Background: I did an experiment in which 30 people watched 3 movie clips ("A", "B", and "C") in a random order. For each song and participant I measured one continuous neural variable named SNR (1 value for each participant and for each song) and one continuous behavioral variable named RT (1 value for each participant and for each song). I am interested in knowing whether there is a linear relationship between SNR and RT for each movie. Also, I am interested in knowing the strength of this relationship to compare across movies. In another words, I want to know whether the greater the SNR the greater the RT for each movie clip, and whether this is enhanced in some movies compared to others.
Here is some dummy data that follows the same structure that I have:
res_table <- tibble(
ID = rep(seq(1,30,1),3), # participant identifier
MOVIES = rep(c("A","B","C"),each=30),
SNR = sample(35:65, 30*3, replace = TRUE),
RT = sample(50:100, 30*3, replace = TRUE)
)
res_table$ID <- factor(res_table$ID, labels = c("1","2","3","4","5","6","7","8","9","10","11","12","13","14","15","16","17","18","19","20","21","22","23","24","25","26","27","28","29","30"))
res_table$MOVIES <- factor(res_table$MOVIES, levels = c("A","B","C"))
Problem: Although I initially performed Pearson Correlations between SNR and RT for each movie, someone told me that to explore these relationships it would be best to "show a scatter plot of the association between SNR and RT, including the linear model and the confidence range for the regression".
I fitted a mixed model as follows:
model.01 <- lmer(SNR~RT*MOVIES + (1|ID), data = res_table)
But from here on I am lost on what I should do. How can I know whether the relationship between SNR and RT at each movie level is significant or not?