Abstract
Ten young women (age 20 to 22 years) and 10 middle-aged women (age 36 to 44 years) served as subjects in choice reaction time, letter classification, and abstract matching-to-sample tasks. In each of seven conditions, the older group responded more slowly than the younger group. Age differences showed a complexity effect. That is, differences between the latencies of young and old subjects increased as the latency of the young subjects increased. Both linear and power functions accurately described the relation between the latencies of the middle-aged and young adult groups. This was true not only for the relation between average latencies but also for the relation between corresponding quartiles of latency distributions. Similar results were observed at the individual level: All middle-aged subjects showed complexity effects, and, for each middle-aged subject, the relation between her latencies and those of the average young adult was well described by linear and power functions. These findings indicate that age-related slowing is apparent by age 40, and that complexity effects are observable in individual performances. This slowing is global and not specific to particular tasks, as indicated by the fact that the latencies of older adults can be predicted directly from those of younger adults without regard to the nature of the task.