Which of the following types of intelligence is most likely to decline in late adulthood?

Cognitive Ability

Cognitive ability is defined as a general mental capability involving reasoning, problem solving, planning, abstract thinking, complex idea comprehension, and learning from experience (Gottfredson, 1997). Probably the most comprehensive taxonomy for cognitive abilities is the three strata model derived by Carroll (1993). The first stratum consists of specific and narrow abilities, the second includes group factors and broad abilities, and the third stratum is general intelligence or g (Ones et al., 2012).

Cognitive ability is widely considered the best predictor of job performance (Schmidt and Hunter, 1998). Meta-analytic reviews and primary studies link cognitive ability to job performance in both United States and European countries (Ispas et al., 2010; Ones et al., 2012). The predictive validity of cognitive ability depends on the complexity of the job with the strongest validity coefficients observed for highly complex jobs (Ones et al., 2012). Cognitive ability impacts job performance through job knowledge acquisition (Borman et al., 1991, 1993); high cognitive ability individuals are better equipped to acquire the knowledge needed to perform their jobs at the highest levels. The meta-analytic estimate of the general cognitive ability–overall job performance relationship is 0.51.

A small number of studies examined citizenship performance or CWB as criteria. Alonso et al. (2008) meta-analyzed 13 studies linking cognitive ability and citizenship performance and found an uncorrected correlation of 0.05. When using CWB as the criterion, the results show validity coefficients around −0.15 to −0.20 (Ones et al., 2012). Dilchert et al. (2007) reported a corrected correlation of −0.33 between cognitive ability and objectively measured CWB.

Predicting Cognitive Ability

Cognitive abilities build upon motor accomplishments, so individual differences in infant motor milestone attainment might plausibly predict later cognitive abilities. Indeed, Joseph Campos and colleagues have argued that self-produced locomotion in the form of crawling has positive consequences for various cognitive skills. For example, self-produced locomotion can enhance perspective-taking skills. These ideas have historical parallels. Bayley and Shirley both considered whether the age of first walking was predictive of preschool mental ability, and both reported that it was; later walking was associated with lower ability scores. However, the strength of the relationship, though statistically significant, was not large, and critics subsequently argued that the relation was due primarily to the influence of a small number of cases where development was greatly delayed. There is little doubt that extreme delays in infant motor development are predictive of poorer later outcomes; the more contentious issue is whether variation in the normal range of motor development predicts later outcomes.

More recently medical researchers considered the potency of individual milestone attainment for predicting later developmental deviations or delays. Like Bayley and Shirley they found a small-to-moderate negative correlation between age of attainment and scores on later intelligence tests. However, the size the relation was too small for clinical diagnostic use (i.e., for predicting individual outcomes).

Another relevant literature has to do with the predictive value of infant development tests for predicting later cognitive abilities. As noted earlier, physical milestones are an important part of infant development tests, so the predictive success of infant development tests bears on whether or not milestone variability has any predictive utility. Generally, scores on infant tests were predictive, but the relationships were too small to allow for predicting later individual outcomes. These findings, together with those discussed above, consistently suggest that individual differences in milestone attainment have some relation to later individual differences cognitive ability. Although there is a relationship, it is too small to allow for individual prediction.

Individual core characteristics and innovation

An individual's cognitive ability provides the foundation for his or her innovative capabilities. Such cognitive abilities include intelligence, perseverance, creative thinking ability, and even pattern recognition. Cognitive ability refers to the functioning usually considered to be a person's mental faculties. In general, the higher an individual's cognitive abilities, the more able that person is to develop innovations and implement innovations from other sources. Leonardo da Vinci and Michaelangelo are perhaps the exemplars of strong cognitive abilities being linked to great innovations.

People with certain personality types have also been found to be more innovative. Those with a more creative personality tend to be more innovative as well. Characteristics that predispose one to innovation include openness to new ideas, perseverance, self-confidence, tolerance of ambiguity, independence, and originality. There are also personality traits that reduce a person's propensity for innovation. These include authoritarianism and being rules oriented. Personality, like cognitive ability, is thought to be a relatively stable aspect of a person, and thus not very amenable to alteration. While there are ways to improve both aspects, intervention techniques are usually aimed at other individual level characteristics.

