scholarly journals It matters to me if you are human: Categorical perception in human and nonhuman agent spectra

2019 ◽  
Author(s):  
Eva Wiese ◽  
Patrick P. Weis
Keyword(s):  
Categorical Perception ◽  
Cognitive Conflict ◽  
Category Boundary ◽  
Uncanny Valley ◽  
Mouse Tracking ◽  
Human Agent ◽  
The Uncanny ◽  
Human Likeness ◽  
The Impact ◽  
Robotic Agent

Humanlike but not perfectly human agents frequently evoke feelings of eeriness, a phenomenon termed the Uncanny Valley (UV). The Categorical Perception Hypothesis proposes that effects associated with the UV are due to uncertainty as to whether to categorize agents falling into the valley as “human” or “nonhuman”. However, since UV studies have traditionally looked at agents of varying human-likeness, it remains unclear whether UV-related effects are due to categorical uncertainty in general or are specifically evoked by categorizations that require decisions regarding an agent’s human-likeness. Here, we used mouse tracking to determine whether agent spectra with (i.e., robot-human) and without (i.e., robot-animal and robot-stuffed animal) a human endpoint cause phenomena related to categorical perception to comparable extents. Specifically, we compared human and nonhuman agent spectra with respect to existence and location of a category boundary (H1-1 and H2-1), as well as the magnitude of cognitive conflict around the boundary (H1-2 and H2-2). The results show that human and nonhuman spectra exhibit category boundaries (H1-1) at which cognitive conflict is higher than for less ambiguous parts of the spectra (H1-2). However, in human agent spectra cognitive conflict maxima were more pronounced than for nonhuman agent spectra (H2-1) and category boundaries were shifted towards the human endpoint of the spectrum (H2-2). Overall, these results suggest a quantitatively, though not qualitatively, different categorization process for spectra containing human endpoints. Possible reasons and the impact for virtual and robotic agent design are discussed.

Cognitive Conflict as Possible Origin of the Uncanny Valley

2017 ◽  
Vol 61 (1) ◽  
pp. 1599-1603 ◽  
Author(s):  
Patrick P. Weis ◽  
Eva Wiese
Keyword(s):  
Positive Affect ◽  
Categorical Perception ◽  
Cognitive Conflict ◽  
Social Robotics ◽  
Linear Increase ◽  
Uncanny Valley ◽  
The Uncanny ◽  
Human Likeness ◽  
Popular View ◽  
Simultaneous Activation

In social robotics, the term Uncanny Valley describes the phenomenon that linear increases in human-likeness of an agent do not entail an equally linear increase in favorable reactions towards that agent. Instead, a pronounced dip or ‘valley’ at around 70% human-likeness emerges. One currently popular view to explain this drop in favorable reactions is delivered by the Categorical Perception Hypothesis. It is suggested that categorization of agents with mixed human and non-human features is associated with additional cognitive costs and that these costs are the cause of the Uncanny Valley. However, the nature of the cognitive costs is still matter of debate. The current study explores whether the cognitive costs associated with stimulus categorization around the Uncanny Valley could be due to cognitive conflict as evoked by simultaneous activation of two categories. Using the mouse tracking technique, we show that cognitive conflict indeed peaks around the Uncanny Valley region of human-likeness. Our findings lay the foundation for investigating the effects of cognitive conflict on positive affect towards agents of around 70% human-likeness, possibly leading to the unraveling of the origins of the Uncanny Valley.

