Inhibition, shifting and updating in relation to psychometric intelligence across ability groups in the psychiatric population

BACKGROUND

Assessment of intelligence and executive function (EF) is common in complex neuropsychiatric practice. Although previous studies have shown that EF and intelligence are related, it is unknown whether these constructs relate to one another in a similar manner across different ability groups (mild intellectual disability, borderline intellectual disability and normal/high intelligence). This study therefore examines the relation between three EFs (inhibition, shifting and updating) and intelligence in a heterogeneous psychiatric sample. It is hypothesised that the strength of the relation between intelligence and the three EFs decreases when the level of intelligence increases, in accordance with Spearman's Law of Diminishing Returns.

METHODS

In a cross-sectional, between and within subject design, one of the three intelligence tests (Kaufman Adolescent and Adult Intelligence Test and Wechsler Adult Intelligence Scale - third and fourth editions) and several EF tests (Stroop Colour-Word Test, Trail Making Test and Spatial Working Memory task) were administered to 250 neuropsychiatric inpatients and outpatients (Mage  = 39.8, standard deviation = 14.3, 52.8% male). Based upon their full-scale IQ score, patients were divided into three ability groups (mild intellectual disability, borderline intellectual disability or normal/high intelligence). The relation between EF and intelligence was assessed through analyses of the correlation pattern; groups were compared using analysis of covariance.

RESULTS

Analyses showed significant correlations between the constructs of EF and intelligence. A significant interaction effect was found for shifting, with highest correlations in the normal to high intelligence group, but not for inhibition and updating.

CONCLUSIONS

Results support a specific role for shifting in this EF-intelligence relation. The correlational pattern of updating and intelligence, as well as the differential relation of shifting and intelligence across ability groups, suggests that EF tasks may not measure distinct EFs in lower intellectual ability but rely on cognitive primitives such as processing speed. EF tasks can be considered less valid indicators of EF ability. Implications in terms of the need for development of specific tasks to measure cognition in low intellectual ability are discussed.

Predictive Value of Traditional Measures of Executive Function on Broad Abilities of the Cattell–Horn–Carroll Theory of Cognitive Abilities

The neuropsychological construct of executive functions (EFs), and the psychometric Cattell–Horn–Carroll (CHC) theory of cognitive abilities are both approaches that attempt to describe cognitive functioning. The coherence between EF and CHC abilities has been mainly studied using factor-analytical techniques. Through multivariate regression analysis, the current study now assesses the integration of these latent constructs in clinical assessment. The predictive power of six widely used executive tasks on five CHC measures (crystallized and fluid intelligence, visual processing, short-term memory, and processing speed) is examined. Results indicate that executive tasks—except for the Stroop and the Tower of London—predict overall performance on the intelligence tests. Differentiation in predicting performance between the CHC abilities is limited, due to a high shared variance between these abilities. It is concluded that executive processes such as planning and inhibition have a unique variance that is not well-represented in intelligence tests. Implications for the use of EF tests and operationalization of CHC measures in clinical practice are discussed.