I'm not well versed but I can give it a shot
As you've probably learned in physics, the E field in a conductor is zero. The E field at the interface between the air outside and the closed conductor has no tangential component (at least for the electrostatic case, don't know about the electrodynamic case). You can see that the tangential E fields are zero using the fact that the potential around a loop is zero for the electrostatic case. Anyway, the conductor re-arranges itself to negate the field introduced by any charges. If you wanted to analyze the field in the air region outside the conductor, you can place a mirror charge 'inside' the conductor with opposite polarity and that will reflect the field you see outside. To think of it another way, if you place a positive charge outside a conductor, the electrons in the conductor will shift towards the positive charge (leaving the side they left more positive) and effectively negate the field inside the conductor. However, since the side near the place where the positive charge is introduced now has more electrons next to it, the field is actually greater outside than it was before (between the positive charge and the surface of the conductor).
In regards to placing the mirror charge, it's usually more than one charge. The idea is to place as many charges as you have to to maintain zero tangential E field around the interface (in your case, a sphere). I imagine the arrangement can get complicated, but I think the other way of thinking about it should be enough.