 This research evaluated the reliability of measures used in brain science and neuroengineering studies. It found that measures at the sensor level had higher average interclass correlation coefficients, ICCs, compared to those at the source level, except for directed between region measures. Additionally, single region measures had higher average ICCs compared to between region measures. Furthermore, nodal metrics displayed highly varying ICCs across regions, while global metrics varied according to nodal metrics. Overall, this study provides valuable insight into the reliability of measures over a long period of time, which can be useful when selecting measures for practical mental monitoring applications. This article was authored by Jonathan Harvey, Anastasios Biserianos, and June Wally.