Trading faces: Complete AI face doubles avoid the uncanny valley

2020 ◽  
Author(s):  
Christopher Welker ◽  
David France ◽  
Alice Henty ◽  
Thalia Wheatley
Keyword(s):  
Artificial Intelligence ◽  
Sexual Orientation ◽  
Uncanny Valley ◽  
Torture Survivors ◽  
The Uncanny ◽  
The Creation ◽  
Full Face ◽  
The Impact

Advances in artificial intelligence (AI) enable the creation of videos in which a person appears to say or do things they did not. The impact of these so-called “deepfakes” hinges on their perceived realness. Here we tested different versions of deepfake faces for Welcome to Chechnya, a documentary that used face swaps to protect the privacy of Chechen torture survivors who were persecuted because of their sexual orientation. AI face swaps that replace an entire face with another were perceived as more human-like and less unsettling compared to partial face swaps that left the survivors’ original eyes unaltered. The full-face swap was deemed the least unsettling even in comparison to the original (unaltered) face. When rendered in full, AI face swaps can appear human and avoid aversive responses in the viewer associated with the uncanny valley.

scholarly journals Sending an Avatar to Do a Human's Job: Compliance with Authority Persists Despite the Uncanny Valley

2015 ◽  
Vol 24 (1) ◽  
pp. 1-23 ◽  
Author(s):  
Himalaya Patel ◽  
Karl F. MacDorman
Keyword(s):  
Ethical Dilemma ◽  
Sensitive Information ◽  
Uncanny Valley ◽  
Human Communication ◽  
Risk Of Falling ◽  
Computer Interfaces ◽  
The Uncanny ◽  
Authority Figure ◽  
Human Likeness ◽  
Motion Quality

Just as physical appearance affects social influence in human communication, it may also affect the processing of advice conveyed through avatars, computer-animated characters, and other human-like interfaces. Although the most persuasive computer interfaces are often the most human-like, they have been predicted to incur the greatest risk of falling into the uncanny valley, the loss of empathy attributed to characters that appear eerily human. Previous studies compared interfaces on the left side of the uncanny valley, namely, those with low human likeness. To examine interfaces with higher human realism, a between-groups factorial experiment was conducted through the internet with 426 midwestern U.S. undergraduates. This experiment presented a hypothetical ethical dilemma followed by the advice of an authority figure. The authority was manipulated in three ways: depiction (digitally recorded or computer animated), motion quality (smooth or jerky), and advice (disclose or refrain from disclosing sensitive information). Of these, only the advice changed opinion about the ethical dilemma, even though the animated depiction was significantly eerier than the human depiction. These results indicate that compliance with an authority persists even when using an uncannily realistic computer-animated double.

scholarly journals Virtual faces evoke only a weak uncanny valley effect: An empirical investigation with controlled virtual face images

2018 ◽  
Author(s):  
Jari Kätsyri ◽  
Beatrice de Gelder ◽  
Tapio Takala
Keyword(s):  
Linear Relationship ◽  
Empirical Investigation ◽  
Subjective Evaluation ◽  
Objective Measures ◽  
Uncanny Valley ◽  
Face Images ◽  
The Uncanny ◽  
Human Likeness ◽  
Upward Curvature ◽  
Virtual Face

The uncanny valley (UV) hypothesis suggests that increasingly human-like robots or virtual characters elicit more familiarity in their observers (positive affinity) with the exception of near-human characters that elicit strong feelings of eeriness (negative affinity). We studied this hypothesis in three experiments with carefully matched images of virtual faces varying from artificial to realistic. We investigated both painted and computer-generated (CG) faces to tap a broad range of human-likeness and to test whether CG faces would be particularly sensitive to the UV effect. Overall, we observed a linear relationship with a slight upward curvature between human-likeness and affinity. In other words, less realistic faces triggered greater eeriness in an accelerating manner. We also observed a weak UV effect for CG faces; however, least human-like faces elicited much more negative affinity in comparison. We conclude that although CG faces elicit a weak UV effect, this effect is not fully analogous to the original UV hypothesis. Instead, the subjective evaluation curve for face images resembles an uncanny slope more than a UV. Based on our results, we also argue that subjective affinity should be contrasted against subjective ratherthan objective measures of human-likeness when testing UV.

scholarly journals Category Processing and the human likeness dimension of the Uncanny Valley Hypothesis: Eye-Tracking Data

