Psychology and Aging 1988, Vol. 3, No. 1,59-74 Copyright 1988 by the American Psychological Association, Inc. 0882-7974/88/S00.75 Aging, Attention, and Intelligence Lazar Stankov University of Sydney, Sydney, New South Wales, Australia The aims of this study were (a) to find out if attentional ability factors that are separate from the well-established ability factors (e.g., fluid intelligence, crystallized intelligence, and short-term acqui- sition and retrieval function) can be identified, and (b) to establish, through the use of part corre- lations, whether attentional abilities play a role in the changes in fluid and crystallized intelligence that occur with increasing age. A battery of 36 tests (19 psychometric tests and 17 measures of attentional processes) were given to 100 people, 20 of whom were in each age decade between 20 and 70. Results indicated that three attentional factors—Search, Concentration, and Attentional Flexibility—exist at the primary-ability level and that all three define fluid intelligence at the second- order level. Results also indicate that the decline in fluid intelligence with increasing age disappears if attentional factors are parted-out. Similarly, the increase in crystallized intelligence with increasing age becomes even greater if one controls for attentional processes. I conclude that changes in atten- tional processes play an important part in changes in human intelligence with age. As people grow old, physical and psychological changes take place in their bodies and minds. Typically, psychological changes are thought of as caused by the physical changes. For example, it is more acceptable to say that decline in intellectual abilities follows changes in the cerebral blood flow because of "hardening" of the arteries, than to say that slower blood flow represents a consequence of lower intelligence. Among the physical changes affecting intelligence, the most important are those of the central nervous system, or neurological changes. Psychometric measures of fluid abilities (Gf) show a decline during the adult years and, therefore, a negative correlation with chronological age. They largely reflect these neurological influences. People tend to believe in the importance of another set of influences on intelligence during the adult years, also. These influences are the accumulation of knowledge, which can occur either through idiosyncratic and incidental learning ex- periences or through formal education and acculturation. These influences tend to offset the effects of neurological de- cline. Idiosyncratic learning is reflected mainly in the improve- ment of performance on fluid intelligence tasks, whereas formal education and acculturation are reflected in performance on crystallized intelligence tests. Psychometric measures of crys- tallized abilities (Gc) show a slight increase or remain at the same level until the retirement age. During the adult years, cor- The work on this article was supported in part by the Australian Na- tional Health and Medical Research Committee Grant and Australian Research Grants Scheme. I am grateful to Kuei Chen, John Crawford, Gerry Fogarty, Mary- Ann McGill, Brett Myors, Richard Roberts, Georgina Spilsbury, and Lucy Sullivan for their help in working on this project. I am also grateful to two unknown reviewers for their unusually thorough and construc- tive criticisms of an earlier draft of this article. Correspondence concerning this article should be addressed to Lazar Stankov, Department of Psychology, University of Sydney, Sydney, New South Wales, 2006, Australia. relation between age and Gc is, therefore, either low positive or zero (see Horn, 1982, 1986). With the development of information-processing models in cognitive psychology during the 1970s, experimental psycholo- gists working outside psychometric traditions, became inter- ested in exploring age-related changes in these processes. Cog- nitive psychologists tended to assume that the elements in their models of cognitive functioning are fundamental to intelli- gence. For example, it was assumed that the decline in fluid intelligence with age can be attributed to the decline in short- term memory capacity rather than vice versa, (see Crawford & Stankov, 1983; Horn, 1982). The treatment of cognitive pro- cessing variables as causal of individual differences in intelli- gence test performances, somewhere in between physical and psychometric domains, has become an entrenched part of psy- chological theory in the 1980s. 1 The justification for treating experimental cognitive variables as basic appears to derive from the notion that psychometric abilities reflect only those cognitive processes that show pro- nounced individual differences. Cognitive psychology is broader in scope, inasmuch as it proceeds in a theoretical 1 One might reasonably argue that intelligence test scores can be treated as a reflection of the overall cognitive proficiency, perhaps bio- logically determined and akin to mass action. In this case, intelligence, rather than cognitive processing variables, would be considered more fundamental. Although there is nothing logically wrong with such a view and, indeed, some psychometricians may strongly endorse it, the general Zeitgeist of psychological thinking at the moment seems to be along the views espoused in the main body of this article. From the information provided in Table 4, a reader of such a persuasion can easily calculate part correlations that would show, say, how much change in concentration can be accounted for by the decline in fluid intelligence. It is conceivable that both intelligence and attention may reflect the operation of a yet another unmeasured source variable (or variables) in their effects on cognitive aging. Present analyses are not designed to disclose this latter possibility. One should nevertheless be open to such an option in future. 59