Age-Related Differences in

Basic Cognitive Processes:

Implications for Work

Timothy A. Salthouse
Georgia Institute of Technology, Atlanta, Georgia,
USA

Negative relationships between age and proficiency in many basic cognitive

processes have been well documented in the research literature, and there is
also evidence that work performance is positively related to the eficiency of
basic cognitive processes. It does not necessarily follow, however, that in-
creased age is a handicap in most occupational situations, because experience
may play an important role in moderating the influence of age on work
performance. The relationships among age, experience, and performance
both in basic cognitive tusks and in more complex work environments are
brieJy discussed.

Two relationships involving cognitive ability have been well docu-

mented and, when considered together, appear to have important impli-
cations for how aging is related to work performance. The first relation-
ship is the negative one between age and measures of the efficiency of
basic cognitive processes, as reflected by performance on cognitive tests
of the type used in many assessments of intellectual or cognitive ability.
Results indicating that increased age is often associated with lower per-
formance on various tests of cognitive ability have accumulated for more
than 70 years (see Salthouse, 1991a, for a review). For example, in a
recent study (Salthouse & Mitchell, 1990), 383 adults between 20 and
83 years of age were administered four standard tests of cognitive ability
(Series Completion, Letter Sets, Surface Development, and Paper Fold-
ing). The correlations between age and performance on these tests were
- .25, - .26, - .30, and - .38, respectively, values that are typical of
many results in the literature.
Of course, not all aspects of cognitive functioning exhibit age-related
declines of this magnitude. In fact, little or no age-related declines have

A version of this paper was originally presented at the 2nd International Stein Conference.
Accepted for publication March 1993.
Address correspondence to Timothy A. Salthouse, School of Psychology, Georgia Institute of
Technology, Atlanta, GA 30332. USA.

Experimenral Aging Research, 20:249-255, 1994

Copyright 0 1994 Taylor & Francis
0361-073x194$10.00 .OO + 249
250 T . A . Salthouse

been observed in the amount of information or knowledge a person pos-

sesses (see Salthouse, 1991a, for a review). Partially because of the
different age trends, researchers have differentiated two broad types of
cognition, one corresponding to the efficiency of processing at the time
of assessment (often termed fluid intelligence) and the other representing
the cumulative products of processing from earlier periods in one’s life
(usually referred to as crystallized intelligence). Both types of cognition
are important, but because it is fairly large and negative, the relationship
between age and the fluid aspects of cognition has received more attention
than has the general absence of a relationship between age and the crys-
tallized aspects of cognition.
The second relationship is that between cognitive ability and perfor-
mance in work environments. A common method of indicating the mag-
nitude of the relationship between a predictor variable, such as cognitive
ability, and job performance is to compute the correlation between the
two variables, which in this context is referred to as the validity coeffi-
cient. Most validity coefficients reported between cognitive ability and
job proficiency have been positive, but estimates of the average magni-
tude of the relationships have varied according to the procedures used to
combine individual values. For example, Ghiselli ( 1973) summarized
results from thousands of studies conducted over many years and reported
a median correlation of approximately .24 between scores on a single
intellectual test and various measures of job proficiency. More recently,
Hunter and Hunter (1984), using procedures that corrected for measure-
ment error and for restriction of range, concluded that the average validity
of cognitive ability tests for the criterion of job proficiency was .45.
Whatever the absolute value of the relationship, however, it seems clear
that higher levels of cognitive ability, as measured by performance on
cognitive tests, are associated with better performance in work situations.
The relationships among age, cognitive ability, and work performance
are illustrated in a somewhat different manner in Figure 1. This figure is
based on separate analyses of the Wechsler Adult Intelligence Scale-
Revised (Wechsler, 1981) standardization data. The top panel in Figure
1 presents data from the tests that make up the Verbal scale, which tend
to reflect cumulative knowledge, whereas the bottom panel illustrates
data from the tests that make up the Performance scale, which all reflect
functioning at the time of assessment. The age trends were derived by
converting the mean summed scale scores (from Wechsler, 1981) into
IQs on the basis of the distribution of scores for adults between 20 and
24 years of age. The occupational differences are based on the data from
Kaufman (1990) and represent the mean scores for people between 20
and 54 years of age in each occupational category.
Age, Cognition, and Work 25 1

