One critique of three-act structure is that the idea is self-perpetuating. Because there’s an industry of books about the subject, aspiring writers are likely to format their work in this particular way, even if it isn’t otherwise advantageous. George Lucas has been open about the fact that he copied the monomyth from Joseph Campbell’s Hero With a Thousand Faces for Star Wars. And when that film made a lot of money, you could bet that studios were interested in repeating every element about it that could work.
A studio’s readers, who decide which scripts to turn into movies, might also be brainwashed by the idea of three-act structure, and thus be more inclined to select screenplays which conform. Aspiring writers aware of all this will also be less expected to stray from the usual format. The most dedicated wannabes, the ones who have done the research and are thus more likely to succeed, will also be encouraged to go with it.
Robert McKee charges $785 for anyone interested in attending his seminar. That includes many professional and aspiring writers. In the last year, he has addressed sold-out crowds in New York, Los Angeles, Montreal, London and Seoul. He has spoken to tens of thousands of wannabe screenwriters, all of whom have chosen to invest in improving their craft. Some will go on to write screenplays good enough to be made into films, so McKee might simply get credit for their success, even if it was going to happen anyway. They might do well writing three-act screenplays, but it’s possible that they would have done just as well without that.
I have a small amount of insider information here. Several years ago, I took a brief class on writing for comic books taught by then-Marvel Comics editor Andy Schmidt. He was a big fan of McKee, and noted that the company paid all of its editors and writers to go to McKee’s seminar. Since then, Marvel has branched into film, producing adaptations of their characters. And those movies tend to abide by three-act structure.
So there is certainly something to the idea that people are rewarded for following the structure. Though it’s unclear if that’s because the people in charge are brainwashed by the feedback loop, or because the experts have come to the conclusion. after a tremendous amount of research, that this particular story form works especially well.
What Would Spidey Do? 
As a film student and amateur script writer, I’ve gotta say that the three act structure works wonders for telling a story that could otherwise be a mess. My first feature film script was a re-write of somebody else’s script, and by applying the basic tenets of the three act structure to their draft I was able to give it more balance and a bit more of an emotional path. For sure there were still plenty of areas for improvement, but the emotional beats were all in place for two hours of entertainment.
As an alternative, Todorov’s five stages of narrative effectively lays out a five-act structure, but it effectively follows the same overall path as a three act structure. As to why humans are hard-wired to respond to this type of story, I don’t really know…
the 3 act structure is a very logical, simple and elegant form. thesis. antithesis. synthesis. as old as humanity probably. nevertheless, I think we will see more variety in mainstream film structure as time goes by. because even if storytelling is nothing new in human culture, storytelling for film… is.
think of music. in the 18th century any symphony had to start with a sonata form, that was pretty much mandatory, no exceptions allowed. and then you had a slow part, usually a theme and variations, and then a minuet or, later, scherzo, and finally another sonata or rondo. structure has been quite rigid, in music, all throughout the baroque and classical eras (although it did evolve/changed), but then 19th century came… and form started to loosen. and it completely exploded in the 20th century. nowadays people still write things they call ‘symphonies’… but they are only titled as such, their form has usually nothing to do with what a ‘symphony’ originally used to be.
so similarly, I presume we are just going through a period of strict-ish-adherence to form. I think we need to become seriously bored with the recipe (which is kind of happening already) in order to move on to new forms. like… why not a 2-act structure? or a 7-act structure… act 1 and 7 are huge… 2 and 6 smaller… 3 and 5 smaller, and 4 really tiny, umm.. and 5 6 7 are mirroring 1 2 3… whatever, I’m just improvising. my point is: in order to reach some solid understanding of the real functionality of a new structure, time will have to pass and a lot of experimenting to be done. otherwise, it’s just a one-hit, an accidental experiment that doesn’t affect mainstream thinking. again, by judging what happened to structure in music, I have to say that real, functional, useful new structure is something that evolves, gains acceptance, and spreads slowly in time.