Cognitive Abilities

Cognitive abilities are key competences that are needed to meet the challenges of job demands, education, and advanced training, societal expectations, and the demands of everyday life of middle-aged adults (Lachman, 2004) (see Modern Human Cognition, Evolution of). Middle adulthood provides many good opportunities for making intellectual contributions, given the position of middle-aged adults in the family, workplace, and society. Hence, middle age is the period of highest performance in a variety of individual domains. Middle-aged adults typically perceive themselves or are perceived by others as having not only more status and responsibility but also more effective intelligence and integrative skills (Lachman et al., 1994). The demands of status and widest responsibilities may encourage the development of these intellectual skills. For example, diverse work environments and workplace conditions may contribute to job-specific and individual differences in the development of cognitive abilities. As such, the development of cognitive abilities is more strongly influenced by environmental factors compared to other developmental periods (Sternberg et al., 2001).

As a consequence of diverse life contexts, changes in cognitive functioning in middle adulthood are varied and multidirectional (Schaie, 2005). Indeed, longitudinal research demonstrates that some aspects of cognitive functioning that depend on experience are maintained or even improved in midlife, while other aspects decrease (Willis and Schaie, 2005). For example, perceptual speed already begins to decline before the middle years. Number ability declines over middle adulthood and afterward. But the years from age 40 to the early 60s also reflect a period of maximum performance on some of the higher order abilities such as inductive reasoning and spatial orientation. These abilities then begin to show a decline in late-middle adulthood, though with many individual differences. As such, it is possible to detect the first indications of developmental decline in middle adulthood (Schaie, 2005).

Several risk factors or protective factors may contribute to the variability in cognitive development (Willis and Schaie, 2005). For example, multiple biosocial risk factors such as hypertension, diabetes, health status, cardiovascular disease, or stressful life circumstances may be associated to midlife cognitive decline. By contrast, several protective factors are associated with cognitive maintenance and plasticity. For example, high levels of education, complex environmental contexts and demanding activities at work, intellectual engagement with a wide range of interests, and high levels of exercise and physical activity are discussed as possible mechanisms than can be targeted in preventive interventions (cf Willis and Schaie, 2005).

Dual-Process Models and Implications for Decision Making in Older Adults

Cognitive ability is one of the possible contributors to choice set-size performance as well as preference, and dual-process models have been postulated to characterize the role of cognitive ability in decision making. When making decisions, theorists proposed that information processing involves two types of procedures (e.g., Epstein, 1994; Kahneman, 2003): System 1, which refers to an affective/experiential system; and System 2, which refers to a more deliberative/analytical system. System 1 can be thought of as automatic, effortless, rigid, heuristic-based, affective, and implicit. It is the kind of decision that can be made almost unconsciously, such as stereotyping. In contrast, System 2 is described as effortful, conscious, analytical, slow, flexible, and more resource intensive. It requires attention and concentration, such as computing and comparing probabilities (Kahneman, 2011; Stanovich & West, 2000). The two systems can work simultaneously, and affective information can also influence deliberative thinking. However, System 2 can become depleted and less efficient with effort. Given the nature of health-related decision making, it is reasonable to assume that such decisions would involve both deliberate and affective components.

A number of researchers have capitalized on dual-process models to better understand life-span changes in decision-making abilities (Peters, Hess, Västfjäll, & Auman, 2007; Peters & Bruine de Bruin, 2012; see also Hess, in this volume). Overall, there is a general consensus by those studying aging and decision making that older adults will perform worse on tasks that are more heavily dependent on System 2 processes relative to those dependent on System 1, based on the findings that aging is associated with normative decline in specific cognitive abilities typically associated with System 2 (Peters & Bruine de Bruin, 2012). For example, changes in working memory and processing speed would more directly impact System 2 type deliberative processes than System 1 type processes (Evans, 2003).

Indeed, there are now ample data to argue that age effects on choice performance and strategies are most likely associated with System 2 type processes (e.g., Hanoch, Wood, & Rice, 2007), especially when the decision-making tasks are cognitively demanding or lack supportive environments for decisions (Finucane, Mertz, Slovic, & Schmidt, 2005; Yoon, Cole, & Lee, 2009). Declines in cognitive abilities make it more difficult for older adults to navigate a complex decision-making environment that requires concentration. For example, older adults are slower in terms of processing speed, which is associated with decreased performance on other cognitive tasks (Salthouse, 1996). Also, although it remains to be investigated in more depth, the tendency for older adults to seek less information in decision-making tasks might be related to decreased working memory capacity (for a review, see Mather, 2006). These findings from the cognitive aging literature imply that aging is associated with declines in fluid abilities, such as speed of processing, working memory, and executive functioning (Schaie & Willis, 2002), precisely the abilities that characterize System 2 processing and functioning. Whether older adults are cognizant of these changes and thus are more likely to actively prefer less demanding choice environments is an open empirical question. Regardless of preference, however, their performance in different choice environments may very well decline if these environments tax System 2 types of processes.