The variations in cognitive ability associated with the occupational

categories presented in Figure 1 are not equivalent to validity coefficients
because they merely reflect differences in the average ability level of
persons in different occupational classifications, and not the degree to
which cognitive ability was related to proficiency or success in those
occupations. The two types of information are nevertheless similar in
indicating that cognitive ability is related to functioning in work environ-
ments, in terms of either level of performance in a given job or the
likelihood of placement in higher status, and more cognitively demand-
ing, positions. Moreover, Figure I indicates that for some measures of
cognitive ability, the age-related differences are large relative to the dif-
ferences across occupational categories.
The two relationships just described could be interpreted as implying
an argument of the following form: (a) Increased age is associated with
lower performance on some cognitive measures, (b) lower performance
on these cognitive measures is associated with poorer job performance or
placement in less cognitively demanding occupations, and therefore ( c )
increased age can be expected to be associated with poorer job perfor-
mance or shifts into occupations with reduced levels of cognitive de-
mands.
Contrary to the implications of this argument, however, most research-
ers who have reviewed the relationship between age and job performance
have concluded that there is little convincing evidence that older workers
are either less productive or less competent than young workers (e.g.,
Davies & Sparrow, 1985; McEvoy & Cascio, 1989; Rhodes, 1983; Wald-
man & Avolio, 1986).
A number of factors might be proposed to account for the discrepancy
between the expectation of poorer work performance with increased age
and results indicating that there is little or no relationship between age
and most aspects of work performance. One possibility is selective attri-
tion in the workplace, such that only the most competent individuals
survive in the job until middle or late adulthood. Another possibility is
that many assessments of job performance may not be very sensitive,
particularly with respect to dimensions of the job that require cognitive
abilities.
However, the most important factor contributing to the presence of age
trends in performance on measures of cognitive ability but not in work
performance may be the amount of relevant experience. That is, increased
age is usually associated with greater experience, and experience is likely
to be advantageous in work situations but not necessarily in tests of
cognitive ability. Although the idea that experience helps minimize the
negative relationship between age and work performance seems quite
25 2 T . A. Salthouse

130 VERBAL SCALE

120
................................................... Professional & Technica
110
Managerial, Clerical, Sale!
100

...................................................
90
................................................... Unskilled

80

70
a I I I I I I
20 30 40 50 60 70

130 PERFORMANCE SCALE

120

110 .................................................. Professional & Technical

Managerial, Clerical, Sales

100

90

80

70
I I I I I I
20 30 40 50 60 70
Chronological Age

Figure 1. Age and occupational differences in performances of the standard-

ization sample ( N = 1,480) on the Verbal and Performance Scales of the Wechs-
ler Adult Intelligence Scale-Revised.
Age, Cognition, and Work 253

plausible, the mechanisms of this influence are not yet obvious. For
example, one possibility is that extensive experience leads to maintained
proficiency of basic cognitive processes that would decline in the absence
of that experience. An alternative possibility is that experience contributes
to stable or improved levels of job performance without affecting the
proficiency of basic cognitive processes. Although both of these interpre-
tations could account for the finding of little or no relationship between
age and work performance, they should be distinguished because they
have quite different implications. For example, if it were discovered that
age-related declines in relevant basic cognitive processes were evident in
persons with extensive experience, then low levels of work performance
might be predicted if older individuals were to switch to a new job where
their previous experience was not applicable.
My colleagues and I have found evidence relevant to these two alter-
native mechanisms by which experience might moderate relationships
between age and work performance in three studies (Salthouse, 1991b;
Salthouse et al., 1990; Salthouse & Mitchell, 1990). All of the studies
were designed to investigate the relationships among age, experience,
and performance on cognitive tests assessing various spatial abilities. For
example, one of the tests used in these projects was a paper-folding test
in which the examinee sees rectangles with illustrations of a sequence of
folds followed by a hole punch, and then is asked to select which pattern
of holes would result if the paper were unfolded. Another test was a
surface development test in which a drawing of an object assembled and
unassembled are presented, and the task is to identify corresponding
regions in the two drawings.
The three separate studies were conducted to determine the effects of
presumably relevant experience on the relationships between age and
performance on these types of tests. The research participants in one
study consisted of 47 practicing architects ages 21-71 (Salthouse et al.,
1990). We chose architects as the research subjects because we assumed
that spatial abilities are needed on an almost daily basis in this occupation.
Despite what can be presumed to be extensive and nearly continuous
experience with work requiring spatial ability, the age trends in this
sample’s performance on measures of spatial ability were nearly identical
to those observed in unselected samples of adults.
The participants in the second study (Salthouse, 1991b) were mainly
engineers and other users of computer-assisted design and computer-
assisted manufacturing systems. The tests in this project were designed
to involve materials relevant to these job categories (e.g., recognizing
objects from orthographic drawings). As in the previous study, however,
there was almost no effect of experience on the magnitude of the age
relationships in these tests.
254 T . A. Salthouse

The third study involved a total of 383 adults who, in addition to

performing several cognitive tests, completed a questionnaire regarding
their experience with activities requiring spatial abilities (Salthouse &
Mitchell, 1990). For example, three of the items asked how frequently
the individual “imagined different arrangements of furniture or other
objects,” “considered how an object or building would look from a
different viewing position,” and “produced or interpreted technical
drawings of three-dimensional objects. ” The major results from this study
were identical to those of the earlier studies in that the relationships
between age and performance on the spatial tests were very similar among
individuals with varying amounts of relevant experience.
These three studies indicate that experience does not moderate the
relationships between age and basic cognitive processes. Because it seems
unlikely that the older architects or engineers who served in these studies
were markedly less competent than their younger colleagues, the greater
experience associated with increased age can be postulated to have led to
a larger accumulation of occupation-specific knowledge. For example, in
the case of architects, greater experience is likely to have resulted in more
extensive knowledge of clients’ needs, the suitability of building materials
and structures to sites, building codes and zoning regulations, and so on.
All of these knowledge factors probably contribute to successful perfor-
mance in the job, and they are almost certain to be positively related to
amount of experience.
The preceding discussion can be summarized as follows. Cognitive
ability is an important determinant of work performance, and some as-
pects of cognitive functioning appear to decrease with increased age.
However, there is little evidence that aging is associated with lower levels
of work performance, perhaps because with age often comes increased
experience. At least two possible mechanisms for experiential influences
can be identified: preservation of a high level of cognitive ability or larger
accumulation of relevant knowledge. Results of recent laboratory-based
research seem inconsistent with the preserved-abilities interpretation, but
a clear need exists for research in actual work settings on the relationships
among age, experience, cognitive ability, job knowledge, and job per-
formance. Research of this type will be difficult, but the efforts would
be justified because of the utility of the resulting information in guiding
job redesign and improving the effectiveness of retraining programs and
other interventions intended to maximize the fit of older adults to work
environments.

REFERENCES

Davies, D. R . , & Sparrow, P. R . (1985). Age and work behavior. In N . Charness (Ed.),
Aging and human peflortnance (pp. 293-332). New York: Wiley.
Age, Cognition, and Work 255

Ghiselli, E. E. (1973). The validity of aptitude tests in personnel selection. Personnel

Psychology, 26, 46 1-477.
Hunter, J . E., & Hunter, R. F. (1984). Validity and utility of alternative predictors of
job performance. Psychological Bulletin, 96, 72-98.
Kaufman, A. S . (1990). Assessing adolescent and adult intelligence. Boston: Allyn &
Bacon.
McEvoy, G. M., & Cascio, W. F. (1989). Cumulative evidence of the relationship
betweeen age and job performance. Journal of Applied Psychology, 74, 11-17.
Rhodes, S. R. (1983). Age-related differences in work attitudes and behavior: A review
and conceptual analysis. Psychological Bulletin, 93, 328-367.
Salthouse, T. A. (1991a). Theoretical perspectives on cognitive aging. Hillsdale, NJ:
Erl baurn.
Salthouse, T. A. (1991b). Age and experience effects on the interpretation of orthographic
drawings of three-dimensional objects. Psychology and Aging, 6, 426-433.
Salthouse, T. A,, Babcock, R. L., Skovronek, E., Mitchell, D. R. D., & Palmon, R.
(1990). Age and experience effects in spatial visualization. Developmental Psychol-
ogy, 26, 128-136.
Salthouse, T. A., & Mitchell, D. R . D. (1990). Effects of age and naturally occurring
experience on spatial visualization performance. Developmental Psychology, 26, 845-
854.
Waldman, D. A., & Avolio, B. J . (1986). A meta-analysis of age differences in job
performance. Journal qf Applied Psychology. 71, 33-38.
Wechsler, D. R. (1981). Wechsler Adult Intelligence Scale-Revised. New York: Psycho-
logical Corporation.