Sustained attention in adulthood: A unique, large-sample, longitudinal and multicohort analysis using the Mackworth Clock-Test.

A cross-sectional sample of men (n = 177) tested in 1962-1964, and men (n = 188) and women (n = 113) tested in 1980-1984, and a longitudinal sample (n = 53) of men tested in 1962-1964 and 1980-1984, were used to investigate age differences and 18-year age changes on a 62-min sensory vigilance task that made virtually no demand on memory, the Mackworth Clock-Test. Age differences and changes in the vigilance decrement were also examined. No age differences or changes were obtained for detection accuracy. Target response time showed significant age-related 18-year longitudinal changes and one significant cross-sectional age difference; the overall age effect was of a U-shaped function with a minimum at middle age. Age differences and changes were obtained in a measure of arousal: skin potential response latency; the overall age effect was also consistent with a U-shaped function. There was little consistent evidence of a differential effect of age on the vigilance decrement. Outcomes contradictory to Surwillo and Quilter (1964) and Quilter, Giambra, and Benson (1983) are reconciled through additional statistical analyses.