1 Set up R environment

library(tidyverse)
library(ggplot2)
library(ggpubr)
library(plyr)
library(magick)
library(png)
library(EBImage)
library(lme4)
library(lmerTest)

2 Set the R working drectory to the main experiment directory.

setwd("/Users/adambarnas/Box/MetaAwareness/data/")

3 Read in the individual subject files.

tbl_all <- list.files(path = "./Rensink_Recontact", pattern = "*.csv", full.names = T, ignore.case = F) %>%
  map_df(~read.csv(., colClasses=c("gender..m.f."="character")))
tbl_all = subset(tbl_all, select = c(user_resp.keys,user_resp.rt,workerId,image_a))
col_idx <- grep("workerId", names(tbl_all))
tbl_all <- tbl_all[, c(col_idx, (1:ncol(tbl_all))[-col_idx])]
tbl_all <- data.frame(na.omit(tbl_all))
tbl_all <- tbl_all %>%
separate(image_a,into=c('database', 'image', NA), sep = "([\\_\\-])")

4 Compute average likelihood rating.

tbl_all_subj_avg <- tbl_all %>%
  group_by(workerId,image) %>%
  dplyr::summarize(average = mean(user_resp.keys)) %>%
  spread(image,average) %>% 
  mutate(subj_avg = rowMeans(.[-1], na.rm = TRUE))
mean(tbl_all_subj_avg$subj_avg)
## [1] 3.633333
tbl_all_img_avg <- data.frame(img_avg = colMeans(tbl_all_subj_avg[,2:49], na.rm = TRUE))
tbl_all_img_avg <- tibble::rownames_to_column(tbl_all_img_avg, "image")

5 Merge Mudsplash and Meta-Awareness data files.

rensink_RTs_raw <- read_csv("Rensink_RTs_raw.csv")
rensink_RTs_raw <- rensink_RTs_raw[, -c(2,3,6:17,19,20)]
rensink_RTs_raw <- rensink_RTs_raw[(rensink_RTs_raw$workerId %in% tbl_all_subj_avg$workerId),]

tbl_all <- tbl_all[order(tbl_all$workerId, tbl_all$image), , drop = FALSE]
rensink_RTs_raw <- rensink_RTs_raw[order(rensink_RTs_raw$workerId, rensink_RTs_raw$image), , drop = FALSE]

rensink_RTs_likelihood <- left_join(tbl_all, rensink_RTs_raw, by = c("workerId", "image"))
colnames(rensink_RTs_likelihood)[2] <- "likelihood_rating"
colnames(rensink_RTs_likelihood)[3] <- "likelihood_rt"
colnames(rensink_RTs_likelihood)[7] <- "detection_rt"
rensink_RTs_likelihood <- rensink_RTs_likelihood[, c(-4,-6)]
rensink_RTs_likelihood <- rensink_RTs_likelihood[,c(1,4,5,2,3)]

6 Compute likelihood rating for each image.

rensink_RTs_likelihood %>%
  ggbarplot(x = "image", y = "likelihood_rating", ylab = "Likelihood of Detecting Change", title = "All images (30 per subject)", fill = "#f7a800", add = "mean_se", font.xtickslab = 8, sort.val = c("asc")) + rotate_x_text() + theme(legend.position = "none")

rensink_RTs_likelihood_no_NA <- rensink_RTs_likelihood %>%
  drop_na()
rensink_RTs_likelihood_no_NA %>% 
  ggbarplot(x = "image", y = "likelihood_rating", ylab = "Likelihood of Detecting Change", title = "'Correct' images", fill = "#f7a800", add = "mean_se", font.xtickslab = 8, sort.val = c("asc")) + rotate_x_text() + theme(legend.position = "none")

7 Count number of ratings

rensink_RTs_likelihood_count <- rensink_RTs_likelihood_no_NA %>% 
  group_by(workerId,image) %>% 
  dplyr::summarize(counts = n()) %>%
  spread(image,counts) %>%
  mutate(sum = rowSums(.[-1], na.rm = TRUE))
#head(tbl_all_counts,10)

rensink_RTs_likelihood_count <- data.frame(count = colSums(rensink_RTs_likelihood_count[,2:49], na.rm = TRUE))
rensink_RTs_likelihood_count <- tibble::rownames_to_column(rensink_RTs_likelihood_count, "image")
rensink_RTs_likelihood_count
##          image count
## 1        Amish     9
## 2         Army     8
## 3        Barns     8
## 4    BarnTrack     8
## 5      Barrels     7
## 6        Beach    11
## 7        Birds     9
## 8         Boat     8
## 9          Bus     9
## 10      Cactus     9
## 11       Camel    10
## 12 CanalBridge     8
## 13      Castle     5
## 14     Chopper     9
## 15     Cockpit     8
## 16 Description     6
## 17      Dinner     8
## 18       Diver    11
## 19      Eating     8
## 20       Egypt     8
## 21  FarmByPond     5
## 22      Farmer     9
## 23     Fishing     9
## 24  Floatplane     8
## 25    Fountain    10
## 26      Harbor     9
## 27     Horizon     8
## 28         Ice     7
## 29       Kayak    10
## 30     Kayaker     8
## 31        Kids     8
## 32        Lake    11
## 33      Market     6
## 34     Marling     8
## 35      Mosque    10
## 36   NotreDame    12
## 37      Nurses     8
## 38     Obelisk     9
## 39  OtherDiver     6
## 40      Pilots     9
## 41        Seal    11
## 42    Soldiers     9
## 43     Station     9
## 44  SummerLake     9
## 45      Turtle     6
## 46       Water     9
## 47      Window     8
## 48        Wine     9

8 Mixed effects model and correlation.

fit0 <- lmer(detection_rt ~ likelihood_rating + (1 | workerId) + (1 | image), data=rensink_RTs_likelihood_no_NA)
summary(fit0)
## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: detection_rt ~ likelihood_rating + (1 | workerId) + (1 | image)
##    Data: rensink_RTs_likelihood_no_NA
## 
## REML criterion at convergence: 1839.1
## 
## Scaled residuals: 
##     Min      1Q  Median      3Q     Max 
## -2.1786 -0.4009 -0.1123  0.2428  4.5075 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev.
##  image    (Intercept) 0.6162   0.785   
##  workerId (Intercept) 4.4929   2.120   
##  Residual             4.3149   2.077   
## Number of obs: 407, groups:  image, 48; workerId, 16
## 
## Fixed effects:
##                   Estimate Std. Error       df t value Pr(>|t|)    
## (Intercept)         8.4320     0.6633  32.0856  12.712 4.54e-14 ***
## likelihood_rating  -0.0961     0.0999 251.4716  -0.962    0.337    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Correlation of Fixed Effects:
##             (Intr)
## liklhd_rtng -0.551
corr <- rensink_RTs_likelihood_no_NA %>% 
  group_by(image) %>% 
  dplyr::summarize(detection_rt = mean(detection_rt), likelihood_rating = mean(likelihood_rating))
corr %>%
  ggscatter(y = "detection_rt", x = "likelihood_rating", ylab = "Change Detection RT (sec)", xlab = "Likelihood of Detecting Change", title = "N = 16", fill = "#f7a800", color = "#f7a800", add = "reg.line", cor.coef = TRUE, cor.coeff.args = list(method = "pearson", label.x = 1, label.y = 5, label.sep = "\n"), xlim = c(1, 5), ylim = c(0, 15))