I think our samples must be different. Admittedly I have small samples of cocaine users, but the behaviors for them are consistent with the mainstream literature. Maybe just a bad sample, but it’s good enough for me to avoid it.