, value_cat, age_group) %>% 
  dplyr::summarise(mean_prop_looking_TD = mean(proportion_looking, na.rm = T))
```

## Create summary table at the participant-level

Get mean accuracy and rt for each participant.

Some munging to get data frame for analysis. Variables needed for each subject: 

* Mean acc 
* Mean rt 
* Signs produced 
* Age 
* Age condition

```{r bind summary statistics}
ss_acc <- bind_rows(acc_ss, acc_ss_adults)
ss_rt <- bind_rows(rt_ss, rt_ss_adults)

# merge acc/rt van1
ss <- left_join(ss_acc, ss_rt, by="Sub.Num")

# merge with demo info
ss <- left_join(ss, filter(sol_demo, include=="yes"), by="Sub.Num")

# clean up variable names in data frame
names(ss)[names(ss)=="Vanilla"] <- "mean_accuracy"
names(ss)[names(ss)=="Vanilla_C_D_D"] <- "mean_incorrect_rt"
names(ss)[names(ss)=="Vanilla_C_T_D"] <- "mean_correct_rt"
```

Add in Tanenhaus proportion looking computation for each participant.

```{r}
ss <- ss_tanenhaus_comp %>% 
  filter(value_cat %in% c("Target", "Distractor", "Signer")) %>% 
  select(Sub.Num, value_cat, age_group, mean_prop_looking_signer, value_cat) %>% 
  dplyr::rename(prop_looking = mean_prop_looking_signer) %>% 
  left_join(ss, ., by = c("Sub.Num"))

ss <- ss_tanenhaus_comp_TD %>% 
  filter(value_cat %in% c("Target", "Distractor", "Signer")) %>% 
  select(Sub.Num, value_cat, age_group, mean_prop_looking_TD, value_cat) %>% 
  left_join(ss, ., by = c("Sub.Num", "value_cat", "age_group"))
```

## First Shift Analysis

First we need to categorize each trial type based on the analysis window.

```{r}
trial_types_analysis_kids <- apply(iChart_kids, 1, trial_type_fun, 
                                   start = analysis.window[1], 
                                   end = analysis.window[2])

# merge trial type information with iChart
iChart_kids <- cbind(iChart_kids, trial_types_analysis_kids)
```

Then we add this information to the summary data frame.

```{r}
# kids 
ss_first_shifts <- iChart_kids %>%
  filter(trial_types_analysis_kids %in% c("C_T", "C_D", "no_shift")) %>% 
  group_by(Sub.Num, trial_types, age_group, Months) %>% 
  dplyr::summarise(count = n()) %>% 
  spread(trial_types, count) %>% 
  ungroup() %>%     
  mutate(C_D = ifelse(is.na(C_D), 0, C_D),
         no_shift = ifelse(is.na(no_shift), 0, no_shift),
         total_trials_shifting = C_D + C_T,
         total_trials = C_D + C_T + no_shift,
         C_D_prop = round(C_D / total_trials_shifting, 2),
         C_T_prop = round(C_T / total_trials_shifting, 2),
         C_D_prop_all = round(C_D / total_trials, 2),
         C_T_prop_all = round(C_T / total_trials, 2))

# add adults
ss_first_shifts <- iChart_adults %>%
  filter(trial_types %in% c("C_T", "C_D", "no_shift")) %>% 
  group_by(Sub.Num, trial_types, age_group, Months) %>% 
  dplyr::summarise(count = n()) %>% 
  spread(trial_types, count) %>% 
  ungroup() %>% 
  mutate(C_D = ifelse(is.na(C_D), 0, C_D),
         no_shift = ifelse(is.na(no_shift), 0, C_D),
         total_trials_shifting = C_D + C_T,
         total_trials = C_D + C_T + no_shift,
         C_D_prop = round(C_D / total_trials_shifting, 2),
         C_T_prop = round(C_T / total_trials_shifting, 2),
         C_D_prop_all = round(C_D / total_trials, 2),
         C_T_prop_all = round(C_T / total_trials, 2)) %>% 
  bind_rows(ss_first_shifts)

# add shifts to full data frame
ss <- ss_first_shifts %>% 
  dplyr::rename(Sub.Num = Sub.Num, C_D_count = C_D, C_T_count = C_T) %>% 
  left_join(ss, by = c("Sub.Num", "age_group"))

