Dressed in cozy pajamas and armed with steaming cups of hot chocolate, millions of families flock to their televisions to watch “The Polar Express,” a film that has arguably become a holiday classic. While “The Polar Express” encourages the audience to believe in Christmas magic and draws upon childhood innocence, it is equally known for eliciting eerie feelings due to hyperrealistic computer animations. NPR’s Aliva Noë described the characters as having an expressive face with “zombie-like” eyes due to the limitations of the motion capture technology utilized. This strange feeling is known as the uncanny valley effect and has been studied by scientists and engineers for the past few decades.
The uncanny valley phenomenon now extends to various different fields, from animation to medicine, and may serve as a prediction of failure in human-technology relationships as more futuristic technologies are integrated into our daily lives.
In 1970, Professor Masahiro Mori, a robotics professor at the Tokyo Institute of Technology, mathematically described human appreciation for human-like robots as a function of human likeness that increases until a certain point where the feeling of appreciation suddenly turns into one of eeriness or repulsion. This area of the graph is analogous to a valley sloping downward, and was termed the uncanny valley. The uncanny valley phenomenon now extends to various different fields, from animation to medicine, and may serve as a prediction of failure in human-technology relationships as more futuristic technologies are integrated into our daily lives.
A recent study at the University of Cambridge provides strong evidence for the neurological origin of the uncanny valley phenomenon. In this study, 21 individuals were asked to rate the likability and human likeness of various groups (humans without physical impairments, humans with physical impairments, humanoid robots, synthetic humans, and android robots) in certain social situations. In keeping with the uncanny valley effect, the participants consistently ranked human or human-like groups higher and were monitored under fMRI to analyze the physiology behind this behavior. The University of Cambridge team found that the medial prefrontal cortex, a structure responsible for higher-order decision making, experienced greater activation during the ranking task. Surprisingly, two parts within the structure built upon one another to generate this response. One region translated human likeness into a “human is here” signal and elicited a strong response when presented with human groups. The other region, called the ventral medial prefrontal cortex, processed the “human is here” signal and turned it into a likability rating; this rating experienced decreases around the boundary between human and non-human groups and is a hallmark of the uncanny valley effect. This study provides a good launching-off point for understanding the neurobiology behind the uncanny valley effect and why the response exists in the first place.
With recent developments in robotic technology, robots are nearing human resemblance and intelligence.
While the University of Cambridge study provides a basis for the physiology behind the uncanny valley effect, several theories for why we feel this repulsion have been proposed by psychologists. For instance, the evolutionary aesthetics theory explains the uncanny valley effect as a product of human preferences for physical attributes indicating health, fertility and overall fitness. Perhaps our strange feeling towards semi-realistic beings can be attributed to low aesthetic value. On another note, the mortality salience hypothesis, another concept in line with evolutionary psychology, suggests that humans are reminded of death upon observing human replicas without human liveliness. As a result, humans experience a feeling of uncanniness as a coping mechanism. While these theories are plausible, further studies must be conducted to confirm the link between human reactions to the hyperrealistic and the psychological factors behind them.
With recent developments in robotic technology, robots are nearing human resemblance and intelligence. Futurists have predicted artificial intelligence surpassing human intelligence in the next few decades. Animation technologies are getting more realistic with each film, and the line between animation and lifelike figures is likely to become blurred to the extent that we recognize them as one and the same. It is only a matter of time before we are able to overcome the uncanny valley.
Journal of Neuroscience (2019). DOI: 10.1523/JNEUROSCI.2956-18.2019
IEEE Robotics & Automation Magazine (2012). DOI: 10.1109/MRA.2012.2192811