Not only do the declines in cognitive functioning appear later in life than we might have expected, but cognitive abilities vary in how they are affected by aging. The Seat- tle Longitudinal Study (see Chapter 15) found that many abilities start a dramatic downward trend beginning in the twenties, with an even greater one as advanced age is reached. There are important exceptions. The cross-sectional data for verbal and numeric abilities indicate a peak in midlife with relatively little change into early old age but a significant decline in the eighties. Take a minute to review some weaknesses of a cross-sectional design. The major one, of course, is that you don't know if your groups are comparable. For instance, would your 80-year-olds, if measured at the age of 40, be similar to your current 40-year-olds? In the Seattle study, both longitudinal and cross-sectional data were collected and analyzed. The only ability that shows profound linear decre ment is perceptual speed. This decline is the result of progressive slowing of neural impulses throughout the central nervous system. Another important finding is the extent of individual variation; that is, some very old people are capable of quite quick responses (Powell & Whitla, 1994; Schaie, 1995). Even more interesting is the fact that most other abilities show a gain from young adulthood into midlife. Intellectual competence generally peaks in the forties and fifties, because people continue to gain experience and without significant physiological loss to offset the gain. Speed of numeric computation declines significantly with age when followed longitudi but in ability does not peak until the sixties and declines only modestly after that age.