While the fluffy, seal-like PARO robot or the big-eyed, cartoon-figurish Pepper makes people smile, almost real-looking humanoid creatures, such as Sophia, horrify most individuals. Especially when she says she wants to destroy humanity. What’s behind the remarkably different perceptions of robots? There is a term called uncanny valley determining the aesthetics of robots, and it will have a significant role in shaping medical robotics, too.

Robot aesthetics – expert level

Functionality, potential users or the eventual place of use determine in many respects how a machinic creature should look. However, those are far from decisive factors when it comes to figuring out the actual shape of a robot. Should it look like R2-D2 from Star Wars, HAL 9000 from 2001: A Space Odyssey, KITT from the Knight Rider or The Terminator himself?

The range of robotic design is incredibly broad with the endpoints resembling R2-D2 as well as androids from the sci-fi series Westworld. The advancement of robotics comes with an array of mechanic creatures ranging from metallic animal-shaped skeletons through humanoid robotic companions to highly human-like androids. The latter ones specifically aim to mimic humans in form and presentation. They not only look recognizably human but also possess several human traits and behaviors.

uncanny valley
Robotic Drummer designed by Leonardo da Vinci at the beginning of the 1500s. Source: www.waikatomuseum.co.nz

Common sense would tell us that with the advancement of technology, androids would become more commonplace – why would you build less realistic creatures when you can construct highly life-like robots, such as Erica? Why would you go for a toylike machine as a social companion robot? Aren’t more human-like creatures able to invoke a better sense of communion and companionship because of the resemblance?

Androids terrify people

It seems that the response is an unambiguous no. The reception for androids, such as Erica, Sophia or Tara, who were all able to mimic human speech, facial expressions or movement, but couldn’t come close to perfectly resembling people, was the feelings of shock, panic with the mixture of awe for the technology. One of the creepiest robots is, for example, Showa Hanako 2, designed for helping dental students practice on her teeth.

No wonder that the daVinci surgical robot, which aids surgeons during operations, looks like an industrial machine arm – instead of Sophia. The Xenex robot, which disinfects a hospital room in minutes, looks like R2-D2 and not like Ava from the movie Ex Machina. Jibo, Pepper, Paro and Buddy, all social companions easing loneliness or treating mental health issues, do rather resemble Wall-E than a human. It seems that – especially in healthcare – there is a trend for non-humanistic robots instead of androids. And what’s the reason for it?

Arriving at uncanny valley

The explanation for the preference of humans for “robotic robots” instead of highly realistic androids is to be found in the aesthetic concept of “uncanny valley” and humans deep-rooted instincts and preferences. The phenomenon was identified in 1970 by a Japanese roboticist Masahito Mori, who noticed that people presented with robots of increasingly realistic human likeness respond with growing empathy, right up to the point where the similarities are almost real. At that point, people are repulsed. The sudden dip in the graph below describing their response gave the phenomenon its name.

Plenty of theories have sprung up in the next decades what causes the sudden abhorrence. Some believe that your brain cannot cope with the cognitive dissonance. When you see a robotic robot, you categorize it as a “robot”, when you look at an android, your brain actually sees a human, but if some details don’t fit the frames of being and acting like a human, it gets terrifying.

Some explain it with the human instinct for avoiding pathogenic creatures either as mates or as possible disease carriers; some say androids switch on humans’ innate fear of death or people start to unconsciously see them as violators of human norms and the meaning of personal identity. One thing is clear. It is deeply rooted.

uncanny valley

So profoundly that apparently not only humans but great apes are also terrified. To test their preference, researchers showed macaque monkeys real pictures, digital caricatures and realistic reconstructions of other monkey faces. To the latter, the macaques repeatedly averted their eyes.

Where do healthcare robots stand on the uncanny valley graph?

The uncanny valley graph shows that up until the point of “almost-real”, robotic robots are cute, likable, relatable and reliable. The ROWA dispenser robot with its metallic arms and robotic movement pattern has a low humanness score, but that’s not as low as in the case of surgical robots, which are rather thought of as simple mechanic arms and not entire creatures. The ROWA is higher on the likeability axis: pharmacists in Budapest even call one “Robi”, a cute nickname for Robert in Hungarian as well as for robot. The robust TUG robot and the streamlined RoboCourier, which make the in-hospital transport of medical devices, drugs, laboratory specimens or sensitive supplies easier, would score around the same values.

Social companion robots, such as the already mentioned Paro, Pepper, Jibo or Buddy would be considered even more likable – while their humanness score is fairly low. Riba or Robot for Interactive Body Assistance, as well as its Japanese version, the Robear shaped like a giant, gentle bear with a cartoonish head, would have the same scores.

The robot dogs of Atlas-Boston Dynamics fall into the same category. How do we know? When researchers and testers handle the little doggos harshly, we empathize with them dearly. Just look at the one below trying to open a door!

Deep down in the valley, creepy robots rule

On the other hand, for The Medical Futurist team, Sophia, Erica or the Actroid-F robot from the Japanese robotics firm, Kokoro, constitute all the features which toss androids deep down into the uncanny valley. It doesn’t matter that they have humanlike expressions, perfect eyes, eyebrows and smooth-looking skin; certain small details cannot be captured so perfectly as not to freak out the audience. Kokoro reportedly announced plans to sell 50 units to museums and hospitals to serve as receptionists or surveyors. We do wonder how that experiment will go down.

At the very moment, there are no robots at the other end of the valley. In science fiction, you have the androids of Westworld – played by human actors – who are totally human-like. Thus our brains do accept their existence. The same goes for the robots in the TV series Humans or David from Prometheus.

Be that as it may, there are robots which cannot be clustered. The origami robot, which can be swallowed and its capsule containing medication dissolves in the patient’s stomach and unfolds itself. It’s a very tiny, foldable machine. Another similarly uncategorizable robot is the blood-drawing machine. Veebot, a blood-drawing robot helps with finding a vein and taking a blood test in less than a minute. And what about exoskeletons? These robotic structures boost human performance and augment capabilities. However, would you call it a robot? Could you place it on the graph? We are not sure about that.

How to live together with robots?

The Medical Futurist believes that in healthcare, “robotic robots” will be more commonplace in the coming decades – at least up until the moment when androids arrive at the level of total humanness.

Looking beyond aesthetics, though, robots, in general, will become more of an everyday phenomenon. Perhaps we will not arrive at Hans Moravec’s, founder of Carnegie Mellon’s Robotics Institute, prediction that robots will emerge as their own species by 2040, but we have to get used to them. They will take over monotonous, repetitive tasks and re-shape how we think about plenty of professions. Although many people fear they will take their jobs (they won’t), they should rather prepare for a cooperative attitude to get the most out of the robot-human relationship. Especially if artificial intelligence will boost it.

Moreover, general respect and appropriate treatment should characterize humans’ handling robots. This determines our values and our morals – not the robots. That’s why it was hard to watch when the robot dog of Boston Dynamics gets kicked – even if it’s for an experiment.

Subscribe To The Medical Futurist℠ Newsletter

  • News shaping the future of healthcare
  • Advice on taking charge of your health
  • Reviews of the latest health technology