1.Research Question Does reward magnitude influence reaction time and decision accuracy?
#Primary Analysis: Independent T-Test
#(Reaction Time)
df <- read.csv("/Users/sake/Downloads/ddm_all_data_experiment_1.csv")
colnames(df)
## [1] "participant" "staircase_threshold" "dem_age"
## [4] "dem_gen" "bias_source" "bias_direction"
## [7] "trial_number" "reference_length" "target_length"
## [10] "line_type" "answer" "rt"
## [13] "accuracy"
t.test(rt ~ bias_source, data = df,
alternative = "two.sided")
##
## Welch Two Sample t-test
##
## data: rt by bias_source
## t = -0.92795, df = 29997, p-value = 0.3534
## alternative hypothesis: true difference in means between group mullerlyer and group payoff is not equal to 0
## 95 percent confidence interval:
## -0.04138641 0.01479042
## sample estimates:
## mean in group mullerlyer mean in group payoff
## 1.067494 1.080792
#Visualization
ggplot(df, aes(x = bias_source, y = rt,
fill = bias_source)) +
geom_boxplot() +
labs(title = "Reaction Time by Reward
Level", y = "Reaction Time (ms)")
3. Secondary Analysis: Chi-Square Test To ensure the Decision Accuracy
variable is utilized, we test if accuracy is independent of reward
level.
# Create a contingency table
table_data <- table(df$bias_source,
df$accuracy)
# Run Chi-Square
chisq.test(table_data)
##
## Pearson's Chi-squared test with Yates' continuity correction
##
## data: table_data
## X-squared = 33.629, df = 1, p-value = 6.67e-09
# Visualization
ggplot(df, aes(x = bias_source, fill =
accuracy)) +
geom_bar(position = "fill") +
labs(title = "Decision Accuracy
Proportion by Reward Level", y =
"Proportion")