The term cognitive ability generally refers to the capacity to mentally process, comprehend, and manipulate information—in short, the ability to learn. For example, reasoning deductively or inductively, grasping general principles from observing the behavior of objects, mentally rotating objects in one’s mind, quickly and accurately comprehending what one is reading, and dealing effectively with mathematical concepts are all cognitive abilities. Cognitive abilities largely constitute what most people intuitively call intelligence. Cognitive abilities are also referred to as cognitive aptitudes.
The scientific study of cognitive abilities has a long and sometimes contentious history. However, researchers’ interest has centered on two common themes: the structure of cognitive abilities (i.e., how many are there, and what do they look like?) and the impact of differences in cognitive abilities on outcomes of importance (i.e., what do they predict?).
The Structure of Cognitive Abilities
The debate over the structure of cognitive abilities is perhaps one of the most storied arguments in differential psychology. At the risk of oversimplifying its history, the debate largely centers on the question of whether there is a single general cognitive ability or many independent specific abilities. Today, most scientists accept a model of cognitive abilities similar to the one proposed by John B. Carroll. In an exhaustive and remarkable endeavor, Carroll reanalyzed more than 400 data sets spanning more than 60 years of research. His results convincingly demonstrated that the psychometric structure of cognitive abilities is best described by a hierarchal model with three basic strata or levels. At the apex is a single general cognitive ability factor, referred to as g, below which are a small number of narrow abilities, often referred to as group factors, each of which sits, in turn, on top of a large number of task-specific abilities. The primary levels of interest are the top level g) and the second level, which contains 8 to 10 narrow abilities.
Definition of the g Factor
The g factor reflects one’s general ability to learn. Formally, it is defined as the “eduction of relations and correlates,” that is, the ability to infer or deduce meaningful principles and concepts from abstract and novel situations. The g-factor is reflected in the pervasive positive correlations among any set of tests or tasks that require any form of cognitive manipulation or processing of information. That is, although more narrow mental abilities (e.g., verbal ability, quantitative ability, visual-spatial ability, short-term memory) can be identified, people who are high (or low) on any individual narrow ability tend to be relatively high (or low) on the others. Reliable measures of cognitive abilities are always positively correlated.
Definition of Narrow Abilities
There remains some slight disagreement and uncertainty regarding the exact specification of narrow abilities. In fact, Carroll himself cautioned that some slight modification or further refinement to those factors may be in order. Nonetheless, the following narrow abilities are included in most models of cognitive abilities:
- Fluid intelligence/reasoning: The ability to apply rules and premises to reach a solution; the ability to discover the underlying characteristics that govern problems
- Quantitative reasoning/skills: The ability to reason, either inductively or deductively, with mathematical concepts, relations, and properties; general knowledge of mathematical concepts
- Crystallized intelligence: The size and sophistication of one’s vocabulary; the ability to comprehend and communicate orally and use communication skills with fluency; the range of general and acculturated knowledge
- General verbal ability: The ability to recognize and decode words or disguised words; the ability to comprehend and communicate with clarity of thought and expression in written discourse; general understanding of language rules of (native) language
- Short-term memory: The ability to form and store mental representations of stimuli and then recognize or recall them after a short duration (memory span, visual memory)
- Long-term associative storage and retrieval: The ability to store and recall previously learned material regardless of whether it is meaningful; the ability to rapidly produce series of ideas, words, or other elaborative information related to a specific theme or object; the ability to rapidly produce novel or uncommon connections among stimuli or solutions to problems
- Visual-spatial processing: The ability to mentally manipulate objects or visual patterns, such as mentally rotating multidimensional objects in space; the ability to quickly discern a meaningful object from partially obscured or vague patterns and stimuli
- Auditory processing: The ability to process speech sounds; phonological awareness; the ability to discriminate speech sounds in normal and distorted contexts; the ability to discriminate tones, tone patterns, pitch, and other variations in sound qualities; the ability to localize sounds in space
- Cognitive processing speed: The ability to rapidly make simple decisions or perform simple tasks; the ability to compare visual symbols; the ability to rapidly manipulate and deal with numbers in elementary ways
Predictive Validity of Cognitive Abilities
That differences in cognitive abilities exist and that they appear to have something to do with differences in a wide array of behaviors has been recognized for several thousand years. Therefore, a wide variety of methods for assessing individual differences in cognitive abilities have been developed, and many have proved useful in understanding or predicting behaviors as varied as academic performance and technical job performance, occupational and economic attainment, delinquency, criminal behavior, accident prone-ness, and mortality, to name just a few.
