Skip Ribbon Commands
Skip to main content
You are editing a CASE STUDY Content Type

The Stability Of Intelligence with Age

Date September 2011
Author Loh Xu Ying
Topic Leadership and Management

The process of growing older is often associated with a gradual decline in intellectual ability, quickness of thinking and memory capacity. For organisations that use cognitive ability tests to assess individuals of different age groups, this raises questions about the applicability of uniform standards of test performance. However, empirical research has shown some unexpected findings with regard to the process and onset of cognitive aging. This article highlights key findings from the research literature and provides recommendations on how to handle cognitive ability testing across different age groups.



The process of getting older is commonly associated with a decline in intellectual capability. Older adults are often perceived to be less nimble at picking up new skills, slower to numerate and have declining memory. However, empirical research on this topic has highlighted how the onset of cognitive decline actually occurs much later than commonly believed. In this article, we explore the process of cognitive aging and highlight some recommendations for organisations that conduct cognitive ability testing for different age groups.


Early views

In the 1900s, the development of cognitive ability tests has enabled the standardised assessment of intelligence on a broader scale. During that period, many large-scale general population studies showed that the trend of cognitive ability peaks around the ages of 20 to 30, followed by a steady decline. The figure below shows a typical chart depicting the apparent decline in cognitive ability tests scores with increasing age, using data on the Standard Progressive Matrices (SPM) obtained from a series of studies conducted between 1939 and 1947 in the United Kingdom (Raven, 2000).

Figure 1. Chart of SPM scores by age from studies between 1939 and 1947 (Raven, 2000).


However, some have called into question the interpretation suggested by these studies, highlighting inadequacies in the study design that failed to detangle the impact of age from other confounding factors that influence intelligence (Schaie, 1983).

Specifically, many of the early studies made use of cross-sectional assessments in which individuals from varying age groups were tested within a single time frame. Thus, the results not only reflected the performance of individuals of different ages but also individuals of different generations. These observed trends could have arisen due to the effects of age or due to generational differences in factors that influence cognitive capacity, such as years and quality of education and exposure to environmental complexity.

Insights from longitudinal studies

Since then, more rigorous longitudinal studies on the issue have led to new insights on the relationship between cognitive ability and age. A particularly significant study was the Seattle Longitudinal Study (Schaie, 1994; Schaie & Hertzog, 1986), which tracked the long-term cognitive performance of individuals in the United States from 1956. Over a period of about 40 years, more than 3,900 adults from age 22 to 84 were tested in seven major testing cycles on a range of cognitive domains.1

The results challenged some earlier views and provided a deeper understanding of how cognitive ability changes with age. As illustrated in the following table charting the relationship between cognitive performance and age, different abilities show different patterns of change over time. Perceptual speed declines across adulthood and numeric ability starts to show moderate decline at age 50. However, it turns out that other abilities such as spatial orientation, verbal ability and inductive reasoning show a slight increase in early adulthood before reaching a plateau that is sustained till age 60. Thus, contrary to earlier studies that show an early onset of cognitive decline, these abilities are relatively stable across adulthood, with declines in performance only evident after 60 (Schaie, 1994).

Figure 2. Chart of longitudinal mean cognitive ability scores by age, from Schaie (1994).


These findings have been replicated in a range of other studies (McDonald-Miszczak, Hertzog & Hultsch, 1995; Zelinski & Burnight, 1997; Baltes, Staudinger, & Lindenberger, 1999), therefore suggesting that generational differences or cohort effects had inflated the cognitive ability differences found in earlier studies on cognitive aging. In fact, studies using more recently accumulated longitudinal data have found compelling evidence for the inflationary influence of cohort effects on the cognitive aging process (Zelinski & Kennison, 2007; Zelinski, Kennison, Watts & Lewis, 2009).


The research described above have primarily studied how fluid intelligence changes with age. Fluid intelligence refers to the ability to reason logically, see complex relationships and solve novel problems, independent of acquired knowledge (Horn, 1967). Fluid intelligence is often contrasted with crystallised intelligence, which refers to the variety of skills and knowledge one acquires from learning and acculturation (Horn & Cattell, 1967) As crystallised intelligence involves knowledge acquisition and learning from experiences, studies on the relationship between crystallised intelligence and age show markedly different trends.

In a noteworthy study by Ackerman & Rolfhus (1999), a wide range of ability tests (fluid intelligence) and knowledge scales (crystallised intelligence) were administered to young adults age 18 to 27, and middle-aged adults age 30 and above. While middle-aged adults performed poorer than young adults on numerical and spatial ability tests, they performed better on the verbal ability tests and vastly out-performed young adults on all of the knowledge tests (spanning the Humanities, Sciences, and Civics), except Chemistry. These results show that when it comes to acquired knowledge, middle-aged adults who have more years of experience on their side outperform young adults by a long stretch.


