Gender effects on reaction time variability and trial-to-trial performance: reply to Deary and Der (2005)

Deary and Der (2005) reported significant gender effects on intraindividual variability in choice reaction time (CRT), and suggested several potential causes. In their procedure it was not possible to examine RTs on a trial-to-trial basis. We therefore investigate an additional possible cause: that men and women differ in their rate of performance improvement across trials, creating an apparent difference in variability. Using data from a large online CRT study, we replicate the finding that women are more variable in RT than men. However, we also demonstrate an interaction between gender and trial number for RT: women were initially slower than men, but became faster than men across a block. When the first two trials of the block are excluded, the gender effects on RT variability disappear.