>>34088902
No, Duolicious doesn't store your account history so I can't run that analysis. But if the reason there's no difference is the hypothesis I gave in my previous post, the results would be the same. i.e. the historical analysis would show no difference between verified and unverified users.

One way to test the hypothesis would be to get an AI model to assign attractiveness scores to images and see how reply rate correlates with attractiveness for verified and unverified users. Although I'm pretty sure the results would be very predictable, so I'm not going to bother doing that.