Although you can measure the effects of subliminal stimuli (e.g. pictures not seen consciously, words not heard consciously) the effects are always statistical and small. This means that if you show a cinema audience of 1000 people, say, a 20 ms exposure of a Mars Bar and then measure how many more Mars Bars are sold in the foyer 10 minutes after the film you might find a 1% increase in Mar Bar Sales. (this experiment hasn't been done, but it sounds plausible to me). But two points: Firstly, these people were going to buy confectionary after the film anyway - you already had people ready to spend money on chocolate. The subliminal picture probably just made the idea of 'Mars Bar' come to mind for some people a couple of seconds earlier as they stood wondering about what selection to make. If you'd had shown a picture of a oysters, i don't think anyone would have been at the popcorn-stand demanding to buy edible molluscs. Secondly, the effect of subliminal messages (in this case mars bar sales) is usually less that the effect of superliminal messages. In other words if you had told everyone "Mars Bars taste great, buy a mars bar when you leave the cinema!!" you would have got many more sales.

So that is the weak (second point) and simple (first point) of the section title. You can see that not only are subliminal messages weak in size, they have to operate on something someone is already doing, or has the mental structures for doing already. Another thing is that research shows that if you tell someone about the expected effect of something subliminal the effect is cancelled out - people use conscious deliberation to monitor their behaviour and the automatic processes that are guiding them. (On a side note i think there is probably a deep relation to the way hypnotism works here).

You asked specifically about subliminal messages in music. There was a case in the eighties where a heavy metal band got sued for (it was alleged) putting subliminal suicidal messages in their songs. Nonsense of course. There was no voice - people who thought they could hear something had just been fooled by the electronic voice phenomena (i.e. wishful thinking). Anyway the band got sued and there was a court investigation into the reality of subliminal messages in music. You may be interested to get hold of it and read it. Conclusion: not possible.

The idea was that Judas Priest (they were the band i think) had inserted the words "Do it" backwards into the song, and that this had convinced two fans to kill themselves. Apart from the idiocy of supposing that a hidden message would overcome a normal conscious aversion to suicide and posssess people, there is no reason to believe that backwards messages (especially hidden backwards messages) can be automatically decoded by the brain.

So the probable reason that some professors you mailed said that this was not a topic of research is that all the research on subliminal messages shows that hidden messages in songs don't work - it would be no easier than getting someone to do something by putting the message in plain view in the lyrics. And millions of people listen to heavy metal lyrics without following the instructions therein (my favourite album when i was younger was called "Kill 'Em All" and i seemed to have survived without embarking on a spree of indiscriminate slaughter). Have you heard to remark that if you play a heavy metal album backwards, you hear a satanic message - backwards!

Does that help answer your questions? Sad to say the reality is less exciting that what rumour and stories seem to say. The interesting thing for me is what how being able to subliminally (and superliminally) prompt people's automatic behaviours says about the nature of our minds and what this means for our theories of individual free will, the self and the responsibility of people like advertisers and film-makers. But this is not specifically about subliminal images - it's about all input our brain takes in. Subliminal messages have no special power - no more than messages we do see but don't think about (like a lot of advertising). But that's another discussion.