Just a mouse? That mostly works if you can map all actions onto a single 3-button mouse. But don't use that third button and make sure that you can adjust your mouse sensitivity on-the-fly. That's my No. 1 reason for people to feel dizzy during exposure to VR environments.
Joysticks seem very intuitive, but might be too exhausting when dealing with patients. A hand-rest on the joysticks helps heaps where participants can rest their hand on while still using the device.
I tried a 3D mouse like the Space Explorer and I had my required degrees of freedom comfortably mapped onto the device and disabled everything else, but this will need a lot of fine-tuning. Since you can twist, turn and move this device, it's easy to mix up rotations and translations and you manually have to fiddle with thresholds and the input that such a device gives you. Scrap that!
Where to now? I'm gonna go the extra mile and fork out some cash for a computer brain interface. I'm currently looking at NeuroSky ad Emotiv. Emotiv seems to be sold out (or whatever their marketing strategy is called). Their advantage seems to be the more sophisticated setup for conducting EEG research with this baby. NeuroSky only uses one dry sensor which might limit it's capabilities for getting meaningful EEG data out of it. On the other hand, the dry sensor is probably more comfortable to wear for participants. As long as it's good enough to have somebody walk through a virtual environment. In addition, NeuroSky already comes with the documentation and .dll to integrate it into Unity. That's a plus for me.
If anyone has any experience with the computer brain interfaces, please get in touch with me. I'd be happy to have a talk about these and hear about some hands-on experience with such devices. If know of any other products, drop me a line. Cheers!
Lastly, I'll most likely be getting a tablet for my line bisection experiments. I was looking at the Wacom Intuos4, but the Bamboo Fun seems tempting with its touch capabilities. I wish it also had the better resolution of the Intuos which is critical for my experiments. Tough choice!