The embedded set procedure. Comment on Galbraith, MacCrimmon, and Steffy.

The reaction-time crossover phenomenon discovered by Rodnick and Shakow has come to be studied mainly with the embedded isotemporal sets procedure and indexed by the difference in reaction time to regular and irregular trials at a 7-second preparatory interval (PI). A 1983 report in this journal by Galbraith et al. (J Nerv Ment Dis 171:670-675) demonstrated that the magnitude of crossover at this PI was influenced by the duration of the preparatory interval immediately prior (PPI) to the first trial of the 7-second isotemporal blocks. Secondary analyses reported here extend the examination of PPI influences and indicate: a) that the crossover phenomenon as assessed in the embedded sets procedure is due mainly to the very slow reaction time to the first trial of the 1-second isotemporal blocks; and b) that crossover in this paradigm is partially due to uncontrolled PPI influences across the 1-, 3-, and 7-second embedded blocks. These are further reasons for returning to the original Rodnick-Shakow method as Galbraith et al. recommended.