Uncanny valley is a concept trying to explain the uneasy feeling that comes from human-looking things but not quite enough [1]. Researchers have come up with multiple theories trying to explain.
This concept was first used by Ernest Jentsch (1906), who considered it to have cognitive causes [2]. Later, it appeared in Masahiro Mori’s essay (1970); in which he argued something’s likeliness to humans increases people’s positive emotions towards the thing, but there is a part of that likeliness, that it looks like human but people don’t feel positive about it, Mori defines this sudden drop as uncanny valley [1].
Theories explaining the causes of the uncanny valley are examined in two main groups as evolutionary and cognitive conflicts. Threat avoidance hypothesis and evolutionary aesthetics hypothesis are examples of evolutionary explanations whereas mind perception hypothesis, violation of expectation hypothesis and categorical uncertainty hypothesis are cognitive conflict based [3].
Threat avoidance hypothesis
According to this hypothesis uncanny valley occurs to avoid the two biggest threats in the evolution, namely, death and disease.
Mortality salience is based on terror management theory which explains how humans deal with being aware of their mortality. In this theory imperfections in similarity cause people anxiety.
In pathogen avoidance people evaluate imperfections to be caused by diseases or genetically transmitting deformities and this awakens a feeling of discomfort [3].
Evolutionary aesthetics hypothesis
According to this hypothesis eeriness is about physical attractiveness. Imperfections on humanness is considered as unattractiveness. Although studies done on this matter are insufficient when tested eeriness negatively correlates to attractiveness [3].
Violation of the expectation hypothesis
This hypothesis is based on humans expecting something humanlike to act exactly like a human and feel uneasy when they don’t [3]. Mori wrote about a robot in his essay that was smiling (it had 29 muscles on its face like a human), when they made the robot smile half as fast as a human the smile looked creepier than it was at normal speed [1]; this could be shown as an example for this theory, people expected the smile to be developed at normal speed but it was slower than it should be so it looked creepier.
Category uncertainty theory
On this hypothesis giving human features to a nonhuman being causes cognitive conflicts. The reason for this is categories of human and nonhuman do not overlap, so when something nonhuman has human features causes negative emotions towards it because it is hard to categorize [4].
Studies about this don’t have a consensus. Studies done on only human-robot categorization supports Mori whereas studies comparing human-robot and human-animal categorizations didn’t show similar results [3].
Mind perception hypothesis
This hypothesis mostly centres around humanlike robots and says if a robot is too similar to a human people might consider it having a mind and emotions. Having a mind and emotions are human traits and giving them to a robot makes people anxious according to this theory [3].
Foundation of this theory is laid by Heather M. Gray, Kurt Gray and Daniel M. Wegner (2007). On this article they said that mind perception consist of two dimensions; agency which is ability to make a decision and act on it, experience on the other hand is a capability to feel. These two dimensions affect people’s judgments. Capability to understand and use these dimensions affect the moral responsibility we consider it should have and the moral judgment we give [5].
Years later Kurt Gray and Daniel M. Wagner (2012) tested this theory.
They used three experiment groups to do it. Experiment 1 observed if a machine’s humanlike appearance caused people to think it can experience and also feel uneasy, Experiment 2 looked if a machine doesn’t look like a human but is able to experience would still be making people uncomfortable, Experiment 3 examined if perceived lack of experience could explain the creepiness of unfeeling humans and philosophical zombies [6].
Results of the Experiment 1 showed; humanlike robots make people more uncomfortable that the robots having mechanical looks, humanlike and mechanical robot didn’t show a significant difference on perceived agency but showed difference in experience (humanlike rated higher on experience) which means uncanny response is caused by experience levels.
In Experimen 2, researchers focused on the robots having different experience levels so the robots looked mechanical. People rated uncanniness higher on robots with experience; agency alone didn’t have any effect on uncanniness, perceived experience is the cause of uncanny valley.
In Experiment 3 participants got the description for 3 different mental capacities (normal, lacking agency, lacking experience) and researchers measured participants’ discomfort towards them. Lacking experience version have been found to be the most disturbing, while lacking agency and normal one didn’t have differences in terms of uncanniness. This experiment showed having emotions constitutes a major part of being human, autonomy and capability has fewer parts in it [6].
REFERENCES
- Mori, M., MacDorman, K. F., & Kageki, N. (2012). The uncanny valley [from the field]. IEEE Robotics & automation magazine, 19(2), 98-100.
- Jentsch, E. (1997). On the Psychology of the Uncanny (1906). Angelaki: Journal of the Theoretical Humanities, 2(1), 7-16.
- Zhang, J., Li, S., Zhang, J. Y., Du, F., Qi, Y., & Liu, X. (2020). A literature review of the research on the uncanny valley. In Cross-Cultural Design. User Experience of Products, Services, and Intelligent Environments: 12th International Conference, CCD 2020, Held as Part of the 22nd HCI International Conference, HCII 2020, Copenhagen, Denmark, July 19–24, 2020, Proceedings, Part I 22(pp. 255-268). Springer International Publishing.
- MacDorman, K. F., & Chattopadhyay, D. (2016). Reducing consistency in human realism increases the uncanny valley effect; increasing category uncertainty does not. Cognition, 146, 190-205.
- Gray, H. M., Gray, K., & Wegner, D. M. (2007). Dimensions of mind perception. science, 315(5812), 619-619.
- Gray, K., & Wegner, D. M. (2012). Feeling robots and human zombies: Mind perception and the uncanny valley. Cognition, 125(1), 125-130.
Supervisor: Gizem Vurgun
