Interrelations of Aging, Knowledge, and Cognitive Performance

Abstract

This chapter reviews the research literature concerned with the relation between age and knowledge and with the role of knowledge on the relations between age and cognitive performance. Although it is generally assumed that accumulation of experience with age leads to greater quantity or quality of knowledge, the empirical evidence from large-scale studies with representative samples of research participants suggests that while there appears to be an increase in knowledge from age 18 to about age 40 or 50, the dominant trend in later years of adulthood is one of either stability or decline. Among the hypotheses discussed to account for the lack of continuous growth in knowledge are generational confounds in education, losses offsetting gains, an asymptote on exposure to new information, and increased specialization of one’s knowledge. Each hypothesis has some plausibility, but it is concluded that the reasons for the failure to find continuous age-related increases in knowledge are still not well understood. Several conceptual models of the role of knowledge on the relations between age and cognition are discussed, including moderation, mediation, and migration. Because interactions of age and knowledge have been inconsistent, and because statistical control of knowledge tends to increase rather than decrease the negative relations between age and measures of cognitive performance, the available empirical evidence seems to favor the migration interpretation. That is, age-related effects on cognition appear to be reduced among people with high levels of knowledge because people tend to “migrate” into higher knowledge groups with increasing age.

The realization that different cognitive variables have different age trends dates back to the earliest empirical studies on aging. For example, several studies published in the 1920s found larger age-related differences for measures of memory and reasoning than for measures of vocabulary and general information (e.g., Foster & Taylor, 1920; Hollingworth, 1927; Willoughby, 1927). Many labels have been used to characterize the two types of cognition (such as new learning versus reliance on stored information, educative versus reproductive abilities, fluid versus crystallized abilities, and mechanics versus pragmatics), but the terms process and product may be the most descriptive because the first term refers to the efficiency of processing at the time of assessment, and the second term refers to the cumulative products of past processing (see also P. Baltes, this volume; Wellman, this volume).

The different age trends can be illustrated with data collected by Jones and Conrad (1933) on three subtests from the Army Alpha test battery (see Figure 12.1). (The vertical axis in most of the figures in this chapter represents performance in standard score or z-score units to facilitate comparison across variables and studies. Sometimes a sample of young adults will be used as the reference distribution as in this figure, and other times the entire sample will be used as the reference distribution. In either case, the units along the vertical axis are standard deviations of the relevant distribution, and thus the scale is informative about the magnitude of the age difference relative to the distribution of scores in the reference sample.)

Notice in that across the range from 20 to 55 years of age the measures of general information and vocabulary tend to remain stable but that increased age is associated with lower scores on the analogies measure. This same general pattern has been reported many times, and it is now widely accepted that different types of cognitive measures can have quite different age trends.

There are two major implications of these findings. The first implication is that because the constituent variables exhibit different age relations, no single number will be meaningful as an index of overall cognitive ability across all of adulthood. That is, if cognitive ability is composed of both product and process aspects, then it is not meaningful to refer to a single age trend in cognitive functioning. Jones and Conrad (1933) expressed this point elegantly by stating that the older adult derives more intellectual power from accumulated stocks of information than do young adults.

The second major implication of the different age trends is that there are at least two quite distinct phenomena to explain in the field of cognitive aging—the age-related decline in process aspects of cognition and either stability or increase with age in product aspects of cognition. The majority of cognitive aging research has focused on process aspects of cognition, in part because of an interest in remediating or preventing age-related cognitive decline. However, in this chapter, I focus on two issues concerned with aging and product or knowledge aspects of cognition. The first issue concerns the relation between age and measures of the quantity of knowledge, which might be expected to be positive because knowledge presumably derives from experience and experience is often assumed to increase continuously with advancing age. The second issue is the role played by knowledge on the relations between age and cognitive performance and particularly the ability of greater knowledge to offset the consequences of age-related decline in process aspects of cognition. This issue is particularly relevant to the P. Bakes and M. Baltes (1990) selective optimization with compensation (SOC) framework because of the possibility that optimizing one’s knowledge in select domains might compensate for declines that may be occurring in other aspects of cognitive functioning (see also M. Baltes & Carstensen, this volume; P. Baltes, this volume; cf. Lerner, Dowling, & Roth, this volume).