There’s an awesome analysis of this in Daniel J. Levitin’s book “This Is Your Brain On Music” If I remember correctly, basically, people actually like a balance of repetition and surprise. They’ve done studies on foetuses in the womb and found that from something like 30 weeks (I can’t remember precisely) they start to build their musical preferences. At a young age children are naturally more inclined towards simple, major harmony before more dissonant harmony begins to appeal to them – partly to do with “prosody” (patterns and intonation found in speech); for example, when we speak to children in an excited and happy way, we go high and major, whereas sad is low and minor, angry is loud and aggressive – all of these are the same descriptions in happy/sad/angry music.

Anyway, these familiarities and associations mean that people start to EXPECT certain conventions, BUT would get totally bored if there was no element of surprise. Levitin compares it to a journey: going from A to B on a long straight road is boring and forgettable, but taking a detour to see some sights still gets you from A to B (as EXPECTed), but with something to make it memorable. That said, if you didn’t actually get to B (removing EXPECTation) and just drove off into the woods for a few hours you’ve gone too far down the unexpected (like trying to play “Free Jazz” to someone who had only ever listened to mainstream pop music).

So, good music is a balance of the expected and the unexpected: hence why so many songs have something slightly different in the second verse or chorus: if it was an exact repetition the listener would lose interest, because there’s nothing unexpected but we still need some familiarity to make sure we get to the end of our journey.

I highly recommend that book if you’re interested in finding out more about the psychology of music – I haven’t really done it much justice in this answer!