There is support for dual-process theories in the area of medical decision-making and aging. Because older adults tend to use more health-related services, more work was done in the health domain versus other areas of decision-making abilities. Hibbard, Slovic, Peters, Finucane, and Tusler (2001) have long been interested in older adults’ abilities to understand health-related (e.g., insurance) information. In one study, they evaluated older and younger adults’ comprehension of health and financial information about health insurance. Their results indicated that older adults are more likely to make mistakes compared to younger adults. Finucane et al. (2005), in a related investigation, focused on the association between age and decision quality by varying the complexity of tasks in a number of related domains: health, financial, and dietary. Their data showed that as the task became more complex, the number of errors increased as well, with older adults experiencing even greater difficulties than their younger participants. As such, one would predict that as the number of choices increases, older adults would be less likely to make optimal decisions compared with younger counterparts.

More evidence supports the relationship between cognitive resources and decision making in aging. Based on a series of studies, Johnson (1990, 1993) had amassed sufficient evidence to show that, when deciding about cars or apartments, older adults tend to evaluate less information, reexamine information more often, need longer time to review information, and use more simplified search strategies. Mata and colleagues (Mata, von Helversen, & Rieskamp, 2010; Mata, Schooler, & Rieskamp, 2007) have provided similar results, using somewhat different tasks. In their investigations, they were interested in the relationship between aging and the ability to utilize adaptive decision strategies in a number of different environmental structures. In line with Johnson’s earlier work, Mata and colleagues found that older adults frequently use less information and require more time to evaluate it in their decision making. Furthermore, older adults often utilized simpler decision strategies due to, according to the authors, declines in cognitive abilities. A meta-analysis by Mata and Nunes (2010) provides further indication that older adults tend to use more heuristic-based decision strategies, as they often search and use less information in their decision-making process. However, other studies (Hess, Queen, & Ennis, 2013; Queen, Hess, Ennis, Dowd, & Grühn, 2013) found smaller differences in search strategies and highlight the importance of individual difference factors like education and search environment in strategy selection across the life span. Taken together, these findings appear to indicate that older adults are more likely than younger adults to adopt simpler strategies in their searches.

Principle 5: The Number of Low Scores Varies by Level of Intelligence

Cognitive abilities are related to intellectual functioning (Horton, 1999; Steinberg, Bieliauskas, Smith, & Ivnik, 2005; Steinberg, Bieliauskas, Smith, Ivnik, & Malec, 2005; Tremont, Hoffman, Scott, & Adams, 1998; Warner, Ernst, Townes, Peel, & Preston, 1987). With the WAIS–IV and WMS–IV, correlations between the FSIQ and the five WMS–IV index scores range from r=0.57 (AMI) to r=0.71 (VWMI) (Wechsler, 2009). As a result of significant and often large correlations between intelligence and cognitive domains, people with below-average intellectual abilities have more low scores on cognitive tests than people with above average intelligence.

Figure 2.7 illustrates the difference in multivariate base rates based on level of intellectual functioning. For example, one in five healthy adults with low average estimated intellectual abilities (i.e., TOPF with simple demographics estimated FSIQ) have five or more WMS–IV scores at least one standard deviation below the mean. However, having five or more WMS–IV scores at or below the 16th percentile is extremely uncommon in healthy adults with high average estimated intellectual abilities (1.4%), and was not found in any of the healthy adults with superior or very superior estimated intellectual abilities (0%) in the standardization sample. Regardless of which cut-off score is used or the number of low scores considered, there is a gradual decline in the prevalence of low scores across increasing levels of intellectual abilities. When interpreting multivariate base rates, it is very important to consider the prevalence of low scores based on level of intelligence. Using education- or demographically-adjusted norms will not remove the effect of intelligence on the base rate of low scores.