# flag chance first shifters = more than 50% posterior mass on guessing strategy
# in latent mixture model
ss <- ss %>% 
  mutate(C_D_prop = ifelse(C_T_prop == 1, 0, C_D_prop),
         age_group_collapsed = ifelse(age_group == "Adults", "Adults", "Kids"))
```

## Median RT 

```{r}
ss_medians <- iChart %>% 
  filter(trial_types == "C_T", GoodRT == T) %>% 
  group_by(Sub.Num, age_group, age_group_collapsed, Months) %>% 
  dplyr::summarise(median_ct_rt = median(RT, na.rm = T))

ss <- ss_medians %>% 
  ungroup() %>% 
  select(Sub.Num, median_ct_rt) %>% 
  left_join(x = ss, by = "Sub.Num")
```

```{r}
ss %>% 
  filter(age_group_collapsed == "Kids") %>% 
  select(Sub.Num) %>% 
  unique() %>% 
  nrow()
```

```{r}
write.csv(ss, "../../data/processed_data/summary_tables/sol_ss_all_600_2500.csv", 
          row.names = F)
```

# Profile Plot Processing

## Get confidence intervals for graph values

```{r, warning=F, message=F, eval=F}
# keeping age bins
createPlots(iChart_kids, startWindow=0, endWindow=4000, RejectLongestGap=FALSE, 
            RejectFirstGap=FALSE, RejectRT=FALSE, color=TRUE, smooth=400, 
            targetEnd=800, carrier="", targets=c(""), 
            group="age_group",  plotStats="PP", miny = 0.4, maxy=0.95, size=15, 
            legend.direction = "vertical", legend.position=c(0.85, 0.9), 
            breaks=c(0.25, 0.50, 0.75), x.target=0.33)

# keeping age bins
createPlots(iChart_adults, startWindow=0, endWindow=4000, RejectLongestGap=FALSE, 
            RejectFirstGap=FALSE, RejectRT=FALSE, color=TRUE, smooth=400,  
            targetEnd=800, carrier="", targets=c(""), 
            group="age_group",  plotStats="PP", miny = 0.4, maxy=0.95, size=15, 
            legend.direction = "vertical", legend.position=c(0.85, 0.9), 
            breaks=c(0.25, 0.50, 0.75), x.target=0.33)

# keeping target signs and separating by stimuli
iChart_v1 <- filter(iChart_kids, stimuli == "V1")
iChart_v2 <- filter(iChart_kids, stimuli == "V2")

createPlots(iChart_v1, startWindow=0, endWindow=4000, RejectLongestGap=FALSE, 
            RejectFirstGap=FALSE, RejectRT=FALSE, color=TRUE, smooth=400, 
            targetEnd=800, carrier="V1", targets=c(""), 
            group="clean_target_img",  plotStats="PP", miny = 0.4, maxy=0.95, size=15, 
            legend.direction = "vertical", legend.position=c(0.85, 0.9), 
            breaks=c(0.25, 0.50, 0.75), x.target=0.33)

createPlots(iChart_v2, startWindow=0, endWindow=4000, RejectLongestGap=FALSE, 
            RejectFirstGap=FALSE, RejectRT=FALSE, color=TRUE, smooth=400, 
            targetEnd=800, carrier="V2", targets=c(""), 
            group="clean_target_img",  plotStats="PP", miny = 0.4, maxy=0.95, size=15, 
            legend.direction = "vertical", legend.position=c(0.85, 0.9), 
            breaks=c(0.25, 0.50, 0.75), x.target=0.33)
```

## Data processing Tanenhaus-style Plot

First, we munge the raw iChart data.

Grab just the eye movement data and group information

```{r}
ss_iChart <- iChart %>% 
  filter(Response == "D", GoodFirstGap == T, GoodLongestGap == T) %>% 
  select(Sub.Num, Tr.Num, stimuli, age_group_collapsed, Response, RT, trial_types, 
         clean_target_img, `0`:`4000`) 
```

Convert to long format

