A systematic replication of the worst performance rule

The worst performance rule (WPR) states that when reaction times are ranked from fastest to slowest and then correlated with intelligence test scores, the slowest reaction times predict participants' intelligence better than the mean or best reaction times. Previous research suggests that the size of the WPR effect may be moderated by the difficulty or g-loading of RT tasks. Because there is substantial variation in sample size, sample characteristics and trial numbers across the different studies, we aim to replicate the WPR in a large, heterogenous sample using a variety of RT task. Our main aim is to replicate that the size of the WPR effect varies systematically with the difficulty or g-loading of tasks.