Picture-Word-N1/02 Stimuli/02-tabulate_stimuli.Rmd

---
title: "Stimuli List"
author: "Jack Taylor"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library(tidyverse)
ggplot2::theme_set(theme_minimal())

practice_stim_long <- read_csv("practice_stim.csv")
stim_tidy <- read_csv("stim_tidy.csv")
```

# Practice Trials List

Column name explanations:

* `item_nr`: An identifier for the matched pair of words

* `image`: The image to be displayed in the study (can right-click, open image in new tab to see larger)

* `string`: The practice trial's word

* `condition`: The condition for the practice trial (A1=congruent, A2=incongruent)

* `Zipf`: Word frequency in Zipf (SUBTLEX-UK)

* `Length`: Word length

* `CNC`: Mean concreteness rating

* `BG`: Mean bigram probability

&nbsp;&nbsp;

```{r, echo=FALSE}
practice_stim_long %>%
  mutate(image = sprintf("![](boss_small/%s.jpg)", filename)) %>%
  mutate(item_nr = sprintf("Practice %i", item_nr)) %>%
  mutate_at(vars(dplyr::matches("Zipf.SUBTLEX_UK")), ~ round(., 2)) %>%
  mutate_at(vars(dplyr::matches("cos_ppmi_sim")), ~ round(as.numeric(.), 5)) %>%
  mutate_at(vars(dplyr::matches("BG.SUBTLEX_UK")), ~ round(., 5)) %>%
  rename(
    Zipf = Zipf.SUBTLEX_UK,
    sim = cos_ppmi_sim,
    CNC = CNC.Brysbaert,
    BG = BG.SUBTLEX_UK,
    Length = Length
  ) %>%
  select(item_nr, image, string, condition, Zipf, Length, CNC, BG) %>%
  knitr::kable()
```

# Experimental Trials List

Column name explanations:

* `item_nr`: An identifier for the matched pair of words

* `image`: The image to be displayed in the study (can right-click, open image in new tab to see larger)

* `perc_agree`: The percentage of name agreement for the congruent word

* `string_A1`: The congruent word

* `string_A2`: The incongruent word

* `lev_dist`: The levenshtein distance between the congruent and incongruent words

* `semantic_sim`: The ppmi associative semantic similarity between the congruent and incongruent words

* `Zipf_A1` & `Zipf_A2`: Word frequency in Zipf (SUBTLEX-UK) for each word

* `Length_A1` & `Length_A2`: Word lengths

* `CNC_A1` & `CNC_A2`: Mean concreteness ratings

* `BG_A1` & `BG_A2`: Mean bigram probabilities

* `OLD20_A1` & `OLD20_A2`: OLD20 values (orthographic neighbourhood density)

&nbsp;&nbsp;

```{r, echo=FALSE}
stim_tidy %>%
  mutate(image = sprintf("![](boss_small/%s.jpg)", filename_A1)) %>%
  select(item_nr, string_A1, image, nb_diff_names_A1, everything(), -filename_A1, -nb_diff_names_A2) %>%
  mutate_at(vars(dplyr::matches("Zipf.SUBTLEX_UK")), ~ round(., 2)) %>%
  mutate_at(vars(dplyr::matches("cos_ppmi_sim")), ~ round(as.numeric(.), 5)) %>%
  mutate_at(vars(dplyr::matches("BG.SUBTLEX_UK")), ~ round(., 5)) %>%
  rename(
    nb_diff_names = nb_diff_names_A1,
    perc_agree = perc_name_agree_A1,
    Zipf_A1 = Zipf.SUBTLEX_UK_A1,
    Zipf_A2 = Zipf.SUBTLEX_UK_A2,
    sim_A1 = cos_ppmi_sim_A1,
    sim_A2 = cos_ppmi_sim_A2,
    CNC_A1 = CNC.Brysbaert_A1,
    CNC_A2 = CNC.Brysbaert_A2,
    BG_A1 = BG.SUBTLEX_UK_A1,
    BG_A2 = BG.SUBTLEX_UK_A2,
    OLD20_A1 = ON.OLD20_A1,
    OLD20_A2 = ON.OLD20_A2,
    lev_dist = levenshtein_distance
  ) %>%
  mutate(
    semantic_sim = c(sim_A1, sim_A2)[c(sim_A1, sim_A2)!=1]
  ) %>%
  arrange(perc_agree) %>%
  select(item_nr, image, perc_agree, string_A1, string_A2, lev_dist, semantic_sim, Zipf_A1, Zipf_A2, Length_A1, Length_A2, CNC_A1, CNC_A2, BG_A1, BG_A2, OLD20_A1, OLD20_A2) %>%
  knitr::kable()
```

## Predictability Distribution

```{r, echo=FALSE}
pred_hist <- stim_tidy %>%
  ggplot(aes(perc_name_agree_A1)) +
  geom_bar(width = 1, fill="grey25") +
  scale_x_binned(
    expand = c(0,0),
    n.breaks = 10,
    limits=c(0,100)
  ) +
  scale_y_continuous(expand = c(0,0)) +
  labs(x = "Predictability (%)", y = "N Images") +
  theme_classic() +
  theme(plot.margin = margin(5, 15, 5, 5, unit="pt"))

ggsave(file.path("fig", "pred_hist.svg"), pred_hist, width=6, height=1)
```