The Cattell–Horn–Carroll (CHC) model of intelligence views creativity as a first-level factor within the second-level factor of broad retrieval ability (Gr), alongside other first-level abilities such as ideational fluency and word fluency. Traditional methods of measuring creativity, however, confound idea quality with idea quantity, which might exaggerate the relationship between creativity scores and verbal fluency factors. Participants (n = 131 adults) completed two divergent thinking tasks (unusual uses for a rope and a box), which were scored using newer methods that effectively separate creativity (scored via subjective ratings) and fluency (scored as number of responses). They then completed 16 verbal fluency tasks that assessed six lower-order Gr factors: word fluency, associational fluency, associative flexibility, ideational fluency, letter fluency, and dissociative ability. Viewed singly, many of the lower-order factors significantly predicted creative quality and fluency. General Gr had substantial effects on creative quality (standardized β = .443) and fluency (β = .339) in a higher-order model as well as in a bifactor model (quality β = .380, fluency β = .327). Moreover, general Gr was the only significant predictor in the bifactor model, suggesting that it, not the specific factors, was most important. All effects were essentially the same after controlling for typing speed and vocabulary knowledge. The findings thus support the CHC view of creativity/originality as a lower-order component of Gr, illuminate the relationships between creativity and first-level Gr factors, extend the study of creativity and intelligence beyond fluid intelligence, and further indicate that creativity is more closely tied to cognitive abilities than creativity research has yet recognized.
How do people come up with clever and creative ideas, and why are some people better at it than others? Most research on these questions has used divergent thinking tasks, which prompt people to generate ideas than can be scored, based on a variety of systems, for creativity (Kaufman et al., 2008 and Plucker and Renzulli, 1999). In the Cattell–Horn–Carroll (CHC) model of cognitive abilities (McGrew, 2005 and McGrew, 2009), idea generation tasks fall under the second-level factor known as broad retrieval ability, abbreviated as Gr ( Carroll, 1993). But as many researchers have argued, traditional methods for assessing divergent thinking yield only a fluency score—the simple number of valid responses—or yield quality scores that are confounded with quantity ( Hocevar, 1979b, Michael and Wright, 1989 and Silvia et al., 2008). Two problems result: (1) divergent thinking tasks might resemble verbal fluency tasks too closely, leading to questions of construct validity, and (2) the weak correlations between creativity and intelligence observed in past work ( Kim, 2005) might be due to weak assessment of creativity, not to a genuinely small effect size.
The present research thus addresses two issues. First, when newer assessment methods that effectively dissociate creativity and fluency are used, how does creativity fit within the Gr domain? The dominance of fluency-based scoring systems available at the time of Carroll's (1993) landmark analysis might have inflated the association of divergent thinking and Gr. Second, how does divergent thinking relate to both the Gr factor and to its first-level factors? What first-level factors contribute the most to generating creative ideas? In the present research, people completed two divergent thinking tasks and 16 Gr tasks that mapped on to six lower-order Gr factors: word fluency, associational fluency, associative flexibility, ideational fluency, letter fluency, and dissociative ability. Using structural equation modeling, we estimated the contributions of the lower-order factors and the higher-order Gr factor—modeled using higher-order and bifactor models—to both the quality and quantity of responses to the divergent thinking tasks.
The present research examined the role of the CHC Gr factor in divergent thinking. Because past assessment methods have confounded creative quality and quantity—or assessed only quantity—the role of Gr in divergent thinking may have been exaggerated. When new assessment and scoring methods are used—ones that emphasize to participants that they should try to be creative and that use subjective scoring to dissociate creativity and fluency—a significant role for the second-level Gr factor appears for both the creative quality and the amount of ideas. Taken together, the findings extend the emerging CHC approach to creative abilities (Nusbaum and Silvia, 2011 and Silvia and Beaty, 2012) and further suggest that intelligence and creativity are more closely linked than creativity theories acknowledge.
