Comparing the effects of different methodological decisions on the error-related negativity and its association with behaviour and gender

Abstract

The error-related negativity (ERN) is an event-related potential (ERP) component that is widely used to study human performance monitoring. However, substantial methodological differences exist across studies and it is unclear to what extent these differences may impact the reliability and replicability of observed effects. The current study used multiple common methodological approaches to ERN measurement on the same dataset in order to clarify the impact of these choices on the component's measured amplitude, psychometric properties, and association with individual differences, specifically behaviour and gender. In a sample of 263 adults, we quantified the ERN using different reference schemes (mastoid and average), baseline correction periods (−100 to 0, −200 to 0, and −500 to −300 ms), amplitude measures (mean, peak, and peak-to-peak), difference scores (subtraction and residual scores), and electrode site scorings (single-electrode and region of interest). This resulted in 72 distinct processing streams and estimates of the ERN. We found that data processing choices affect not just the measured amplitude of the ERN (range = −12.60–1.38 μV), but also measures of internal consistency (α range = 0.49–0.77) and test-retest reliability (r range = 0.40–0.71). Moreover, these different combinations of methods affected the strength of associations between the ERN and post-error slowing, as well as the magnitude and direction of gender effects on the ERN. Together, these results illustrate the importance of considering methodological influences on ERN measurement. Future studies comparing the effects of different methodological choices on ERPs and their psychometric properties are needed.