Moreover, a series of training studies, conducted in Germany (Baltes et al., 1984; Baltes et al., 1992) and the U.S. (Schaie, 1986; Schaie and Willis, 1986; Willis and Schaie, 1994), have shown that older individuals still have a great deal of potential plasticity, or reserve capacity for development. The results demonstrated that intervention can lead to significant gains in skills such as problem-solving tasks (Denney, 1979), perceptual speed (Hoyer, Labouvie, and Baltes, 1973), and fluid cognition (Baltes and Lindenberger, 1988; Willis, 1987). Intervention research generally targeted those skills which have been shown to decline the most (i.e., fluid cognition and processes representative of the mechanisms of cognition).

In general, results from intervention studies convincingly demonstrated the remarkable plasticity of human cognition in the elderly (see Willis, 1987 for a review). In the German studies, better performance was demonstrated for (1) target training (Baltes and Willis, 1982; Willis, Blieszner, and Baltes, 1981), (2) independent self-practice (Baltes et al., 1989; Hayslip, 1989a, 1989b), and (3) removed time constraints (Hofland, Willis, and Baltes, 1981). Willis and Schaie, 1986; Schaie and Willis, 1986; Willis and Schaie, 1994) obtained similar findings within a longitudinal design.

These results were replicated in a second follow-up study conducted in 1991 with both new participants and participants from the original training study. Specifically, results from the Seattle Training Study, a component of the Seattle Longitudinal Study (Schaie, 1996) indicated that the performance of the elderly can be successfully impacted in such a way that older adults' performance is boosted back to the level at which they performed more than a decade before. The Seattle researchers set up five one-hour sessions aimed at training the elderly adults' spatial and reasoning skills. The training had differential impact on certain subgroups of the elderly population. For those who had shown decline on either of the Primary Mental Skill Test subtests over the preceding fourteen-year period, training was effective in returning their performance nearly to the original level. For those who had remained stable over the preceding fourteen-year period, training raised their performance beyond the level they performed at fourteen years prior to the training. In addition, the training has been found to be effective, not only in the short run, but over seven years (Neely and Backman, 1993; Willis and Nesselroade, 1990).

One of the outcomes of these studies is the realization that longer and more structured training seems to be necessary for remediation in the very old (Schaie, 1994; Willis, 1989). The importance of these studies is that they suggest that cognitive decline in many individuals may be due to disuse of certain cognitive skills, and that remediation is possible for a significant number of participants, especially for the young-old (Schaie, 1994; Willis, 1990; Willis and Schaie, 1994).

The developmental trajectory of everyday cognition has been examined by a number of researchers (see Berg, in press; Berg and Klaczynski, 1996, for review). The summary of the field today is that the pattern of age differences in everyday cognition differs dramatically depending on how problems to be solved are defined and what criteria are used for optimal problem solving. For example, Berg, Klaczynski, Calderone, and Strough (1994), studying participants' own ratings of how effective they were in solving their own everyday problems, did not find any age differences. Denny and her colleagues (Denney and Palmer, 1981; Denney and Pearce, 1989) utilized the number of "safe and effective solutions" as the criterion for optimal problem solving and found that the highest number of such solutions was generated by middle-aged adults, with both younger and older adults offering fewer solutions. Cornelius and Caspi (1987), using the closeness between participants' ratings of strategy effectiveness and a "prototype" of the optimal everyday problem solver as the criteria, found an increase in everyday problem-solving skill with adult age.