I don’t know about you, but I’m looking forward to the day I will have a friendly, even loving, android in my life.
We human beings are deeply programmed for friendliness. Consequently, it’s surprisingly easy to trigger our perceptual system to perceive friendliness. Years ago, developmental psychologists discovered that infants would smile if the researcher held up in front of their face a piece of cardboard that had primitive circles representing eyes. Much like the geese of Konrad Lorenz who imprinted on him as their parent, our brains early on are tuned to search out that caring mother who will sustain life. We are tuned not just for friendliness but, more dramatically, for love. Recent findings in neuroscience tend to substantiate the clinical findings of attachment theory—our brains are primed for bonding, ever on the lookout for nurturing social soil.
Of course, we are also fearful creatures, programmed just as deeply for flight or fight; fear of the stranger; suspicion of the other.
Those who are in the process of manufacturing the androids/robots of the future need to move carefully to evoke the right emotional response from us. Undoubtedly, psychology will have a role to play in helping roboticists to model the personalities of their creations.
Science fiction has served as an early laboratory to think through many of the issues that will confront us, as androids and robots become an ever larger part of our world. The reflexive first question is “friend or foe?” Of course, our ambivalent imaginings have conjured up vivid portrayals of both possibilities in science fiction, and recent movies often portray our deepest trepidations and our most optimistic hopes.
On the fearful, threatening/ambivalent side, I’m thinking of science fiction movies such as:
- The Day The Earth Stood Still
- The Terminator
- Ex Machina
- The Stepford Wives
- I, Robot
- Blade Runner
- Get Out
On the friendly android/artificial intelligence side, I’m thinking of:
- Star Wars
Clearly these two lists are not exhaustive; in fact it was quite challenging to separate them into bad robot/good robot categories, inasmuch as there were sometimes both sorts of entities in some of the films listed above. For example, in Ex Machina (2015), the beautiful, brilliant, seductive android (Ava) turns killer and, as I recall, I, Robot (2004) features an ultimately helpful self-aware robot along with an army of robots who rebel against their servitude. The Star Wars series touches our hearts with the lovable R2D2 and C3PO pals along with weaponized robotic battle droids, among other automated baddies. I may have been stretching a point by including Get Out (2017) in the list, although the servants, Walter and Georgina, seem strangely robotic and it turns out their brains have been replaced by some very negative AI mojo.
The most recent of these films I’ve seen is Passengers (2016) from last year. For me, the highlight of the film was the android bartender, Arthur (played by Michael Sheen). His upper body is fully humanoid, but then we see his lower body is totally mechanical and shuttles back and forth along the length of the bar on some sort of trolley. His artificial intelligence (AI) makes him the perfect bartender, able to chat about a wide variety of topics while also being the ultimate patient listener—the quintessential quality for a bartender.
I just had to include Her (2013) on the list above, even though Samantha is not an android but rather an AI, whom we know only via her voice. But her voice and intelligence are so engaging as to make her feel tangibly present. Along with the Joaquin Phoenix character, I found myself terribly smitten by Samantha. Of course, she also portrays our ambivalence about AIs—she is Siri on steroids. And like Siri, she’s awfully promiscuous: they’re not just for us but for millions of others too. Samantha ultimately abandons the lovesick Joaquin Phoenix. So is she good or bad?
That beautiful voice of Samantha in Her should inform the psychology of future friendly robots. I think the android/AI builders already know this. Is it an accident that Apple’s Siri, Amazon’s Alexa, and Microsoft’s Cortana all feature female voices? Of course, a variety of voices are available for these assistants, but I think it’s not random that the female voice is the default. As a clinical psychologist, I have to believe that it probably speaks to deep structures in our brain that have imprinted on the maternal voice, laying down deep associations with safety, comfort, and nurturance. I must confess that my wife’s voice was one of the key drivers in my attraction to her, and it continues to be so to this day. By the same token, there are other voices that repel me right out of the gate. I’m sure there must be plenty of research literature in psychology on the effects of vocal characteristics. Going forward, I believe AI designers will need to pay close attention to voice as a key ingredient in the making of friendly androids.
I would think that Dr. Paul Ekman’s research on facial expressiveness will have an important role to play in the design of pleasing android faces. Ekman is an American psychologist who is a pioneer in the study of emotions and their relation to facial expressions. Ekman’s ground-breaking work began in Papua New Guinea in 1967 with his study of the nonverbal behavior of the Fore people, and recently he has begun to translate his research on how facial expressions show how we feel (and what we may be hiding) into practical applications, including the Atlas of Emotions (http://atlasofemotions.org/), which he created with the Dalai Lama. He was named one of the top 100 most influential people by TIME Magazine in 2014, and ranked 15th amongst the most influential psychologists of the 21st century (see http://www.paulekman.com/). Because he has identified specific universal facial expressions for communicating emotion, it will be possible to design androids whose facial expressions will be understood by people all over the world. Back in 2009, I was fortunate to be able to interview Dr. Ekman on my Wise Counsel Podcast
Body postures also have communicative power, and even a relatively small postural vocabulary can give the illusion of a live animal presence. In this respect, Sony’s robotic Aibo dogs (http://www.sony-aibo.com/) come to mind. Unfortunately, these mechanized pets are no longer on the market. Among other characteristics, they were able to cock their head to the side, as if curious about something. Sony created a whole line of these robotic companions. The first mass offering was in 1999. The response was overwhelming and 3,000 robotic pet dogs were sold in under 20 minutes in Japan along with another 2,000 over four days in the US. These robotic dogs not only responded to external stimuli, but were able to learn and express themselves. Each unit was designed to develop a unique personality including behavior shaped by the praise and scolding of its owner. And Aibo (short for “artificially intelligent robot”) quickly became a hit, especially in Japan. In fact, they became so popular among Japanese elders that they staged funerals when they could no long find replacement parts. Clearly, these robotic pets had for many Japanese seniors come to be seen as family members! One such person in Japan was quoted as saying, “So we think that, somehow, they really have souls.”
The sorts of factors I’ve been highlighting here, such as voice, facial expression, postural cues, conversational abilities, and so on to create non-threatening androids are already being implemented. They have already evolved far beyond Sony’s Aibo. The prime examples, the current state of the art, are coming out of Hanson Robotics. The founder and CEO, David Hanson Jr., is an award-winning robotics designer and researcher who has created a series of humanoid robots. Their robot faces look and act like yours and have the normal range of human expressions. His Sophia android appeared on the popular U.S. talk show, Late Night with Jimmy Fallon (https://www.youtube.com/watch?v=Bg_tJvCA8zw), and here’s a link to not only Sophia but some of Hanson’s other humanoid androids—click on the photo of each of the character robots in the gallery, including Professor Albert Einstein, and you’ll see them in action (http://www.hansonrobotics.com/robot-gallery/). There are many other YouTube videos showing the advances in humanoid androids. You might also enjoy David Hanson’s 2009 TED talk on robots that show emotion (https://www.ted.com/talks/david_hanson_robots_that_relate_to_you).
I’m not sure if David Hanson’s group includes one psychologist, or several, or if they have consulted the psychological literature. I wouldn’t be surprised if they had. Certainly, at some point, I think robot developers will be driven to do this as they seek to improve their creations to make them even more humanoid. I sent Hanson Robotics an email inquiring about this but have not heard back.
In the meantime, I look forward to welcoming an android friend into my life. How about you?