'patio11 makes great points, but his advice is mostly actionable for desktop software and web applications, which are a completely different beast than mobile apps.
Mobile applications won't sell at high prices unless they have a real core competency that solves a real need that doesn't exist yet - indie games and to-do list apps have little hope of achieving that kind of benefit.
Aside from that, there is such a low barrier to entry that they need that kind of incredible edge over the competition to survive at high prices.
Take a look at the high-priced apps and you'll notice a trend - they're typically not games/entertainment/weather/"fun" apps. They're real tools that can be used in production or they're productivity suites.