An icebreaker
A psychologist devised a simple test in 1970 to check whether an animal recognizes its own reflection, by marking it and watching if it touches the mark. Researchers now build similar self-models into robots and AI, though what either 'passing' actually proves stays disputed.
In 1970 he ran the first version of the test on chimpanzees, marking their faces and watching whether they touched the mark while looking in a mirror; he still maintains only great apes and humans pass it conclusively.
He argued that self-awareness likely comes in degrees rather than as a pass-or-fail line, and that a test built around vision can miss it in animals that rely on other senses.
Leading a robotics lab, he has spent over two decades building robots that learn 3D models of their own bodies from video of themselves, describing self-modeling as a primitive but genuine form of self-awareness in a machine.