A wise man once said, "In order for us to understand the rat, we must become the rat." Now, it's quite possible that no one has ever actually said this, but that doesn't matter. But what does matter is that it's the driving sentiment behind a project that's seeking to bridge the gap that's separating us humans from rodents. By using telepresence, immersive virtual reality, and robotic technologies, researchers are hoping to see things from a rat's point of view.
The technique, which was developed by Mandayam Srinivasan and computer scientists working at UCL and the University of Barcelona, call it "beaming." It works by having two different subjects, one human and one rodent, interact with each other through a virtual reality interface.
The experiment worked like this:
Human participants were asked to put on a head mounted VR display and interact with a virtual human that appeared on their screen. But this human wasn't a human at all - it was a digital representation of a rat.
The rat itself, which was 12 km (7.5 miles) away, had its movements tracked, and this tracking data was transmitted over the internet to the computers running the VR simulation. So, when the rat moved, the virtual human moved accordingly. In other words, Srinivasan had scaled the rat up in size, allowing human participants to interact with it on a more human scale.
As for the rat, it was placed in a small room alongside a robot (which regrettably looked nothing like a rat; the feeding dispenser sticking out from it certainly didn't help). This robot would scurry around the room in accordance to the movements of the human far away. So similarly, the rat was now able to interact with a virtual human on a rat-like scale.
Once everyone was hooked up and ready to go, Srinivasan asked the human participants to play two rounds of a five-minute game. The game was specifically designed to encourage the two subjects to approach one another.
Interestingly, in one of the games, humans were told that they were interacting with another human (when in reality the avatar was being controlled by the rat's movements). The researchers were hoping to see if human reactions to avatars changed if they thought humans were involved.
What Srinivasan and his colleagues discovered was that the human-controlled rat-sized robot did not bring about big changes in rat behavior (sorry for the downer). If anything, the rat tended to ignore the robot and hug the walls in a possible attempt to get out (gee, I wonder why).
But this isn't necessarily a bad result. First, it was a good proof of concept; the technology certainly worked. And second, it could allow for more intimate forms of research, observation, and interaction at different scales and perspectives. The fact that the robot didn't really affect the rat's behavior is a good thing - it didn't introduce too many external factors. Working at this level, human researchers could gain insights about animal behavior they never considered before. You can check out the entire study at PLOS.
Images via UCL.