The strength of research conclusions must follow from the strength of evidence. In this commentary I raise several issues with a recent paper “Brain structure changes induced by attention bias modification training” (Abend et al. 2019). I follow the paper’s five highlights to discuss; the absence of discussion of negative ABM results, the low power of the study itself. Centrally, I raise the concern that the conclusions rest all their weight on a single statistically significant group x time interaction form brain-wide analysis. Most importantly, this test is not internally consistent following statcheck. The reported result requires checking, and potentially a correction that would shift the results entirely.