Theoretical Perspectives Relevant to Developmental Psychology

A discussion of cognitive changes associated with old age within the framework of the first three sections.

Old age is often viewed as a period of cognitive or intellectual decline. Older adults are commonly described as unable to learn new skill, rigid, suffering from a decline in memory, or "senile". However, this stereotypic view of cognitive change in old age is inaccurate. Cognitive and intellectual decline in old age in not universal (Schaie, 1974). In fact, cognitive changes associated with old age are better described as multidimensional and multidirectional, and characterized by large individual differences and plasticity. The myth of universal cognitive decline is ascribed to problems in definition, measurement, and methodology.

The three problems cited above are related to those issues discussed in the first three sections of this paper. The purpose of this section is to discuss cognitive changes associated with old age within the framework of the previous discussions. It should be emphasized, however, that this paper does not constitute a complete review of the literature concerning cognitive changes in later adulthood. Rather, cognitive change will be discussed only as it relates to the theoretical issues already described.

Cognitive Change as it Relates to Section One

This problem of definition in the study of cognitive change in old age is confounded as researchers and theorists propose definitions that reflect their paradigmatic and world views. For example, a structuralist would view cognitive change as unidirectional, whereas, a life-span developmental psychologist would view cognitive change as multidirectional. As developmental psychology as a field is currently undergoing a paradigmatic shift, differences exist in the way that cognitive change is studied and conceptualized. An analysis of the life-span developmental perspective of cognitive change will be used to highlight the effect of a paradigmatic view on theoretical conceptualization.

The life-span developmental perspective of cognitive change in later adulthood has been outlined by Schaie (1974) and Baltes and Schaie (1976, 1977). According to this view, cognitive change in old age is multidirectional, multidimensional, subject to large individual differences, and plasticity. Each of these will be discussed below.

According to the contextual world view, change is the result of the interaction between ontogenetic, evolutionary, and social change conditions. Therefore, cognitive change in old age is described as multidirectional (Baltes and Schaie, 1976,1977; Nesselroade & Baltes, 1974). This can be viewed two ways. First, change can occur in different directions for different cognitive dimensions. For example, crystallized intelligence is thought to increase with age (Horn,, 1970; Horn & Donaldson, 1976). Secondly, change can be multidirectional for a single dimension. The intervention work of Hartley, Harker, and Walsh (1980) and Fozard (1980) provide examples of such multidirectional change. Harley et al. (1980) found that performance on memory tasks could be improved following decline through the training of encoding strategies. Fozard (1980) also demonstrated improved memory performance through the use of task redesign and training. Thus, these interventions lead to an improvement following a decline in memory performance. Therefore, cognitive change can be multidirectional.

Cognitive change can also be described as multidimensional. Horn (1970) proposes that intelligence is made up of combination of genetically-based skills (fluid intelligence) and experientially-based skills (crystallized intelligence). Therefore, according to Horn, intelligence is a bidimensional concept. Schaie (1970) further breaks down Horn's (1970) categories to include verbal meaning, space, reasoning, and number (crystallized intelligence), and word fluency (fluid intelligence). In addition, Schaie (1970) examines psychomotor speed in relation to cognitive ability. Therefore, Schaie proposes a multidimensional view of cognitive ability.

The dimensions proposed above are all affected by ontogenetic, evolutionary, and environmental influences. Therefore, there are large individual differences in cognitive abilities and cognitive change. For example, Schaie (1973, cited in Schaie, 1974) noted that during a 14-year study of individuals within several age-cohorts, some individuals experienced decline, some experienced a gain, and some remained stable on cognitive dimensions. Therefore, cognitive change is not a fixed, universal phenomena tied to chronological age. Rather, cognitive abilities are tied to life-style, health, economics, etc. This theoretical position proposes that only through the study of the individual microenvironment will an understanding of individual differences in cognitive change be attained (Schaie (1974). Terminal decline (Jarvik & Falek, 1963; Riegel & Riegel, 1972) represents another example of individual differences in cognitive change as this phenomena occurs at different points across the chronological life-span for different individuals.

