Using artificial intelligence Samani has constructed a system that simulates the physiology of a human in love. Hormones drive most of what we feel when we are in love. Infatuation, deep love and even orgasms are influenced by our levels of oxytocin, dopamine, serotonin and endorphins. Samani has created an artificial endocrine system that mirrors our own, and he's developed quite a compelling product. (See researchers interacting with a little round R2-D2-like robot in the video below.)
These robots exhibit six emotions—happy, sad, fear, surprise, disgust, anger—manifested in different movements, flashing LED lights, vibrations, etc. In this video watch as the needy little thing displays dangerous jealousy. And while you're watching pay attention to your own reaction.
Samani stresses that the robots will have to be in active communication with the human, adjusting its "lovotic state" to inputs and feedback from the human. So a great deal of psychological research is required in building such a system.
From his overview: A wealth of information about a person’s emotions and state of mind can be drawn from facial expressions, voice, blood pressure, temperature, gesture, etc. The affective system of the robot would need to analyze system inputs to generate suitable states and behaviors for the robot in real-time. The affective system should be modeled as closely to the human being as possible in order to be an emotionally engaging system.
Curiously, Samani found that while 36% of surveyed people said they could imagine a robot loving them, only 19% said they could love a robot. Perhaps we underestimate our ability to adapt and anthropomorphize. After all, as soon as an object starts behaving like us, we start projecting. Just watch a three-year old cuddle a Baby Alive doll. And that thing only eats and defecates. Hasbro's tag line says it all: "[Baby Alive] I love the way you make me feel, you're so real."
Imagine how eerily easy it might be to attach to a more challenging and subtle object.