A particularly noteworthy example of a Gumbel-min prediction is shown here. The ROC predicted from g' (calculated from a single yes/no point) closely matches the ROC reconstruction derived independently from forced-choice judgments. The Gaussian model cannot even make a prediction in this case.
Comments
Are you familiar with this work by Adva Levi, Caren Rotello & Yonatan Goshen-Gottstein?
https://osf.io/preprints/osf/9m3sr_v1
They show da is superior to many other measures. It would be interesting to see how the two measures complement each other