Baltes and Schaie (1976) propose that cognitive changes associated with old age are not fixed and irreversible. Rather, they support the notion of plasticity with respect to cognitive abilities in old age. Concomitant with this theoretical position is the promotion of intervention techniques. For example, Baltes and Labouvie (1973) advocate for the conduction of experimental, intervention research with respect to cognitive ability and old age. Subsequent research has supported the notion of plasticity with intervention leading to a gain in cognitive abilities performance (Birkhill & Schaie, 1975; Fozard, 1980; Hartley et al., 1980). Therefore, evidence is beginning to appear in the literature that supports the notion of cognitive plasticity in old age.

As described above, cognitive change is viewed by life-span developmental psychologists (contextual world view) to be multidirectional, multidimensional, subject to large individual differences, and plasticity. This is the result of the interaction of ontogenetic, socio-cultural, and evolutionary influence.

Cognitive Change as it Relates to Section Two

How one examines cognitive changes associated with old age is also influenced by how one defines such terms as development and aging. The life-span developmental perspective views cognitive changes associated with old age as part of normal development. Aging is viewed not as a disease state but rather as a normal developmental process. Thus, even changes, such as in fluid intelligence, that are associated with decline are viewed as normal as opposed to as a disease process.

In addition, the life-span developmental perspective is concerned with the description, explanation, and optimization of behavior. This theoretical perspective can be applied to the study of cognitive change. First, the life-span developmental perspective of cognitive change is described by Schaie (1974) and Baltes and Schaie (1976, 1977). As stated above, this perspective views cognitive change in old age as multidirectional, multidimensional, subject to large individual differences, and plasticity. Cognitive change is explained as the result of ontogenetic, evolutionary, and socio-cultural influences. Optimization is underscored by the emphasis on intervention strategies with the life-span developmental perspective (Baltes & Labouvie, 1973). Much of the research on cognitive change currently being undertaken is from this life-span perspective.

Cognitive Change as it Relates to Section Three

The study of cognitive change in old age is confounded by the different methodological and research strategies employed by researchers within developmental psychology. Methodologies commonly used include the cross-sectional method, longitudinal method, and sequential methods. However, the results of these methods are not always in agreement. This is in part due to the differences in the three underlying dimensions being studied: age change, age difference, and cultural change (Schaie, 1970). Each of these will be discussed below.

Much of the research on intellectual development and age differences has used a cross-sectional methodology. However, cross-sectional designs are confounded by cohort differences. Four major cohort differences with respect to performance on cognitive tasks have been uncovered. First, older adults tend to be more cautious than younger adults (Okun, 1976). This cautiousness tends to slow them down, thus, leading to a lower overall performance than younger adults. In addition, this cautiousness leads them to make errors of omission as opposed to commission (Wallach & Kogan, 1961). In other words, if an older adult is unsure of the answer they tend to leave it blank as opposed to guessing. This also leads to overall lower performance. Second, the majority of standardized tests have been standardized on a younger cohort. It has been argued that these tests are cohort specific. Thus, the younger group will tend to do better because that is the population for which the instruments were standardized (Schaie, 1974). Third, there are major educational differences between cohorts with increased educational levels with each successive cohort. This can be contrasted with the increased performance on tasks of intellectual ability with each successive generation. Thus, differences may be due to educational level as opposed to cognitive ability (Schaie, 1974). Fourth, there is a slowing of the peripheral and central nervous systems beyond the age of 55. Thus, where cognitive tests require a speed component, older subjects tend to perform poorly in comparison to younger subjects (Botwinick, 1977). These nervous system changes are impacted by a variety of cultural factors related to exercise and diet both heavily influenced by cohort-related factors. Thus, cohort differences are confounded with age differences.

All four of the cohort differences describe above underscore the dilemma of performance versus capacity/ability. Differences in measures of cognitive ability may be due to performance differences rather than cognitive capacity or ability. This is also a problem of design. However, it is not specific to any one design. Rather, it highlights the fact that cognitive ability can only be inferred and not directly measured (Botwinick, 1977).

