i've known of someone that made dmmda-2 from it's hyrdrogeniodide adduct by ammonolysis with excellent results.
i think it's a handy route to all of the essential amphetamines whose double bond is allylic and not internal.
trisubstituted (trimethoxylated) phenylpropenes have been known to form a stable carbobcation during the pinacol rearrangement process leading to a tetrahydrofuran type dimer the reference eludes me but the ring activation does stabilize the benzylic carbocation leading to this unwanted by product in short probably no.
this reference is a tetrahedron letters article circa 2000
you know what there are articles where they just take the crude methanol extract of nutmeg and brominate that then ammonolysis to get around 50% mmda i'll do that sometime for shits and giigles and tell you how it went.
i'm a bit curious about mmda