IQ is related to the worst performance rule in a memory task involving children

Abstract

Larson and Alderton [Intelligence 14 (1990) 309.] observed that on elementary cognitive tasks (ECTs), worst performance (i.e., slowest reaction time (RT)) was a better predictor of IQ than best performance (i.e., fastest RT). This suggested to them a Worst Performance rule, which states that “The worst RT trials reveal more about intelligence than do other portions of the RT distribution” (p. 310). The present study examined whether the Worst Performance rule could be applied to a strategic memory task in which children were asked to remember different lists of categorizable words (e.g., boat, bus, car, banana, apple, peach). The results revealed strong support for the Worst Performance rule, with worst performance (i.e., fewest words recalled) predicting more unique variance in IQ than any other variable (e.g., most words recalled; fewest strategies used). These results demonstrate that the Worst Performance rule can be used to predict IQ using performance measures other than RT and in tasks other than ECTs.