Cross-sectional studies are also confounded with the phenomena of terminal drop (Riegel & Riegel, 1972). A decline in cognitive ability is often reported in cross-sectional studies. However, as stated previously in section three, as there is an increased incidence of mortality with increased age, the overall decrease in intellectual performance as a group may be simply the result of sampling bias. When only survivors are examined, intellectual performance remains unchanged (Riegel & Riegel, 1972).

Thus, much of the age differences reported in cognitive ability may be due to cohort differences and differential mortality/terminal drop.

Longitudinal studies of cognitive ability and age change have also been conducted. However, these studies are confounded by selection bias, selective subject attrition, retest familiarization, and history effects. For example, subjects who volunteer for longitudinal studies tend to perform better on intellectual tasks than those who do not volunteer (Baltes, 1968). Also, subjects who remain in the study or survive are qualitatively different than those who do not survive. Those who remain in the study tend to perform better on cognitive tasks than those who drop out (Jarvik & Falek, 1963; Riegel & Riegel, 1972). Due to these biases, longitudinal studies typically report a gain in intellectual performance with age.

In summary, cross-sectional studies have reported a drop in intellectual performance with age, while longitudinal studies have reported a gain in intellectual performance with age (Schaie, 1970). This conflict however is due to differences in methodology. For example, terminal decline manifests itself in cross-sectional studies as a decline in intelligence with age. However, in longitudinal studies, these subjects drop out, thus a gain in intellectual performance with age is manifested as an artifact of subject mortality.

In part to resolve this conflict, sequential strategies have been employed to study cognitive changes associated with old age (Schaie, 1970, 1974). These strategies have demonstrated that most cognitive abilities experience a gain or remain stable in later adulthood. Only psychomotor speed and word fluency (fluid intelligence) exhibit a decrement with increased age. This decrement however is much less than that reported in cross-sectional studies and plasticity is demonstrated through remediation. Therefore, the theoretical position of universal decline in old age is proposed by Schaie (1974) and Baltes and Schaie (1976, 1977) to be a methodological artifact.

Summary

It is generally accepted that cognitive abilities change with age. However, what change occurs and what is the direction of the change is not so well accepted. This is in part due to paradigmatic differences with the field of developmental psychology, and in part due to methodological differences in the study of cognitive change. Until these basic issues are resolve, consensus concerning cognitive change in old age will not be achieved.

Shaie has proposed the following conclusions based on cross-sequential studies of intellectual ability and age:

  1. Reliable decrements can not be found for all abilities for all persons. Decline is not likely at all until very late in life.

  2. Decline is most evident where speed of response and peripheral nervous system (PNS) functioning are involved.

  3. Declines will be evident for most abilities for individuals of any age who have severe central nervous system (CNS) disease (e.g. Alzheimer's Disease) and for those in their 50s and 60s who live in deprived environments.

  4. Data on independent samples (cross-sectional studies) will over estimate loss for abilities (no benefit of practice). Dependent samples (longitudinal studies) will better measure age changes for those in good health and stimulating environments and underestimate loss for those in poor health and impoverished environments.

  5. Cohort effects (e.g. generational differences in the amount of education) account for more variance in intelligence than do ontogenetic (age-related, maturational) biologically based factors.

  6. Individual differences in what skills decline as well as the extent of the declines are substantial.

Based upon the above, it is clear that cross-cultural, cross-sequential, multifactorial research needs to be conducted related to intellectual and cognitive functioning with age. It is imperative that this research needs to include the impact of culture, context, and socioeconomic factors on these abilities and attention needs to be paid to careful selection of large, stratified random samples. Generalizations from the volumes of cross-sectional research on aging and intelligence, cognition, learning, and memory based on limited samples needs to be examined critically and the results viewed with skepticism. This is particularly important as inaccurate conclusions based in methodological artifacts have a way of becoming reified and often result in self-fulfilling prophecies.



1998 copyright Linda M. Woolf


Back to the Introduction.



Back to Woolf Gerontology Page

Back to Woolf Home Page