Yesterday I spoke about the dragons in my DF / DFRPG megadungeon game, Felltower.
I mentioned, briefly, that the danger of dragon fights and the concern of the players that they don't really know how to kill them means that they've avoided deliberate dragon encounters. They've had conversations about "Is this technically a dragon?" about other monsters, in case they might mistakenly fight a dragon and get stuck in a dangerous fight with a monster they don't know how to kill. It's a lethal game and guessing wrong can mean making up a new character.
In a way, this really comes down to a GM tip - whatever extra importance you openly apply to something, your players will potentially magnify it many times. To put it another way, if you make something out to be a big deal, it's likely your players will think it's a very big deal.
Dragons in my current game, as above, are an example.
Dragons in my previous GURPS game are another.
I didn't want dragons to just be a casual encounter in that game. I subscribed to this theory of dragons. They were big deals. Huge. Named NPCs. Seen but not encountered unless you went after them. Fabulously wealthy, but also powerful and dangerous. They could be hunted and slain by only the boldest and best equipped, or negotiated with and spoke to in the hopes of acquiring their pent-up knowledge of the ages.
So naturally, dragons were a huge aspect of my game, right?
No, of course not. One dragon was seen flying. The group (wisely) hid from it, as it was early in the campaign and they weren't ready.
The rest of the campaign?
No dragons. Not one was investigated. Not one was spoken to. Not one was seen. Not one was researched to see where it lived, even when names came up. Not one was attacked. No one sat around the table and said, "We should hunt down one of the dragons and kill it, we'll have all the money we need for the rest of the game!"
When dragons did come up, it was quickly agreed that the group wasn't ready, that should be tabled for another time.
Any why not? I'd said they were dangerous, important, wealthy, knowledgeable, lethal, and so on. Who looks at their character sheet and says, "Well, I'm ready for the toughest possible monster in the game"? Not that many players - not a majority, anyway. It's always better - for in-game and out-of-game reasons - to put off danger you don't understand or can't be sure of and solve the easily solvable.
I made an effort to have dragons front-and-center in my current game. There are dragons you can't just march up to and kill - ones that are potentially shatteringly powerful and absurdly wealthy.
But equally, some dragons just sit in a major dungeon entrance on a pile of loot, and can be slain in a single encounter by PCs not entirely ready for a dragon fight.
Even so, we've had one accidental and zero purposeful encounters with dragons since. That's even with a party member, often acting as a party leader, who is obsessed with fighting and killing them. He's been unable to make the case for a dragon hunt, and those who don't want to go are careful to keep him away from a potential fight so he won't rush off and start one.
Another good example are the gates in Felltower. I put them down so they'd be ways to get to many different kinds of adventure or locations they just didn't fit within my megadungeon. So, magical doorway to them. My PCs finally went through one, only out of terror at something else and a series of mistakes that boxed them in near the gate with no other way out. Otherwise, gates are always "We aren't ready, we don't know enough, we might not be able to find our way back." My own logical setup to avoid "toe in the water" gaming - you can't always just go in, look around, and then come back - has meant it's a big deal to step through a gate.
My players regard gates, then, as a really big deal. Because you can't test the waters, you have to commit, it's safer just not to commit because you don't know what you're getting into. A solution would be to allow for "toe in the water" go-and-return gates to be totally standard just like two-way doors, not one-way, are standard. But then we're right back to every gate just being a "go see, and we'll come back someday when we're prepared for certain victory." With an added dose of GM prep and loss of sense of wonder, since gates are no longer doorways to strange and wondrous adventure but just fancy doors to larger dungeon rooms.
In the game world, and real world, that has logic - don't have fights, perform executions; know your enemy like you know yourself; failure to plan is planning to fail; look before you leap. But the war stories we all teach years after the game really happen when you leap and then look.
It ends up being a clash of logic. On one hand, the logic of the GM seeking to make something important and different - dragons aren't just bags of HP slain by dealing enough injury to them in any fashion you choose, gates not simply being glorified doors you can casually walk through and back as you please. On the other, the proper caution of players who don't want their paper man dead or trapped in a different world today, even if it means a potentially less story-worthy session.
I do have three potential solutions, however, to keep the logic but encourage particular actions.
Increase the risk-to-reward
This can be an in-game reward (more treasure with dragons, behind gates, etc.) or a meta-game award (extra XP for killing dragons, extra XP for crossing gates) or a real-world reward (free beer for fighting dragons and going through gates!)
Making every single dragon fight a +1 XP in GURPS or DFRPG - or every gate count as "lots of exploration" for a 2 XP session in my rules - means there is a real basic motivation for giving it a go. You know it will pay off
Simply make more encounters with "important" monsters or passage through gates required. Have a dragon attack the PCs - like in Skyrim, where they'll just drop down from the sky and attack you. Make a gate the only way into or out of an area. Rocks fall, the big doors slam shut, everyone is stuck in the room with the open gate.
Make them common
Make these things more common. Again, Skyrim has you fight so many dragons you need the game to track the count. Gates could be all over the place, so avoiding them is like adventuring with a "we don't open doors" philosophy. You can keep these things special but common enough that you must encounter them - have a dragon nest on top of the dungeon after a while, have important bits of powerful artifacts or keys to great wealth known to be past a gate. It's still a choice to deal with them or not, but you're giving up a lot to avoid the perceived magnified risk.