The way I would approach it (and I could be wrong) would be to find out where each side of each switch goes; depending on how it it wired (and I suppose it could be wired very weird, but...), one side of each switch will go to the positive terminal of the battery (or to a voltage source/regulator on-board the PCB in the transmitter, which may be lower than the battery voltage), or to the negative terminal (common ground).
Hopefully, one side goes to common ground; it makes things easier to do (because the components are more available). If this is the case, what you want to do is hook up an NPN switching transistor between the side going to ground and the other side, so that the transistor becomes the switch (the collector goes to the side of the switch not connected to ground, the emitter goes to the other side, or ground); a common 2N2222 NPN should work OK for this task, since the current flowing should be minimal. You will want a bias resistor on the base of the transistor as well; see here for a method on how to calculate such a base resistor value:
You can also try starting with a 1k ohm, sometimes you can get away with lower values (there's a way to calculate the proper base resistor, I am sure there's a calculator somewhere on the internet, actually), but don't go below 330 ohms. The other side of this resistor you hook up to a digital pin on the Arduino, where you control it as normal.
If you are really paranoid about your Arduino, you could hook up an opto-isolator between the two (or you could try to find an opto-isolator with an NPN detector that has a high enough current rating to handle the switching; anything over 25 ma should do - try to get around 50-100 ma to be "safe".
Now - if the switch control on the positive side, its basically the same kind of scheme, except with a PNP transistor...
Something else you could try would be to wire the Arduino to the receiver, so that you can directly control the h-bridges/switches that control the truck. You would have to do a bit of reverse engineering, and maybe some probing with a multi-meter after some schematic drawing and reverse engineering (taking guesses at components, diagraming the schematic wiring of the receiver PCB) - it isn't easy, but it can be done (sometimes its easier to trace back from the motors to the PCB)...[/edit]