Predictive Validity of g
A wealth of data has confirmed that g is predictive (at levels of both theoretical and practical significance) of individual differences in a wide range of academic, occupational, and social outcomes. Of interest to industrial and organizational psychologists, g is consistently the best single predictor of job training performance and technical job performance across a wide range of ages, settings, and domains. There is no job for which g is not at least a moderately strong predictor of technical performance.
For example, meta-analytic estimates place the predictive validity of g for technical job performance around .50 to .60 and .56 for performance in job training programs. The exact value varies depending on the nature of the job in question, however. For example, the correlation between g and technical performance for managerial and many information-dependent white-collar jobs is typically in the range of .50 to .61. By contrast, the correlation for manual labor jobs is typically around .25. It is the cognitive complexity of the occupation that moderates the predictive validity of g. In addition, g maintains its predictive validity across experience levels. In fact, some studies show that the predictive validity of g actually increases as experience increases. At first, this may seem counterintuitive: After all, people with more experience have had the opportunity to learn more and hence acquire greater levels of job-specific knowledge and skills. All things being equal, an employee with greater experience will perform better than an inexperienced employee. However, all things are not equal. It is the ability to profit from experience that is of importance, and g essentially reflects the ability to learn from experience. Those with higher g learn faster; thus, as experience increases, differences in knowledge and skills attributable to become increasing exaggerated.
Predictive Validity of Narrow Abilities
Unlike g the predictive validity of narrow abilities is tied more closely to the nature of the criteria that one seeks to predict. Thus, it is more difficult to speak of generalized findings regarding the predictive validity of narrow abilities. In addition, given the robustness of g’s predictive validity, the value of narrow abilities has typically been gauged by their incremental validity. That is, in order to merit attention as practically significant, evidence for the incremental contribution of narrow abilities above and beyond g should be substantial, dependable, and related to meaningful outcomes. However, even the most optimistic interpretation of the existing empirical literature would fall short of this standard for practical significance. That is, narrow abilities typically do not add significant amounts of incremental predictive validity above and beyond g in the prediction of academic or technical job performance. For example, after accounting for the variance attributable to g, the inclusion of the set of narrow abilities typically increases the correlation with technical job performance by less than .05. Thus, the narrow abilities, despite their psychological significance, may have only practical significance in situations in which the range of general ability has been restricted (e.g., among a group of doctoral graduate students, for whom prescreening produces significant range restriction in g or in which there is a single, domain-specific criterion to be predicted (i.e., development of reading skills).
On the other hand, there are times when one is less interested in predicting between-person differences in performance and more interested in matching an individual person’s profile of narrow abilities (i.e., their relative strengths and weakness) with the narrow ability demands of the work or educational environment. This corresponds to the dominant perspective within the vocational counseling literature. Indeed, it is well-known that g is an insufficient descriptor of work demands; occupations display distinct patterns of narrow ability demands. As such, the assessment of within-person differences in narrow abilities may be especially useful in personnel classification and academic and career counseling.
Arguably, it is the potential breadth and magnitude of the impact of individual differences in cognitive abilities that makes their study of great interest to both scientists and the general public. Although scientific treatment of cognitive abilities did not appear until the late 19th century, theory development and a wealth of empirical data accumulated since then supports two conclusions: (1) The psychometric structure of cognitive abilities is best modeled as a three-tiered hierarchical structure with a general mental ability factor, g, at the apex, and (2) general mental ability is of great functional importance in virtually every aspect of life.
- Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. New York: Cambridge University Press.
- Gottfredson, L. S. (1997). Why g matters: The complexity of everyday life. Intelligence, 24, 79-132.
- Hunter, J. E., & Hunter, R. F. (1984). Validity and utility of alternative predictors of job performance. Psychological Bulletin, 96, 72-98.
- Jensen, A. R. (1998). The g factor: The science of mental ability. Westport, CT: Praeger.
- Lubinski, D., & Dawis, R. V. (1992). Aptitudes, skills, and proficiencies. In M. Dunnette & L. M. Hough (Eds.), Handbook of industrial and organizational psychology (Vol. 3, 2nd ed., pp. 1-59). Palo Alto, CA: Consulting Psychologists Press.
- Reeve, C. L., & Hakel, M. D. (2002). Asking the right questions about g. Human Performance, 15, 47-74.
- Schmidt, F. L., & Hunter, J. E. (1998). The validity and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124, 262-274.