```{r}
ss_iChart_long <- ss_iChart %>% 
  gather(key = Time.ms, value = value, `0`:`4000`) %>% 
  filter(value %in% c("0", "0.5", "1")) %>% 
  mutate(value_cat = factor(value, labels = c("Distractor", "Signer", "Target")),
         Time.ms_numeric = to.n(Time.ms)) 
```    

Summarize for each participant - get proportion looking at each time slice

```{r}
ms_iChart_count <- ss_iChart_long %>% 
  group_by(Sub.Num, Time.ms, age_group_collapsed, value_cat) %>% 
  dplyr::summarise(count = ifelse(n() == 0, 0, n())) %>% 
  dplyr::summarise(sum_count = sum(count))

ms_iChart <- as.data.frame(xtabs(~ Sub.Num + value_cat + Time.ms, 
                                 data = ss_iChart_long)) %>% 
  left_join(y = ms_iChart_count, by = c("Sub.Num", "Time.ms")) %>% 
  mutate(proportion_looking = Freq / sum_count,
         age_group_factor = as.factor(age_group_collapsed))
```

Get means and CIs for proportion looking at each time slice across particpants

```{r}
ms_mean_iChart <- ms_iChart %>% 
  group_by(Time.ms, age_group_collapsed, value_cat) %>% 
  dplyr::summarise(mean_prop_looking = mean(proportion_looking),
                   ci_low = ci.low(proportion_looking),
                   ci_high = ci.high(proportion_looking))
```

Write graph values to CSV.

```{r}
write.csv(x = ms_mean_iChart, 
          file = "../../data/processed_data/graph_values/sol_gvals_all.csv", 
          row.names = F)
```

## Tanenhaus plot codas vs. deaf

```{r coda-deaf-plot-munging}
# grab just the eye movement data and group information
ss_iChart_coda <- iChart %>% 
  filter(age_group != "Adults", Response == "D", GoodFirstGap == T, GoodLongestGap == T) %>% 
  select(Sub.Num, hearing_status_participant, Response, `0`:`4000`) 

# convert to long format
ss_iChart_long_coda <- ss_iChart_coda %>% 
  gather(key = Time.ms, value = value, `0`:`4000`) %>% 
  filter(value %in% c("0", "0.5", "1")) %>% 
  mutate(value_cat = factor(value, labels = c("Distractor", "Signer", "Target")),
         Time.ms_numeric = to.n(Time.ms)) 

# dplyr::summarise
ms_iChart_count_coda <- ss_iChart_long_coda %>% 
  group_by(Sub.Num, Time.ms, hearing_status_participant, value_cat) %>% 
  dplyr::summarise(count = ifelse(n() == 0, 0, n())) %>% 
  dplyr::summarise(sum_count = sum(count))

ms_iChart_coda <- as.data.frame(xtabs(~ Sub.Num + value_cat + Time.ms,
                                      data = ss_iChart_long_coda)) %>% 
  left_join(y = ms_iChart_count_coda, by = c("Time.ms", "Sub.Num")) %>% 
  mutate(proportion_looking = Freq / sum_count,
         hearing_status_participant_factor = as.factor(hearing_status_participant))

# get means and CIs for proportion looking at each time slice across particpants
ms_mean_iChart_coda <- ms_iChart_coda %>% 
  group_by(Time.ms, hearing_status_participant, value_cat) %>% 
  dplyr::summarise(mean_prop_looking = mean(proportion_looking, na.rm = T),
                   ci_low = ci.low(proportion_looking),
                   ci_high = ci.high(proportion_looking))
```

Write participant-level graph values to CSV.

```{r}
write.csv(x = ms_mean_iChart_coda, 
          file = "../../data/processed_data/graph_values/sol_gvals_coda.csv", 
          row.names = F)
```

### Save trial-level data 

Mean accuracy and RT for each trial

```{r}
ss_trial_level_df <- iChart %>% 
  select(Sub.Num, Sex, Response, Tr.Num, clean_target_img, stimuli, age_code,
         age_peek_months = Months, hearing_status_participant, trial_types, 
          RT, Accuracy, signs_produced, include, GoodRT, reason_excluded) %>% 
  mutate(correct = ifelse(trial_types == "C_T", 1, 0))
```


```{r}
write.csv(x = ss_trial_level_df, 
          file = "../../data/processed_data/summary_tables/sol_trial_level_df.csv", 
          row.names = F)
```