On the whole, research suggests that while reaction time and information processing speed display a gradual decline over adulthood, there is relative maintenance of reasoning ability across adulthood, with declines only noticeable around age 50 to 60. Moreover, intellectual capabilities across adulthood, particularly crystallised intelligence, arguably increase as individuals accumulate far greater knowledge across a wide range of domains.

In the recruitment context, this distinction between fluid and crystallised intelligence highlights the need to consider both 'raw' intellectual ability and acquired skills and knowledge when assessing the intellectual capacity of applicants. While the assessment of acquired job-relevant skills may be difficult for fresh graduates, it becomes increasingly important for mid-career applicants who have the years of experience behind them. A key consideration in making selection decisions is the extent to which these mid-career applicants have acquired and internalised the necessary skills and knowledge from those experiences.

For organisations that administer cognitive ability tests to different age groups, these findings also have several implications for the interpretation of test scores. Despite the common belief that younger candidates have a distinct advantage in cognitive ability tests, research shows that decline in intellectual capability is only noticeable after the age of 50 to 60. This suggests that similar standards of cognitive ability performance can be applied to candidates of different age groups. Nonetheless, the evidence of cohort effects highlight that it is always ideal to use age-appropriate norm groups for the interpretation of test scores, especially in speeded tests or tests of numerical ability.


Loh Yu Xing is Manager, Selection & Assessment, in the Centre for Leadership Development in the Civil Service College. The Centre provides leadership assessment, leadership development and leadership research services.


01. Cognitive domains assessed in the study: Perceptual Speed — The ability to find figures, make comparisons, and carry out other simple tasks involving visual perception with speed and accuracy; Numeric Ability — The ability to understand numerical relationships and compute simple arithmetic functions; Spatial Orientation — The ability to visualize and mentally manipulate spatial configurations, maintain orientation with respect to spatial objects, and perceive relationships among objects in space; Verbal Ability — Language knowledge and comprehension measured by assessing the scope of a person's recognition vocabulary; Inductive Reasoning — The ability to educe novel concepts or relationships (Schaie, Dutta & Willis, 1991).

Ackerman, P. L., & Rolfhus, E. L. (1999). The locus of adult intelligence: knowledge, abilities, and nonability traits. Psychology and Aging, 14, 314–330.
Baltes, P. B., Staudinger, U. M., & Lindenberger, U. (1999). Lifespan psychology: theory and application to intellectual functioning. Annual Review of Psychology, 50, 471–507.
Horn, J. L. (1967). Intelligence: Why it grows, why it declines. Society, 5, 23–31.
Horn, J. L., & Cattell, R. B. (1967). Age differences in fluid and crystallized intelligence. Acta Psychologica, 26, 107–129.
McDonald-Miszczak, L. A., Hertzog, C., & Hultsch, D. F. (1995). Stability and accuracy of meta-memory in adulthood and aging: A longitudinal analysis. Psychology and Aging, 10, 553–564.
Raven, J. (2000). The Raven's Progressive Matrices: Change and stability over culture and time. Cognitive Psychology, 41, 1–48.
Schaie, K. W. (1983). The Seattle Longitudinal Study: A twenty-one year exploration of psychometric intelligence in adulthood. In K. W. Schaie (Ed.), Longitudinal studies of adult psychological development (pp. 64–135). New York: Guilford Press.
Schaie, K. W. (1994). The course of adult intellectual development. American Psychologist, 49, 304–313.
Schaie, K. W. & Hertzog, C. (1986). Towards a comprehensive model of adult intellectual development: Contributions of the Seattle Longitudinal Study. In R. J. Sternberg (Ed.), Advances in the psychology of human intelligence (Vol. 3, pp. 79–118). Hillsdale, NJ:Erlbaum.
Schaie, K. W., Dutta., R., & Willis, S.L. (1991). The relationship between rigidity-flexibility and cognitive abilities in adulthood. Psychology and Aging, 6, 371–383.
Zelinski, E. M., & Burnight, K. P. (1997). Sixteen-year longitudinal and time lag changes in memory and cognition in older adults. Psychology and Aging, 12, 503–513.
Zelinski, E. M., & Kennison, R. F. (2007). Not your father's test scores: Cohort reduces psychometric aging effects. Psychology and Aging, 22, 546–557.
Zelinski, E. M., Kennison, R. F., Watts, A., & Lewis, K. L. (2009). Convergence between cross-sectional and longitudinal studies: Cohort matters. In H. B. Bosworth & C. Hertzog (Eds.), Aging and Cognition (pp. 101–118). Washington, DC: American Psychological Association.
Last updated: 10 September 2012
Copyright © 2014 Civil Service College. All rights reserved