Within-subject reliability, occasion specificity, and validity of fluctuations of the Stroop and go/no-go tasks in ecological momentary assessment.

Following the (revised) latent state-trait theory, the present study investigates the within-subject reliability, occasion specificity, common consistency, and construct validity of cognitive control measures in an intensive longitudinal design. These indices were calculated applying dynamic structural equation modeling while accounting for autoregressive effects and trait change. In two studies, participants completed two cognitive control tasks (Stroop and go/no-go) and answered questions about goal pursuit, self-control, executive functions, and situational aspects, multiple times per day. The sample (aged 18-30 years in both studies) consisted of 21 participants (14 female) in the pilot study and 70 participants (48 female) in the main study. Findings indicated poor within-subject reliability for the Stroop task error rate and reaction time difference between congruent and incongruent trials and moderate to good within-subject reliability for the go/no-go task error rate and reaction time. Occasion specificity-the systematic variance accounted for by state residuals-was at a modest level (between 1.4% and 11.1%) for the Stroop error rate and reaction time difference, and at a moderate level (between 16.1% and 37.2% for the go/no-go error rate and reaction time) in the two studies. Common consistency-the variance accounted for by latent trait variables-was at a moderate to high level for all of the investigated scores. Indicative of construct validity, the Stroop and go/no-go task error rates correlated positively with each other on the within- and between-subject level. Within-subject correlations between task scores and subjective self-control measures were very small and mostly nonsignificant.