Corrigendum: The Potential of Functional Near-Infrared Spectroscopy-Based Neurofeedback—A Systematic Review and Recommendations for Best Practice(Front. Neurosci., (2020), 14, (594), 10.3389/fnins.2020.00594)

Abstract

In the original review article, we miscalculated the statistical power and sensitivity of the included studies which led to an overestimation of power and sensitivity. The miscalculations arose from an easy-to-miss default option for repeated measures (mixed) ANOVAs in the statistical software program GPower (Faul et al., 2007), which co-author (RTT) became aware of after our review was published. The default option defines a variable in the effect size calculation (Ƞ2p) in such a way that the common usage of small, medium, and large effects sizes for the interaction of repeated measures (mixed) ANOVAs (f) do not hold true. If using the default option, the power calculations will account for the correlations between repeated measures a second time, and in turn erroneously increase power. In fact, the G*Power software itself recommends another option (“as in Cohen, 1988 – recommended”). We recalculated the statistical power and sensitivity of the included studies using the recommended effect-size estimation according to Cohen (1988), instead of the default option1. Furthermore, we noticed a few other errors in the sensitivity and power analyses which we corrected: