Normal performance variability on a dichotic CV test across nine onset-time-asynchrony conditions: application of a binomial-distribution model.
The purpose of the present study was to generate normative data for a dichotic listening task involving the identification of dichotic consonant vowels at interchannel onset-time asynchronies of 120, 90, 60, 30, and 0 msec, under right- and left-lag conditions. The subjects were 30 healthy young adults, all of whom passed rigorous preliminary testing to ensure normal auditory function. Both group and individual data were analyzed. When the data were collapsed across all subjects, the results were similar to those of previous studies. However, the individual data, analyzed according to a model based on the binomial distribution, demonstrated that the collapsed group data were misleading since they did not predict normal variability. Both the group and the individual results suggested that the 90- and 0-msec conditions are the most useful clinically. Further research involving patients afflicted with lesions, in which the data are gathered and analyzed in a manner similar to that of the present study, was recommended.