A Pilot Study of the Convergent and Discriminant Validity of the Dutch Version of the Parametric Go/No-Go Task.

Abstract

The parametric go/no-go (PGNG) task is a computerized task that is designed to measure cognitive flexibility, response inhibition, and working memory. The PGNG task has been shown to measure core executive functions (EFs) in a psychometrically sound, brief, and ecologically valid manner. To analyze the psychometric properties of the Dutch version of the PGNG task in a convenience sample of nonclinical adults. The sample consisted of 74 highly educated adults, with an average age of 36 years. Forty-two participants completed test battery A to investigate the task's convergent validity; 36 participants completed test battery B to investigate the task's discriminant validity. The results were analyzed using a repeated-measures ANOVA, Friedman's test, paired-samples t test, and correlation analyses. Level 3 of the PGNG task places increased demands on sustained attention, response inhibition, and set-shifting. Several moderate correlations between level 3 and a complex EFs measure supported the convergent validity of this level of the PGNG task. The convergent validity of levels 1 and 2 was not supported. No significant correlations were found between PGNG levels and non-EF tests, supporting discriminant validity. Our study included a rather homogenous sample of highly educated participants, which might explain the convergent validity of level 3 of the Dutch version of the PGNG task. Hence, to overcome these potentially confounding factors, the Dutch version of the PGNG task should be investigated in a larger and more heterogeneous population in terms of age and educational level.