Early on we learn that music has strong beats and weak beats (and medium beats too, oh my!). Knowing the placements of these beats allows us to recognise the metre of a piece when listening (and adjust for changing metres), and tells us where to move to in a bar. Even when we start disregarding multiple beats in a bar (in faster tempi) or breaking down beats into further subdivisions, it's all pretty straightforward. Move to the strong beat(s). I think even those without conducting experience can easily follow along to a piece of classical music and know where the downbeats belong (even if they don't know specifically it's a "downbeat").
Rock music seems to treat the idea of downbeat somewhat differently. The standard rock beat has a snare hit on beats two and four--are they accented off beats? Can a syncopation that lasts for an entire song--and in thousands of songs--still even be considered a syncopation? Or is it even an accent? It could be that the bass pedal we frequently hear on beat one (and three, though sometimes as two eighth notes or just on the second half of the beat) is interpreted as "strong" and the subsequent snare hit is then processed as "weak." After all, music is always about context. But I'm not convinced. After all, the bass drum is "softer," more subliminal, often blending with a note played on the bass guitar, whereas the snare is a sharp contrast: it's very noticeable. When a band gets the audience to clap along, we clap on the "offbeats" with the snare. I think this can be partly explained by the timbral quality of our claps, but isn't it also because we feel the music moving towards the second and fourth beats? We bob our heads on two and four. So if this is no longer a syncopation, does that mean our entire view of rock music has been shifted by one beat? This article, while not directly relating to these musings do make an interesting related point.
The snare hit also helps determine the overall apparent speed of the music. When there is a shift from two snare hits to four per bar, suddenly the music is galloping along, even if nothing else has changed. Where are the strong beats now? It's as if we're using hypermetres now, but also compressed the bars (or something). I'm not quite sure how to explain that one, but I have noticed that speeding up the snare affects the overall apparent tempo more than any other factor. Now we have four downbeats. Another peculiar effect occurs when there is a snare on every beat, but it has been shifted to the second half of the beat. This sounds perfectly normal while you're listening to it. But it has happened before that I've gotten in the car and the CD has resumed playing in one of these particular passages. Two things happen: first, I recognise that the snare beat in rock means "strong." Then I correct the measures to hear those snare hits as being on the first half of the beat. It's an extremely confusing experience because my brain is in conflict. It knows that the music sounds wrong because it knows the other things going on (vocals, keys, etc.) now sound displaced. But it has trouble reconciling the (actual) displacement of the snare hits. It has taken me up to a minute (or until the drumming pattern changes) for me to correct the rhythm.
This standard use of bass and snare drums to denote first/third and second/fourth beats obviously has its limitations. This will only work in common time, and shifting the snare to beats one and three would create a truly bizarre "syncopation." In shifting to triple metres, I've found the most common use of snare to be on every beat, at the beginning of every bar, or at the beginning of every other bar (effectively turning it into duple time subdivided into triplets). This is, of course a generalisation, and I'm talking about establishing a regular rhythmic pattern.
The uses of bass drum and snare drum/hi-hat in trance and dance/electronica is a completely different matter, however. One, perhaps better left for another time.