Crystallized intelligence increases through emerging and young adulthood and then

Openness to Experience reflects one of the five fundamental dimensions of personality (McCrae, 1993–1994). Broadly, it refers to individual differences in the proneness to be original, complex, creative, and open to new ideas (cf. John & Srivastava, 1999). As such, it includes a motivational component, which is based on a general interest in novelty, complexity, and tolerance of ambiguity. In addition, it contains a cognitive component, which refers to the manner in which information is processed and organized. According to Costa and McCrae (1995), Openness to Experience is composed of six facets: Fantasy (a tendency toward a vivid imagination and fantasy life), Aesthetics (a tendency to appreciate art, music, and poetry), Feelings (being receptive to inner emotional states and valuing emotional experience), Actions (an inclination to try new activities, visit new places, and try new foods), Ideas (a tendency to be intellectually curious and open to new ideas and an active pursuit of intellectual interests for their own sake), and Values (a readiness to re-examine traditional social, religious, and political values). Although the conceptualization of Openness and, hence, its label have been the subject of some scientific debate (cf. McCrae & Costa, 1997), in the present study we use the term “Openness to Experience” throughout, because we administered the scale with the same name from the NEO-FFI (Costa & McCrae, 1992b).

A common assumption is that Openness to Experience as a personality trait refers to a enduring pattern of thoughts, feelings, and behaviors. However, there is both cross-sectional and longitudinal evidence for small, but systematic age changes or age-related differences in personality traits at various ages (e.g., Allemand et al., 2007, Allemand et al., 2008, Allemand et al., 2008, Roberts et al., 2003, Roberts et al., 2006, Terracciano et al., 2005). With respect to Openness to Experience, Roberts et al. (2003) concluded that cross-sectional studies show a slight decrease with age across adulthood (e.g., Costa et al., 1986, McCrae et al., 1999). For example, in a large sample of Internet users aged 21–60 years, Openness to Experience showed a small decline with age (Srivastava, John, Gosling, & Potter, 2003). Srivastava et al.’s findings were similar to those reported in McCrae et al.’s (1999) multi-national studies with a total sample size of over 12,000 adults, where, across cultures, the median correlations of age with Openness to Experience was −0.08. Recently, in a sample of Medicare patients aged 65–100 years, Weiss et al. (2005) reported a similar negative correlation between age and Openness to Experience (r = −.07). Longitudinally, Roberts et al. (2006) demonstrated that, on average, people show increases in Openness to Experience in the college years. During adulthood, Openness to Experience remains largely unchanged, but declines slightly after the age of 60. A similar picture emerged from studies focusing on adulthood and old age. For example, across a 6-year longitudinal time span Small, Hertzog, Hultsch, and Dixon (2003) found a moderate decline in Openness for adults initially aged 55–85 years. Likewise, Schaie, Willis, and Caskie (2004) reported a modest longitudinal increase of Openness to Experience until age 46, a plateau until the late sixties, and a modest decline thereafter. Terracciano et al. (2005) offered a more differentiated picture of the developmental trajectory of Openness to Experience by examining the six facets of Openness (cf. Costa & McCrae, 1995). Following the general trend, the facets Openness to Values, Openness to Feelings, and Openness to Actions showed a small linear decline from age 30 to 90. By contrast, the remaining three facets of Openness, i.e., Openness to Aesthetics, Openness to Ideas, and Openness to Fantasy, exhibited almost no decline, on average. These latter findings indicate that merely considering Openness as a whole might result in an overly simplified picture of age-related changes in Openness.

