Test-Retest Reliability and repeatability of Behavioral and Electrophysiological Markers in an Eriksen Flanker Task

Cognitive control processes, specifically interference control and error monitoring, are often impaired across neuropsychiatric disorders and have been proposed as transdiagnostic markers of psychopathology and important treatment targets. Accurately probing them, however, requires understanding the psychometric properties of the measures used to assess cognitive control, including their intra- and interindividual stability over time. Using an Eriksen Flanker Task, we tested behavioral and electrophysiological readouts of cognitive control in 36 healthy individuals (26 female, 10 male, M age{+/-}SD=33.18{+/-}14.49, range=19-68) and evaluated their test-retest reliability across 48 hours by calculating Pearson correlations and Intraclass Correlation Coefficients (ICCs) to assess group-level stability. Moreover, we assessed repeatability through Coefficients of Variation (CVs) and Bland-Altman statistics, to investigate the degree of change in participants absolute scores. We found satisfactory-to-excellent test-retest reliability for most cognitive control measures, with condition-specific metrics generally being more reliable than difference scores. Regarding repeatability, we observed considerable intraindividual variability in absolute scores over time, which differed widely between participants. These results demonstrate that measurements of cognitive control may display substantial intraindividual variability across sessions despite demonstrating high test-retest reliability and vice versa. Our findings expand the current literature by providing novel information about the stability of behavioral and physiological markers of cognitive control over time. Moreover, they may have important implications for the application and evaluation of clinical interventions by highlighting the usefulness of considering repeatability measures in addition to the more commonly reported test-retest reliability metrics, when tracking changes over time in clinically relevant processes within single individuals.