3.1 Cognitive Training

Given that cognitive abilities can and do decline with age, often resulting in difficulty in performing everyday tasks, the possibility of cognitive training to prevent, slow, or reverse age-related cognitive decline has been investigated. A growing number of studies now support the protective effects of intellectual stimulation on cognitive abilities for older adults without dementia. Early studies in the area of cognitive training were conducted within the Adult Development and Enrichment Project (ADEPT) and the SLS. Both of these studies provided 5 hours of strategy training, preceded and followed by cognitive assessment. Significant cognitive training gains were observed in both studies for the specific abilities that were trained.

A large randomized clinical trial, Advanced Cognitive Training for Independent and Vital Elderly (ACTIVE), recently evaluated the impact of three promising cognitive interventions—speed of processing training, memory training, and reasoning training—on the maintenance of both cognitive and day-to-day abilities in community-living older adults. The study showed that for all three interventions, there were significant and specific improvements in cognitive ability as well as an increased benefit of additional booster training. The amount of training gain for cognitive abilities was equal to or greater than the amount of decline that would be expected in older adults without dementia over 4 to 14 years of aging in the absence of any training.

Disturbances in Cognitive Abilities of Executive Function

Disturbances in cognitive abilities of executive function are common sequelae of TBI in both adult and pediatric populations, particularly when the injury compromises the prefrontal network (Coelho et al., 1995; Levin, Goldstein, Williams, & Eisenberg, 1991; Sohlberg & Mateer, 1989; Ylvisaker & Fenney, 1996). These executive functions include cognitive capacities involved in (a) attainment of a future goal such as the goal setting, planning and problem solving necessary for completing a project, (b) selfawareness and self-regulation, and (c) self-inhibition (Pennington, 1991). These cognitive abilities may be associated with or contribute to the deficits in higher-level discourse processing in children (Chapman, Levin & Lawyer, 1998; Ylvisaker & Fenney, 1996) and in adults (Coelho et al., 1995; Hartley, 1995; MacDonald & Johnson, 1996). It is necessary to consider the relationship between discourse function and cognitive abilities of executive control to manage effectively the cognitive–communicative disability manifested in TBI.

Special Cognitive Abilities and Cognitive Style

The cognitive abilities most frequently reported in literature include fluency, flexibility and originality of ideas, complemented with analytical and critical reasoning, besides metaphorical and analogical thinking. It is noted that the divergent thinking abilities gained prominence with Guilford’s (1967) publications on the structure of intellect theory. The assessment of these abilities has frequently been used inadequately as a general indicator of an individual’s creativity (Alencar, 1996b; Alencar & Fleith, in press), in spite of Guilford’s position that a constellation of multiple abilities characterizes creative thinking. According to him:

“the creative potential is very complex, and at times and in different ways involves abilities outside the divergent-production and the transformation categories, which are most important in that connection.” (Guilford, 1967, pp. 169–170)

The role of other cognitive abilities has been increasingly recognized in creative production, as has the contribution of convergent thinking (Cropley, 2006).

In addition, Mumford, Baughman and Sager (2003) argued that creative thought also involves the identification of new and viable solutions, and considered creative thought as a form of complex problem solving, asserting that:

“the combination and reorganization of extent knowledge structures may represent the key cognitive process underlying the generation of new ideas.” (p. 24)

These scholars have listed the following cognitive processes as appearing to play a role in creative thought: problem construction or problem definition; information encoding; category search; category selection; category combination and reorganization; idea evaluation; implementation planning; and monitoring.

The cognitive style refers to how people approach problems and their solutions, including how they generate new ideas. The perspective is on “how people are creative” or preferred ways of expressing or using one’s creativity (Treffinger, 2003). The best known classification of cognitive styles was proposed by Kirton (1987), author of the theory of adaptation–innovation. It considers that any individual can be located in a continuum that varies from the ability “to do things better” to the ability “to do things in a different way”. Those who present an adaptive style are characterized by precision, efficiency, discipline, attention to norms. On the other hand, the ones that present an innovative style tend to be undisciplined, rule breakers, and in face of a problem, they try to reorganize or restructure it with less predictable responses and a greater level of originality. The cognitive style that is described as innovator is the one that is linked to greater creativity levels. Other classifications of cognitive styles have been proposed by scholars, such as Sternberg (1997; Lubart & Sternberg, 1995) and Wechsler (2006).