Well ya got a central nervous system senior research scientist (chemist) in the crowd somewhere around here, I don't know who he is, but he's a bit forgetfull these days. (he's a bit daft and has a difficult time doing things like erasing his own posts, because his fingers keep hitting the wrong buttons when he's all dyskinetic ).
I don't think one has to get quite into the explanations for DEX as has been stated. DEX is the mirror image of LEVOMORPHAN, which is a narcotic mu agonist.

"Dextromethorphan crosses the blood-brain barrier, and the following pharmacological actions have been reported:

NMDA glutamatergic receptor antagonist
Dopamine reuptake inhibitor[3]
σ1 and σ2 receptor agonist (Zhou & Musacchio, 1991)
α3β4 nicotinic receptor antagonist[4]
Serotonin reuptake inhibitor[5]"

Now this gets back to a discussion that we had awhile back where, I believe it was fiona who asked something about low dose Naltrexone (which is a dopamine antagonist). During this discussion, this medicinal chemist (wish I could remember his name! ) started a line about his believing that narcotic agonists, and antagonists bear a superficial resemblance to Apomorphine, where, if you look at that molecules structure has the dopamine structure "buried" in what we would call a lipophilic molecule, and lipophilic molecules are often brain penetrable, so they have "CNS effects". Nearly all narcotics of the "phenanthrene" class have the dopamine structure embedded within them, and have "extra structural details" that allow the molecule increased lipophilicity ("fat loving", increased solubility in fatty tissue) which is a property involved in passive blood brain barrier penetration of a molecule, since the brain is literally all fat no muscle.
So, these types of molecules get into the brain, wheras dopamine itself can't. The fact that the dopamine structure is preserved in the larger molecule, is unquestionable to say that these molecules could have dopaminergic effects. I noticed it from Oxycontin. Therapeutic doses allowed me to cut my sinemet intake by half. And now, using Apomorphine, a molecule devoid of narcotic properties, but still retaining the dopamine structure within the molecule, and the most potent dopamine agonist known; the user suggests that in the phenanthrene like molecular structure, there lies the next best chance at creating the next "best D agonist", one with no emetic effects, a long half life, orally active, easy on the liver enzymes, and most of all, being a blockbuster antiparkinson drug.
HOwever, we're not there yet and we don't even know if anybody is working on it; most of the pharma companies bowing to the supremecy of Dopa. Many pharm companies that had active D agonist projects just stopped pouring money into this area. IN fact, CNS D agonism is considered a liability as far as receptor binding profiles of new "ethical" drug discovery.
In a nutty shell, Dextromethorphan may be a simple weak D agonist, a molecule whose properties as far as they relate to an antiparkinsons drug which have not been optimized.
That's how an old school pharmaceutical discovery science would at first look at the reported phenomenon that is reported for this molecule. Femtomolar uncharacterized enzyme inhibitory effects is a good reason not to pour money into it's further investigation, but relate it to apomorphine's receptor binding profile, and dopamine structure similarities itself and you might get a few "experts" to give it some more thought. THat would be nice, but will it, or is it happening? Your guesses are as good as this guy I seem to know, what's his name, damn, I forget!
By the way, will somebody just open the door and let us out of here!