A possible relation between Openness to Experience and intelligence has been investigated in a number of studies during the last years. Typically, it was found that Openness to Experience shows substantive correlations with measures of intelligence (e.g., Chamorro-Premuzic et al., 2005, Goff and Ackerman, 1992, Moutafi et al., 2003). Specifically, in adult samples, Openness to Experience has been shown to relate to general intelligence with correlations ranging from about r = .15 (Moutafi et al., 2003) to r = .42 (Holland, Dollinger, Holland, & MacDonald, 1995). In a meta-analysis based on 135 studies, Ackerman and Heggestad (1997) examined a variety of personality and intelligence measures regarding their intercorrelations. Distinguishing between fluid and crystallized intelligence, the authors found that Openness correlated weakly with fluid intelligence (r = .08), while the association with crystallized intelligence was moderate (r = .30). Recently, Ashton and colleagues (2000) applied the Multidimensional Aptitude Battery to assess fluid and crystallized intelligence and the Personality Research Form scales to measure Openness in a sample of 508 adolescents and adults. They reported that Openness correlated r = .18 with fluid intelligence and r = .37 with crystallized intelligence. While a number of studies have found that Openness is a strong predictor of crystallized intelligence but only weakly related to fluid intelligence (e.g., Bates and Shieles, 2003, Costa and McCrae, 1992a), in some studies crystallized and fluid intelligence did not differ in their relation to Openness to Experience (e.g., Austin et al., 1997, Holland et al., 1995, McCrae, 1993–1994). Notwithstanding, the typical finding appears to be that, in adult samples, Openness is weakly correlated with fluid intelligence, while correlations with crystallized intelligence are in the medium range (rs being around .35) (Ackerman and Goff, 1994, Goff and Ackerman, 1992, Rocklin, 1994).

Studies that distinguished between the facets of Openness to Experience and their relations to intelligence are scarce. Regarding general intelligence, the facet Openness to Ideas repeatedly emerged as a significant predictor (Harris, 2004, Moutafi et al., 2003). Goff and Ackerman (1992) found that Openness for Ideas explained 10% of variance in crystallized intelligence, but only 2% in fluid intelligence. By contrast, Openness for Values and Openness for Aesthetics explained 2% and 3% of variance in crystallized intelligence, respectively (cf. Ackerman & Goff, 1994). Focusing on fluid intelligence, Moutafi, Furnham, and Crump (2006) showed that the two facets Openness to Ideas and Openness to Actions were positively correlated with fluid intelligence (rs = .20, .07, respectively). In sum, it appears that the facet Openness to Ideas is associated with both fluid and, in particular, crystallized intelligence.

In order to account for the Openness–intelligence associations reported above, Chamorro-Premuzic and Furnham (2004) have recently argued that some personality traits may play a significant role in the process of skill acquisition in that they may influence choices to engage or invest in particular domains of knowledge. Hence, Openness to Experience might lead to engaging in intellectually beneficial activities, which, in turn, may strengthen the development of intellectual abilities, particularly crystallized intelligence. Similar arguments have been offered by Ackerman, 1994, Ackerman, 1996, who emphasized the role of non-ability attributes in the development of intelligence. Based on the assumption that fluid intelligence is cumulatively invested into specific domains of knowledge and, ultimately, transforms into crystallized intelligence, the intensity and direction of fluid intelligence investment over a longer period of time may be determined by motivation, interests, and personality traits, e.g., Openness to Experience (Ackerman, 1994, Ackerman, 1996, Ackerman and Heggestad, 1997). This rationale is inspired by Cattell’s (1963) investment hypothesis, which posits that fluid intelligence turns into crystallized intelligence by continuously being directed into specific areas of knowledge. As outlined in his triadic theory, Cattell (1987) posited that for channelling fluid intelligence, personality plays an important role, along with specific training and experience. Specifically, the variety of exposure, the time and energy spent, the reinforcement schedules in particular areas of experience, and the sentiments and motivational systems that grow up around certain activities are believed to be influenced by personality traits to a substantive extent. Openness to Experience, in turn, is descriptive of a tendency for individuals to immerse themselves in a task and to fully comprehend an area of interest (McCrae, 1996). As such, Openness to Experience might affect the direction and intensity of the investment of fluid intelligence and, therefore, lead to individual differences in the breadth and depth of the acquisition of knowledge and expertise, that is, crystallized intelligence (Ackerman & Beier, 2003).