2013 ◽  
Vol 4 ◽  
Author(s):  
Marcus Cheetham ◽  
Ivana Pavlovic ◽  
Nicola Jordan ◽  
Pascal Suter ◽  
Lutz Jancke
Keyword(s):  
Eye Tracking ◽  
Uncanny Valley ◽  
Tracking Data ◽  
The Uncanny ◽  
Human Likeness

Human-likeness assessment for the Uncanny Valley Hypothesis

2017 ◽  
Vol 13 (3) ◽  
Author(s):  
Paweł Łupkowski ◽  
Marek Rybka ◽  
Dagmara Dziedzic ◽  
Wojciech Włodarczyk
Keyword(s):  
Uncanny Valley ◽  
The Uncanny ◽  
Internal Part ◽  
Human Likeness ◽  
Typical Model ◽  
The Way

AbstractThe Uncanny Valley Hypothesis (UVH, proposed in the 1970s) suggests that looking at or interacting with almost human-like artificial characters would trigger eeriness or discomfort. We studied how well subjects can assess degrees of human likeness for computer-generated characters. We conducted two studies, where subjects were asked to assess human likeness of given computer-generated models (Study 1) and to point the most typical model for a given category (Study 2). The results suggest that evaluation of the way human likeness is assessed should be an internal part of UVH research.

scholarly journals Uncanny but not confusing: Multisite study of perceptual category confusion in the Uncanny Valley

2019 ◽  
Author(s):  
Maya B Mathur ◽  
David Reichling ◽  
FRANCESCA LUNARDINI ◽  
Alice Geminiani ◽  
Alberto Antonietti ◽  
...  
Keyword(s):  
Category Boundary ◽  
Causal Role ◽  
Uncanny Valley ◽  
Mediation Analyses ◽  
The Uncanny ◽  
Perceptual Category ◽  
Real Robot ◽  
Negative Feelings ◽  
Human Faces

Android robots that are close, but imperfect, likenesses of humans can provoke negative feelings of dislike and eeriness in humans (“Uncanny Valley” effect). We investigated whether category confusion between the perceptual categories of “robot” and “human” contributes to Uncanny Valley aversion. Using a novel, validated corpus of 182 images of real robot and human faces, we precisely estimated the shape of the Uncanny Valley and the location of the perceived robot/human boundary. To implicitly measure confusion, we tracked 358 subjects’ mouse trajectories as they categorized the faces. We observed a clear Uncanny Valley and a pattern of categorization supporting a perceived categorical boundary. Yet, in contrast to predictions of the category confusion mechanism hypothesis, the Uncanny Valley and category boundary locations did not coincide, and mediation analyses further failed to support a causal role of category confusion. These results suggest category confusion does not explain the Uncanny Valley effect.

scholarly journals Uncanny Valley in Video Games: An Overview

Homo Ludens ◽  
2019 ◽  
pp. 135-148
Author(s):  
Dawid Ratajczyk
Keyword(s):  
Video Games ◽  
Negative Emotions ◽  
Empirical Studies ◽  
Humanoid Robots ◽  
Uncanny Valley ◽  
Virtual Characters ◽  
The Uncanny ◽  
Game Industry ◽  
In The Beginning ◽  
The Impact

The uncanny valley is an idea proposed by Masahiro Mori (1970) regarding negative emotions present in contacts with almost humanlike characters. In the beginning, it was considered only in the context of humanoid robots, but this context was broadened by the development of highly realistic animations and video games. Particularly evident are players’ interests in the uncanny valley. Recently there have been a growing number of reports from empirical studies regarding participants’ perception of highly realistic characters. In the paper, a review of publications concerning the uncanny valley hypothesis in video games is presented, as are deliberations about the impact of the uncanny valley on the game industry. According to the results, there is a need to recognise which attributes of virtual characters cause the uncanny valley effect.

Sign in / Sign up

Export Citation Format

Share Document