In Experiment I. the effects of the range of the distribution of preparatory intervals (PIs) on the effect of PIon RT was demonstrated, and disparate conclusions about the development of preparatory set were thereby accounted for to some degree. In Experiment II, both range of the PIs and their average value were examined in a factorial design for their effects on RT, and it was concluded that, for any average value, greater range will result in greater PI effects; and for any range, greater average values will result in smaller PI effects on RT.In a recent study (Elliott, 1970) of age differences in simple reaction time (RT), it was shown that when a number of different preparatory intervals (PIs-the time from a warning signal to a reaction signal) are presented randomly in a series of trials, the effect of Plan RT was much greater among young children than it was among adults. The form of the RT by PI function, called the preparatory set effect, was familiar as the one usually resulting from random presen tations of members of a set of PIs of the values used (1, 2, 4,8, and 16 sec): The shortest PI resulted in the longest RTs, the median PI produced the quickest RTs, and the longest PI resulted in long RTs, so that the function produced was U shaped.The form of the function is usually accounted for as follows. For the shortest PI of a set like the one used, it is especially difficult to be ready to respond for two reasons: The contingent probability of its occurrence, given that a reaction signal has not yet occurred, is the lowest among all the PIs; further, since the shortest PI is usually preceded by a longer PI, adaptation level effects work especially strongly against one's being ready for it.The deficit that appears with the longest PI only appears when that PI is both absolutely and relatively so long that a S finds it difficult to maintain a readiness to respond: If it is not much longer than the other members in the set, RTs may be very quick to the longest PI, but there seems to be no circumstance in which they will be quick to the shortest one, given a random presentation of the PIs.In the author's 1970 paper, the young children produced steeper (larger) preparatory set effects than the adults, and this fact was interpreted to be in keeping with data showing that such effects were also greater in other groups that are slower than normal young adults: e.g., the aged, the retarded, and the psychotic. Persons in these groups presumably neither get ready to react as quickly, nor maintain a readiness to react as long. The problem is that the present author had earlier reported "These studies were conducted with support from Grants HD04047. HDOl571. and HDl2648 from the National Institute of Child Health and Human Development. The assistance of Sara Black and Richard Lyczak is very gratefully acknowledged.(1964) that young children showed preparatory set effects that were slightly smaller than those produced by adults. He had argued at the time that the children might have been so inattentive as to be imp...