I'm working on a serious game capabilities demo and am looking for some advice on ways to script something and the approach to take. I want to have some key decision points in the scenario where the player is asked questions and needs to select between a set of choices. Here's the current plan:

1. The player is in a driving game, where they can move a vehicle freely inside a virtual environment.

2. When they enter within a certain set of X,Y,Z coordinates, the game is paused and a GUISkin is used to show a graphic on a higher layer with buttons for choices.

3. Once the player makes their choices, the GUISkin changes to a new image based on their input to show if they made the right decision. Decisions are also kept as part of the player's profile to assess their training throughout the course of the game.

4. After the player clicks a button to continue, the GUISkin is removed and they are returned to live game play.

Well that's my thoughts, but there may be a better approach to take. Any advice or links to online resources would be much appreciated. I'm still learning my way around Unity 3D and I've never used sensing listeners in Unity 3D before, but have worked with similar scripting in the virtual world of Second Life with the LSL scripting language.

Also if I use GUISkins, would I just do this all within the same scene or is there an advantage to breaking things up among several scenes when you're running a virtual environment? I wanted to avoid loading the same assets multiple times. My current game demo is setup with 3 scenes (start, game, and end).