“The uncanny valley itself is where dwell monsters, in the classic sense of the word. Frankenstein’s creation, the undead, the ingeniously twisted demons of anime and their inspirations from legend and myth, and indeed all the walking terrors and horrors of man’s imagining belong here. In essence, they tend to be warped funhouse-mirror images of humanity.” (Bryant, para. 11)

As the empathy scale takes the nose dive into the uncanny valley, an interesting shift occurs. Though the robots themselves are only slightly more humanoid in nature to their counterparts on the other edge of the curve, the revulsion experienced by those humans that are interacting with them is drastically increased. As discussed earlier, Freud viewed this shift into the uncanny or unhomely as likened to the experience of repressed fears and memories. The familiar is experienced and felt once again in a strange and incomplete form. In the case of the design of robots or near-humans, this is the point where they are in fact very human in aesthetic, but because of how close they are to being perfect, and the fact that they are not yet perfect, they are likened to monsters; the observer focuses on their faults, and the overall experience is very unsettling. It is close, but not quite, and like corpses, the un-dead, puppets or monsters, these near-humans bring forth human fears of ghosts, death and nightmares.

The first examples of this dip into the uncanny valley are not as frightening as what will come as we journey deeper, but they are definitely a bit ‘off’ in their attempts to portray ‘real’ human emotions and relationships. Displayed in the three images at the top of the left-hand bar are ‘avatars’ recently submitted in an online contest entitled "The Uncanny Valley Expo”, an online competition to capture the “few precious seconds, [when] you get the genuine sense of interacting with a real person” (James, para. 1) through an electronic avatar. An avatar traditionally “refers to the bodily manifestation of a higher being, or the Supreme Being onto planet Earth” (Wikipedia, “Avatar”, para. 1), and in the case of the electronic worlds involved in gaming, chat or second life, these computer-generated selves represent the players in the virtual world. They are very customizable and have many human features, but they are definitely found past the uncanny peak on the graph, as their expressions are often empty and dead. Even in this contest, the characters are still lacking soul even as the users attempt to make them appear more human.

Again moving towards more and more human in appearance, we’ll now go back to robots, in this case, the NS5 from “I, Robot”. The NS5 is the newer, better robot that ultimately turns evil and manifests all the uncanny valley fears that the main character, Spooner, has towards the usually helpful automatons of his futuristic world. The movie spends most of its length focusing on the character of Sonny, an NS5 with a heart and, perhaps, a soul. However, as a prime example of the uncanny valley phenomenon, the viewers, and the other characters in the film, are constantly pulled backwards and forwards while deciding whether the robot is, in fact, evil, or whether the robot is to be loved, or empathized with. In the end, Sonny does turn out to be good, but aesthetically, he is deep in the valley, and is very hard to empathize with at all. Though obviously robotic, his human features; a very detailed face, soft plastic skinning over his chest, arms, legs and head, bright blue eyes and human-like movements; as they are combined with his robotic elements; visible cords and metal limbs, the faint views of his brain and motors functioning beneath the white shell, his lack of skin in key areas (neck, stomach, hands, feet); serve not to make the robot himself more acceptable, but instead make him seem more ‘off’, creepy or demonic. In sharp contrast to the friendly, loveable and cartoonish NS4, the NS5s are much more articulated and human-like, but at the same time easier to believe as potentially evil because of our aversion to their closer to human but still so far off form.

Returning once more to animation, a similar element of very human but very demonic is the recent efforts that have been made within animated films to create 3D virtual characters that are closer and closer to photorealistic. Imperfections are added, actors are motion captured, thousands of dollars and millions of hours are spent working on details and subtleties, all in an attempt to create characters that mystify and suspend belief, appearing more and more real. However, there has yet to be a perfect electronic human created, and because of this, the most recent attempts, though painstakingly detailed, still reside deeply and firmly in the uncanny valley. Unlike the avatars discussed above, who, though still uncanny, don’t attempt to duplicate their users, these characters try, with wrinkles and freckles and bouncy hair, to fool the viewer into believing them to be human. Within the movies, “Final Fantasy: The Spirits Within”, and “The Polar Express”, characters were created and laboured upon in order to make them appear to be more and more real. However, the ultimate effect is somewhat opposite to what the intention was, instead of becoming more believable, they become more and more disconnected from the viewer, as they fail to allow the viewer to identify with their obviously fake characteristics. In Final Fantasy: the Spirits Within, the “protagonist Dr. Aki Ross was animated in so much detail that it took an hour and a half to create each individual frame in which she appeared.”(Brook, para. 10). But in the end, the film opened to mixed reviews, “At first it's fun to watch the characters," wrote Peter Travers in Rolling Stone, "[b]ut then you notice a coldness in the eyes, a mechanical quality in the movements." (Brook, para. 10). In response to The Polar Express, a reviewer commented that, “Why, with all the millions pumped into the production of this technological "masterpiece," do all the children still look so creepy?” (Jenkins, para. 3). Though viewers will readily accept electronic environments and effects, they are less inclined to allow for discrepancies in the rendering human beings. “Neuroscientists argue that our brains have evolved specific mechanisms for face recognition, because being able to recognize something "wrong" in someone else's face has long been crucial to survival.” (Thompson, para. 7). In this way, the harder that animators try to make their electronic characters real, the deeper they will fall into the valley, until they manage to achieve photo-realistic characters. However, this does beg the question, if photo-realism is the only solution, why animate humans instead of using real actors?

Regardless, now that we have ourselves trudged deeply into the valley there is still one last type of ‘robot’ or semi-human rendering that needs to be presented before the balance shifts into robots that are nearly indistinguishable from human beings. These last examples are very very close to perfection, they are basically human in form, but in each case there is still something not quite right. In the cases of the golem from “The Golem ”, the Somnambulist, from “The Cabinet of Dr. Calgari”, the Robot Maria in Fritz Lang’s “Metropolis”, and Pris from “Blade Runner”, all of these characters are 99% human, but have one piece that prevents them from ending up on the other side of the uncanny valley. The golem has his sheer massive size, clunky movements and clay hair, the somnambulist, his frightening motion and sunken Frankenstein-like features, Robot-Maria has the nagging issue of her one half-closing eye, and Pris, though only in the later scenes, is dropped onto the creepy end of the valley by her use of black makeup around her eyes, her too-perfect somersaults, and the scene when she sits silent and motionless among the other dolls of the designer’s home. All of these characters are perhaps the most creepy of the ones discussed in this section, simply because of the fact that they are so close to ‘getting away with it’, if the viewer hadn’t looked too close, or had merely passed them by, they could have brushed them off as fully human. Being portrayed as malicious or demonic within their films doesn’t help with empathy either, the golem, somnambulist, Maria and Pris are all involved with murders or deadly circumstances, and through the people around them letting their guard down, cause chaos and monstrous results in their wakes.

Now that you are familiar with the middle zone of the uncanny valley graph, you can continue onwards to part 3 or go back to the introduction, or part 1. You can also view the full references page by clicking on the links below.