Slime mold is a very curious substance. Lab experiments have found that some types of the organisms exhibit signs of "intelligence" -- in the form of being able to learn and remember -- even though they're "no more than a bag of amoebae encased in a thin slime sheath," as slime mold expert John Tyler Bonner put it.
Ella Gale, who researches memristors and brain-like computers at the University of the West of England, has translated some of a particular slime mold's more interesting behaviors into human emotion -- as expressed by a robot face.
Physarum polycephalum can navigate a maze, find food with uncommon swiftness, and retreat from light. Placing the mold on a bed of 64 microelectrodes, along with some oat flakes, Gale shone light onto the fungus. As it moved toward the food or shied away from the light, it produced electrical signals, which Gale then converted into sound frequencies.
Gale and her team then split these sound frequencies into pieces, and linked them to a positive or negative emotion, depending on whether the mold was moving toward food (positive) or away from light (negative), along with an intensity based on the sound's volume.
A negative, high-intensity emotion might be anger, for example, while a positive, low-intensity emotion might be contentment. The researchers then had an expressive robot from Hanson Robotics act out the emotions while the soundtrack played. (If all this is a bit confusing, be sure to watch the video below.)
Gale is currently researching exactly how something so unlike a brain can learn like one, and is hoping this experiment might help her figure it out.