A noteworthy feature of the approaches aiming to account for the personality–intelligence relation is that they draw on developmental processes to explain a link between Openness to Experience, fluid intelligence, and, especially, crystallized intelligence. Fluid intelligence is invested into specific domains of knowledge over time, which leads to the cumulative evolvement of crystallized intelligence. In accordance with this assumption, a number of studies have shown that fluid and crystallized intelligence follow different trajectories across the adult lifespan. While fluid intelligence tends to decrease linearly after the age of thirty, crystallized intelligence remains stable into old age (e.g., Horn and Cattell, 1966, Horn and Hofer, 1992, Salthouse, 1991, Schaie, 2005). In keeping with a developmental view, Openness to Experience might explain some proportion of the age changes and age-related differences in fluid and crystallized intelligence. If one assumes that a person high in Openness to Experience engages to a larger extent in novel activities, this may, through a lifetime of practice, lead to higher levels of fluid and crystallized intelligence (Chamorro-Premuzic et al., 2005, Schaie et al., 2004). Specifically, if the process of engaging in a manifold of novel activities is cumulative across the lifespan, one might expect that the relation between Openness to Experience and fluid and crystallized intelligence would be stronger in older than in younger or middle-aged adults. As Hofer and Sliwinski (2001) have demonstrated, if two variables develop together on the individual level, this should lead to a dedifferentiation across time or, cross-sectionally, age groups. That is, if the development of Openness to Experience within the individual affects the development of fluid and crystallized intelligence, over time Openness to Experience as well as fluid and crystallized should become more strongly related because coupled intraindividual processes lead to more pronounced and stronger associated interindividual differences. Empirically, such a dedifferentiation would find expression as a cross-sectional interaction between age, Openness and intelligence.

The main goal of the present study was to more closely examine the interplay between individual differences in Openness to Experience and fluid and crystallized intelligence in two large, representative samples of middle-aged and old participants from Germany. Specifically, if the investment hypothesis should hold, we would expect the linkage between Openness to Experience and, especially, crystallized intelligence to be stronger in the older age group (Cattell, 1987, McCrae, 1987). That is, there should be an interaction of age group with the Openness–intelligence link. A related objective was to investigate the Openness–intelligence association on the facet level rather than for Openness to Experience as a whole, which has the potential to provide new insights into the processes that govern personality–intelligence relations. Because we utilized the short form of the NEO-Personality Inventory, the NEO-FFI (Costa & McCrae, 1992b), which does not contain facets, we examined whether nine items designated to measure Openness to Experience fell into three item clusters. The allocation of items to item clusters was motivated by previous work of Saucier, 1998, Chapman, 2007, who demonstrated that item clusters of Openness to Experience as found in the NEO-FFI highly corresponded to the Openness facets of the NEO-PI-R. Specifically, the item cluster Aesthetic Interests was strongly related to the facet Openness to Aesthetics (r = .83), Intellectual Interests mirrored Openness to Ideas (r = .90), and Unconventionality was highly similar to Openness to Values (r = .75). Note that item clusters are somewhat broader and based on fewer items than facets.

Methodologically, we built on and extended previous studies in several respects. While in most previous studies analyses have been conducted using sum or scale scores as measures of Openness to Experience (or its facets) (e.g., Ackerman and Goff, 1994, Baker and Bichsel, 2006, Chamorro-Premuzic and Furnham, 2004), we used structural equation modeling. Conducting analyses on the latent variable level required two additional methodological refinements: First, in order to make factor scores of both Openness to Experience and intelligence comparable across age groups, we investigated different degrees of measurement invariance (Zimprich, Allemand, & Hornung, 2006). As Horn and McArdle (1992) have argued, if evidence supporting a measure’s invariance is lacking, conclusions based on that measure are, at best, ambiguous and, at least, incorrect. We, thus, aimed at establishing strong measurement invariance for all constructs involved in the analyses (cf. Meredith, 1993). Eventually, we more closely examined the multivariate interplay among the three item clusters of Openness to Experience and fluid and crystallized intelligence by conducting a commonality analysis (Pedhazur, 1982).