Personality and the Measurement of Intelligence

Abstract

SUMMARY.A re-analysis is presented of some data purporting to show that stable children differ from labile ones with respect to the structure of their intellectual abilities. The hypothesis is supported, and additional data are presented tending to show that theories of linear independence between cognitive and non-cognitive areas may have to be supplemented by theories stressing non-linear dependence.It is usually maintained that intelligence is statistically independent of temperamental factors such as neuroticism and extraversion and the evidence does, indeed, show little cause to doubt lack of correlation between the cognitive and the conative-affective sides of personality. (Cf. Cattell, 1963, for a recent study and discussion). However, it would be unwise to equate statistical independence with lack of interaction ; most studies reported in the literature have used statistical methods based on product-moment correlations, thus setting orthogonality equal to linear independence, and failing to allow for the possibility of curvilinear regression. This failure to take into account more complex modes of causation may have arisen from the fact that psychometric procedures have been developed very much in isolation, and without connection with the large body of experimental psychology. Eysenck (1957) has argued that the study of temperament and of intelligence can be enriched tremendously by regarding the performance of personality and intelligence tests from the point of view of experimental psychology, considering it as subject to the well known laws of learning theory, and making predictions from these. The usefulness of this approach to the study of personality variables, such as neuroticism and extraversion, has been demonstrated in several publications (Eysenck, 1960(Eysenck, , 1964. In this paper, we shall be concerned with a consideration of a similar approach to intelligence test problem solution.It has been argued (Eysenck, 1957) that the performance of a typical intelligence test may be regarded as an instance of massed practice, in which very similar tasks are attempted repeatedly without the interposition of a programmed rest pause TJnder these conditions, we would expect reactive inhibition to build up and interfere with the proper execution of the tasks. We would also expect that extraverted subjects, liable as they are to greater accumulation of inhibition, would show work curves different from those produced by introverted subjects, an expectation shown to be verified by two experimental studies a t a high level of significance (Eysenck, 1957, pp. 132-133). In another study, Eysenck (1959) predicted that " in the process of solving the sixty problems of the Morrisby Compound Series Test . . . extraverts would show greater reactive inhibition, and consequently falling off in performance during the last quarter of the test as compared with the first three-quarters." The results showed " that extraverts show greater work decrement . . . by taking longer to obtain correct solutjons toward the end of...

Intelligence–personality associations reconsidered: The importance of distinguishing between general and narrow dimensions of intelligence

“…For example, Austin et al (1997) found a significant positive quadratic relationship between neuroticism and intelligence, contradicting the past arguments by showing that higher test performance generally occurs for those with either high or low levels of neuroticism. Similarly, the neuroticism-differentiation hypothesis suggests that cognitive abilities are less differentiated at increased levels of neuroticism, and conversely more differentiated at decreased levels of neuroticism (Eysenck & White, 1964; see also Austin et al, 1997). In other words, g should account for more of the total observed variance in a battery of ability tests at high levels of neuroticism than it should at low levels.…”

Personality development from 12 to 18 years of age: changes in mean levels and structure of traits

“…Cattell (1957), for example, was convinced that many 'pure' personality factors have some substantial intellectual ability component and that the general ability influences some of these genuinely personality manifestations such as scope of interests or achievement motivation. Empirical studies, however, have typically found only a modest correlation between measures of personality and intelligence (Eysenck, 1994;Goff & Ackerman, 1992;Zeidner, 1995; for a metareview see Ackerman & Heggestad, 1997). Openness to Experience as measured by the Revised NEO Personality Inventory seems to be consistently related to psychometrically measured intelligence and other cognitive abilities (McCrae & Costa, 1997;Moutafi, Furnham, & Crump, 2003).…”

In the heat of the moment: On the effect of state neuroticism on task performance

“…The notion of a curvilinear relationship between neuroticism and cognitive performance has been discussed previously on a conceptual level (Eysenck, & White, 1964;Brand et al, 1994). The reported study contributes with empirical evidence to this discussion.…”

“…If trait neuroticism is identified with arousal or autonomic drive (Eysenck & White, 1964;Lynn & Gordon, 1961), and assuming that tasks in cognitive ability tests like the Raven"s Progressive Matrices are of moderate difficulty to most individuals, it follows that both high and low levels of trait neuroticism are less effective than 4 moderate levels of trait neuroticism in terms of performance on such tasks. Note however, that the Yerkes-Dodson law refers to within-person differences in the subjective experience of arousal when dealing with a cognitive task, which is arguably different from between-person structural differences in neuroticism as typically studied.…”

“…The argument that neuroticism might be related non-linearly to cognitive performance was proposed as early as the 1960s (Eysenck & White, 1964;Lynn & Gordon, 1961). Using a student sample Lynn and Gordon (1961) observed a negative quadratic effect of trait neuroticism on performance in an intelligence test (Raven"s Progressive Matrices).…”