Though Pixar has seen much success from staying safe and not toying with the possibility of falling into the valley, many other major studios are beginning to take greater risks in their attempts to approach realistic human renderings. Weta Digital, the digital effects studio behind Lord of the Rings, Avatar, and the Planet of the Apes reboot trilogy, has created realistic non-human renderings that engage audiences just as much (or even more than) cartoonish human renderings, but it can be argued that no studio has successfully rendered a realistic human character that does not trigger the Uncanny Valley effect.
There remains limited research investigating the small components that contribute to the Uncanny Valley effect. Most audiences view human figures as a whole picture, but the various components that make up for the form itself - lighting, color, skin tone, reflections, shape, texture, movement, voice, muscle tone, and surrounding environment all contribute to the overall form and therefore to the Uncanny Valley. In Mori’s original graph he distinguishes between a still figure and a moving one, where a moving figure elicits an even more negative response than a figure which remains still. There is potential that advancement in any one specific area could lead to artists overcoming the Uncanny Valley, and there is also potential that computer-generated technology is not yet advanced enough to accurately render models that are realistic enough to fool the human mind.
Various theories have been proposed as to why the Uncanny Valley effect exists, and all of the major theories point toward a perceived threat of death by the individual experiencing the effect. There are five major theories pertaining to the existence of the Uncanny Valley: mortality salience, threat to human identity, conflicting perceptual cues, mate selection, and danger avoidance.
Mortality salience, or awareness of the inevitability of one’s own death, is derived from the theory of terror management \cite{Greenberg} and has been proposed as one explanation for the existence of the Uncanny Valley. In an attempt to show the significance of this theory as it pertains to the Uncanny Valley, MacDorman proposed a study stemming from the original hypothesis regarding mortality salience, which states that “if having a worldview guards people from anxiety about the inevitability of death (e.g., by giving a literal or symbolic explanation of how death is transcended), those who have been subliminally reminded of death will react more favorably to information that supports their worldview and less favorably to information that undermines it” \cite{MacDorman2005}. MacDorman’s study compared the reactions of two test groups to images supporting their worldview with one test group having previously been shown an image of a real human and the second group having previously been shown an image of a humanoid robot. Ultimately, it was concluded that the results of the study were favorable and suggesting that an uncanny-looking android may be assumed as such because it elicits a fear of death. The study “attempted to verify this through questions designed to measure such distal terror management defenses as worldview protection. [...] On average, the group exposed to an image of an uncanny robot consistently preferred information sources that supported their worldview relative to the control group.” MacDorman’s findings suggest that figures known to trigger the Uncanny Valley effect in humans do so because they exist as reminders of humans’ own mortality, but leave us wondering what specific attributes of those figures are indicative of death.
It has been proposed that human-looking robots pose a threat to the uniqueness of the human form, thus threatening the accepted identity humans have established of themselves. MacDorman and Entenzari investigated individual human traits as predictors of sensitivity to the Uncanny Valley, suggesting that individuals who hold human identity with higher regard tend to be more sensitive to the Uncanny Valley \cite{Macdorman2015}, while Ferrari and Paladino showed that anthropomorphism of robots presented a challenge to human distinctiveness and social identity \cite{Ferrari2016}. In summary, humans are prone to recognizing artificial yet realistic representations of the human form as a threat to their own identity.
Another proposed theory suggests that a disconnect between perceptual cues contributes to the effect of the Uncanny Valley. For instance, a robot with a humanoid appearance but robotic-sounding voice is more likely to fall into Uncanny Valley than would a robot whose appearance and voice match each other in human likeness. Mitchell et al. supported this claim with research into the perceived eeriness of figures with mismatched realism levels in appearance and voice \cite{Mitchell2011}. In design, it is assumed important to maintain equal levels of realism in visual appearance and behavior \cite{Garau_2003}. Additionally, Chattopadhyay & MacDorman showed that “reducing realism consistency did not make objects appear less familiar, but only animals and humans, thereby eliciting cold, eerie feelings" \cite{MacDorman2016}. Living things being the only things to generate negative emotions helps to suggest a threat of death as a driving force behind the Uncanny Valley.
The theory of mate selection as a reason behind the Uncanny Valley does not immediately suggest a threat of death, but rather suggests that individuals possess an innate desire to pass on their genetic material before they die. A study by Green et al. suggested that uncanniness in computer-generated faces was lessened by facial proportions deemed conventionally attractive, which, when combined with the well-established theory that perceived attractiveness is based on a desire to select a mate in good health with high fertility and a compatible immune system, leads to the belief that the Uncanny Valley phenomenon is a means of selecting mates that ensure a high change of producing healthy and fertile offspring, thus passing on genetic material to the next generation \cite{Green2008}.
The final, and perhaps most obvious theory on the Uncanny Valley, suggests that the Uncanny Valley exists as a means of avoiding dangerous and potentially life-threatening situations. A paper by Rozin and Fallon in 1987 suggested the Uncanny Valley phenomenon to be a means of disease avoidance, where figures triggering the effect were assumed diseased and potential carriers of illness \cite{Rozin1987}. Moosa and Ud-Dean added to this idea, suggesting that this phenomenon occurred due to a desire to avoid danger, and that a corpse could be an indicator of an imminent source of danger beyond just disease \cite{Moosa2010}.
While it can be assumed that no one of these theories on the Uncanny Valley is entirely correct and that the Uncanny Valley effect may be a complex combination of many factors, one thing is clear: the Uncanny Valley exists due to an instinctual desire among humans to avoid their inevitable deaths. However, it has since been shown that this effect stems beyond just humans and extends to affect chimpanzees and macaques as well \cite{Aronson2009}. In both chimpanzees and macaques, each species displayed an obvious aversion to computer-generated images of their own species as compared to a real image of the species displayed alongside it.
With this, it can be suggested that there exists an evolutionary component to the Uncanny Valley driving the phenomenon, which has been inherited over thousands of years from a common ancestor. Though the ancestors of humans are no longer living, there are many species alive today with which humans share a common ancestor: