Perceptual discrimination difficulty and familiarity in the Uncanny Valley: more like a "Happy Valley"

Ascribing consciousness to artificial intelligence: human-AI interaction and its carry-over effects on human-human interaction

The question of whether artificial intelligence (AI) can be considered conscious and therefore should be evaluated through a moral lens has surfaced in recent years. In this paper, we argue that whether AI is conscious is less of a concern than the fact that AI can be considered conscious by users during human-AI interaction, because this ascription of consciousness can lead to carry-over effects on human-human interaction. When AI is viewed as conscious like a human, then how people treat AI appears to carry over into how they treat other people due to activating schemas that are congruent to those activated during interactions with humans. In light of this potential, we might consider regulating how we treat AI, or how we build AI to evoke certain kinds of treatment from users, but not because AI is inherently sentient. This argument focuses on humanlike, social actor AI such as chatbots, digital voice assistants, and social robots. In the first part of the paper, we provide evidence for carry-over effects between perceptions of AI consciousness and behavior toward humans through literature on human-computer interaction, human-AI interaction, and the psychology of artificial agents. In the second part of the paper, we detail how the mechanism of schema activation can allow us to test consciousness perception as a driver of carry-over effects between human-AI interaction and human-human interaction. In essence, perceiving AI as conscious like a human, thereby activating congruent mind schemas during interaction, is a driver for behaviors and perceptions of AI that can carry over into how we treat humans. Therefore, the fact that people can ascribe humanlike consciousness to AI is worth considering, and moral protection for AI is also worth considering, regardless of AI's inherent conscious or moral status.

The inversion effect on the cubic humanness-uncanniness relation in humanlike agents

The uncanny valley describes the typically nonlinear relation between the esthetic appeal of artificial entities and their human likeness. The effect has been attributed to specialized (configural) processing that increases sensitivity to deviations from human norms. We investigate this effect in computer-generated, humanlike android and human faces using dynamic facial expressions. Angry and happy expressions with varying degrees of synchrony were presented upright and inverted and rated on their eeriness, strangeness, and human likeness. A sigmoidal function of human likeness and uncanniness ("uncanny slope") was found for upright expressions and a linear relation for inverted faces. While the function is not indicative of an uncanny valley, the results support the view that configural processing moderates the effect of human likeness on uncanniness and extend its role to dynamic facial expressions.

The deviation-from-familiarity effect: Expertise increases uncanniness of deviating exemplars

Humanlike entities deviating from the norm of human appearance are perceived as strange or uncanny. Explanations for the eeriness of deviating humanlike entities include ideas specific to human or animal stimuli like mate selection, avoidance of threat or disease, or dehumanization; however, deviation from highly familiar categories may provide a better explanation. Here it is tested whether experts and novices in a novel (greeble) category show different patterns of abnormality, attractiveness, and uncanniness responses to distorted and averaged greebles. Greeble-trained participants assessed the abnormality, attractiveness, uncanniness of normal, averaged, and distorted greebles and their responses were compared to participants who had not previously seen greebles. The data show that distorted greebles were more uncanny than normal greebles only in the training condition, and distorted greebles were more uncanny in the training compared to the control condition. In addition, averaged greebles were not more attractive than normal greebles regardless of condition. The results suggest uncanniness is elicited by deviations from stimulus categories of expertise rather than being a purely biological human- or animal-specific response.

Autistic people outperform neurotypicals in a cartoon version of the Reading the Mind in the Eyes

Prior research suggests that while autistic people may demonstrate poorer facial emotion recognition when stimuli are human, these differences lessen when stimuli are anthropomorphic. To investigate this further, this work explores emotion recognition in autistic and neurotypical adults (n = 196). Groups were compared on a standard and a cartoon version of the Reading the Mind in the Eyes test. Results indicated that autistic individuals were not significantly different from neurotypicals on the standard version. However, autistic people outperformed neurotypicals on the cartoon version. The implications for these findings regarding emotion recognition deficits and the social motivation account of autism are discussed and support the view of socio‐cognitive differences rather than deficits in this population.

Accepting Human-like Avatars in Social and Professional Roles

Humans report perceptions of unease or eeriness as humanoid/android robots and digital avatars approach human-like physical resemblance, a phenomenon alluded by the Uncanny Valley theory. This study extends the discussions on interactions and acceptance of digital avatars with findings from three experiments. In the first, perceptive evaluation of actors in clips from computer-generated animation and a live-action version of the same movie was examined. In the second experiment, we considered short clips with highly realistic digital avatars to measure recognition ability, the extent of eeriness, and specific physical features identified as unreal. The fixation area and pupil size variation recorded using an eye tracker were analyzed to infer attention to the body, face, and emotional response, respectively. Building on these findings, the third experiment looked at acceptance in roles requiring human skill, empathy, and cognitive ability. The results show that based on perceptions from physical attributes, the eeriness scores diverge from the uncanny valley theory as human-likeness increases. The realistic CGI and mocap technology could have helped cross the valley. Visual attention inferred from gaze behavior was similar for live-action and CGI. At the same time, we observe pupil size changes reflecting emotions like eeriness when the avatars either talked or smiled. Proficiency and acceptance scores were lower for roles requiring complex social cognition processes, such as friends and judicial decision-making. Interestingly, real-life stereotypes of gender roles were transferred to digital avatars too. The findings suggest an ambiguity in accepting human-like avatars in social and professional interactions, emphasizing the need for a multi-dimensional approach when applying the uncanny valley theory. A detailed and contextual examination is imperative as technological advancements have placed humans closer to co-existing with digital or physical android/humanoid robots.

A Meta-analysis of the Uncanny Valley's Independent and Dependent Variables

The uncanny valley (UV) effect is a negative affective reaction to human-looking artificial entities. It hinders comfortable, trust-based interactions with android robots and virtual characters. Despite extensive research, a consensus has not formed on its theoretical basis or methodologies. We conducted a meta-analysis to assess operationalizations of human likeness (independent variable) and the UV effect (dependent variable). Of 468 studies, 72 met the inclusion criteria. These studies employed 10 different stimulus creation techniques, 39 affect measures, and 14 indirect measures. Based on 247 effect sizes, a three-level meta-analysis model revealed the UV effect had a large effect size, Hedges’ g = 1.01 [0.80, 1.22]. A mixed-effects meta-regression model with creation technique as the moderator variable revealed face distortion produced the largest effect size, g = 1.46 [0.69, 2.24], followed by distinct entities, g = 1.20 [1.02, 1.38], realism render, g = 0.99 [0.62, 1.36], and morphing, g = 0.94 [0.64, 1.24]. Affective indices producing the largest effects were threatening, likable, aesthetics, familiarity , and eeriness , and indirect measures were dislike frequency, categorization reaction time, like frequency, avoidance , and viewing duration . This meta-analysis—the first on the UV effect—provides a methodological foundation and design principles for future research.

Creepy cats and strange high houses: Support for configural processing in testing predictions of nine uncanny valley theories

In 1970, Masahiro Mori proposed the uncanny valley (UV), a region in a human-likeness continuum where an entity risks eliciting a cold, eerie, repellent feeling. Recent studies have shown that this feeling can be elicited by entities modeled not only on humans but also nonhuman animals. The perceptual and cognitive mechanisms underlying the UV effect are not well understood, although many theories have been proposed to explain them. To test the predictions of nine classes of theories, a within-subjects experiment was conducted with 136 participants. The theories' predictions were compared with ratings of 10 classes of stimuli on eeriness and coldness indices. One type of theory, configural processing, predicted eight out of nine significant effects. Atypicality, in its extended form, in which the uncanny valley effect is amplified by the stimulus appearing more human, also predicted eight. Threat avoidance predicted seven; atypicality, perceptual mismatch, and mismatch+ predicted six; category+, novelty avoidance, mate selection, and psychopathy avoidance predicted five; and category uncertainty predicted three. Empathy's main prediction was not supported. Given that the number of significant effects predicted depends partly on our choice of hypotheses, a detailed consideration of each result is advised. We do, however, note the methodological value of examining many competing theories in the same experiment.

Pre-exposure to Ambiguous Faces Modulates Top-Down Control of Attentional Orienting to Counterpredictive Gaze Cues

Understanding and reacting to others' nonverbal social signals, such as changes in gaze direction (i.e., gaze cue), are essential for social interactions, as it is important for processes such as joint attention and mentalizing. Although attentional orienting in response to gaze cues has a strong reflexive component, accumulating evidence shows that it can be top-down controlled by context information regarding the signals' social relevance. For example, when a gazer is believed to be an entity "with a mind" (i.e., mind perception), people exert more top-down control on attention orienting. Although increasing an agent's physical human-likeness can enhance mind perception, it could have negative consequences on top-down control of social attention when a gazer's physical appearance is categorically ambiguous (i.e., difficult to categorize as human or nonhuman), as resolving this ambiguity would require using cognitive resources that otherwise could be used to top-down control attention orienting. To examine this question, we used mouse-tracking to explore if categorically ambiguous agents are associated with increased processing costs (Experiment 1), whether categorically ambiguous stimuli negatively impact top-down control of social attention (Experiment 2), and if resolving the conflict related to the agent's categorical ambiguity (using exposure) would restore top-down control to orient attention (Experiment 3). The findings suggest that categorically ambiguous stimuli are associated with cognitive conflict, which negatively impact the ability to exert top-down control on attentional orienting in a counterpredicitive gaze-cueing paradigm; this negative impact, however, is attenuated when being pre-exposed to the stimuli prior to the gaze-cueing task. Taken together, these findings suggest that manipulating physical human-likeness is a powerful way to affect mind perception in human-robot interaction (HRI) but has a diminishing returns effect on social attention when it is categorically ambiguous due to drainage of cognitive resources and impairment of top-down control.

Macaque Gaze Responses to the Primatar: A Virtual Macaque Head for Social Cognition Research

Following the expanding use and applications of virtual reality in everyday life, realistic virtual stimuli are of increasing interest in cognitive studies. They allow for control of features such as gaze, expression, appearance, and movement, which may help to overcome limitations of using photographs or video recordings to study social responses. In using virtual stimuli however, one must be careful to avoid the uncanny valley effect, where realistic stimuli can be perceived as eerie, and induce an aversion response. At the same time, it is important to establish whether responses to virtual stimuli mirror responses to depictions of a real conspecific. In the current study, we describe the development of a new virtual monkey head with realistic facial features for experiments with nonhuman primates, the "Primatar." As a first step toward validation, we assessed how monkeys respond to facial images of a prototype of this Primatar compared to images of real monkeys (RMs), and an unrealistic model. We also compared gaze responses between original images and scrambled as well as obfuscated versions of these images. We measured looking time to images in six freely moving long-tailed macaques (Macaca fascicularis) and gaze exploration behavior in three rhesus macaques (Macaca mulatta). Both groups showed more signs of overt attention to original images than scrambled or obfuscated images. In addition, we found no evidence for an uncanny valley effect; since for both groups, looking times did not differ between real, realistic, or unrealistic images. These results provide important data for further development of our Primatar for use in social cognition studies and more generally for cognitive research with virtual stimuli in nonhuman primates. Future research on the absence of an uncanny valley effect in macaques is needed, to elucidate the roots of this mechanism in humans.

Emotional appraisal processing of computer-generated facial expressions: an functional near-infrared spectroscopy study

OBJECTIVE

The current study aims to investigate whether computer generated (CG) expressions of emotion evoke similar emotional appraisal processing in the lateral orbitofrontal cortex (lOFC) compared to real human expressions, as well as how speech cues would influence the processing.

METHODS

Functional near-infrared spectroscopy was used to measure the neural activations in the prefrontal cortex during emotion recognition task. Thirty normal participants were asked to view videos of dynamic facial expressions and selected the emotions that were best matches with the expressions.

RESULTS

CG expressions evoked less activation in the lOFC comparing to real human expressions. Furthermore, speech cues increased the activation in the lOFC for CG expressions but not real expressions.

CONCLUSION

Comparing to real expressions, CG expressions evoked less appraisal processing related to motivational values although this disadvantage can be compensated to some extent by presenting the expressions with speech cues.

Dissociating affective and cognitive dimensions of uncertainty by altering regulatory focus

Cognitive uncertainty is evidenced across learning, memory, and decision-making tasks. Uncertainty has also been examined in studies of positive affect and preference by manipulating stimulus presentation frequency. Despite the extensive research in both of these areas, there has been little systematic study into the relationship between affective and cognitive uncertainty. Using a categorization task, the present study examined changes in cognitive and affective uncertainty by manipulating stimulus presentation frequency and processing focus (i.e., promotion v. prevention focus). Following training, participants categorized stimuli and provided ratings of both typicality and negative affect. Results indicated that cognitive uncertainty was influenced by a categorical representation of stimuli whereas affective uncertainty was also influenced by exemplar presentation frequency during training. We additionally found that when the training was framed in terms of the avoidance of errors (i.e., a prevention focus), categorization performance was affected across the stimulus continuum whereas affective ratings remained unchanged.

Seeing minds in others: Mind perception modulates low-level social-cognitive performance and relates to ventromedial prefrontal structures

In social interactions, we rely on nonverbal cues like gaze direction to understand the behavior of others. How we react to these cues is affected by whether they are believed to originate from an entity with a mind, capable of having internal states (i.e., mind perception). While prior work has established a set of neural regions linked to social-cognitive processes like mind perception, the degree to which activation within this network relates to performance in subsequent social-cognitive tasks remains unclear. In the current study, participants performed a mind perception task (i.e., judging the likelihood that faces, varying in physical human-likeness, have internal states) while event-related fMRI was collected. Afterwards, participants performed a social attention task outside the scanner, during which they were cued by the gaze of the same faces that they previously judged within the mind perception task. Parametric analyses of the fMRI data revealed that activity within ventromedial prefrontal cortex (vmPFC) was related to both mind ratings inside the scanner and gaze-cueing performance outside the scanner. In addition, other social brain regions were related to gaze-cueing performance, including frontal areas like the left insula, dorsolateral prefrontal cortex, and inferior frontal gyrus, as well as temporal areas like the left temporo-parietal junction and bilateral temporal gyri. The findings suggest that functions subserved by the vmPFC are relevant to both mind perception and social attention, implicating a role of vmPFC in the top-down modulation of low-level social-cognitive processes.

Affective responses to coherence in high and low risk scenarios

Presenting information in a coherent fashion has been shown to increase processing fluency, which in turn influences affective responses. The pattern of responses have been explained by two apparently competing accounts: hedonic marking (response to fluency is positive) and fluency amplification (response to fluency can be positive or negative, depending on stimuli valence). This paper proposes that these accounts are not competing explanations, but separate mechanisms, serving different purposes. Therefore, their individual contributions to overall affective responses should be observable. In three experiments, participants were presented with businesses scenarios, with riskiness (valence) and coherence (fluency) manipulated, and affective responses recorded. Results suggested that increasing the fluency of stimuli increases positive affect. If the stimulus is negative, then increasing fluency simultaneously increases negative affect. These affective responses appeared to cancel each other out (Experiment 1) when measured using self-report bipolar scales. However, separate measurement of positive and negative affect, either using unipolar scales (Experiment 2) or using facial electromyography (Experiment 3), provided evidence for co-occurring positive and negative affective responses, and therefore the co-existence of hedonic marking and fluency amplification mechanisms.

Evaluating the replicability of the uncanny valley effect

The uncanny valley (UV) effect refers to an eerie feeling of unfamiliarity people get while observing or interacting with robots that resemble humans almost but not quite perfectly. The effect is not well understood, and it is also unclear how well results from previous research on the UV can be replicated. In six studies, both in the laboratory and online (N = 1343), we attempted to replicate the UV effect with various stimuli used in previous research. In Studies 1 and 2 we failed to replicate the UV effect with CGI stimuli created using the so-called morphing technique (a robot image morphed into a human image, resulting in a supposedly creepy robot-human image). In Studies 3a and 3b we found a prominent UV effect using pre-evaluated, non-morphed and photorealistic robot pictures. Finally, in exploratory Studies 4a and 4b we found the UV effect using morphed and photorealistic human and robot pictures. Our results suggest that the UV effect is more robust when elicited by pre-validated or prima facie uncanny robot pictures than by non-photorealistic images generated using the morphing technique. We argue that photorealistic pictures are more suitable than less realistic CGI pictures as stimuli for research attempting to elicit the UV effect – however, our results do not invalidate any previous research on the UV effect using morphing techniques, but point to their domain of applicability and context sensitivity.

The uncanny valley effect in typically developing children and its absence in children with autism spectrum disorders

Robots and virtual reality are gaining popularity in the intervention of children with autism spectrum disorder (ASD). To shed light on children’s attitudes towards robots and characters in virtual reality, this study aims to examine whether children with ASD show the uncanny valley effect. We varied the realism of facial appearance by morphing a cartoon face into a human face, and induced perceptual mismatch by enlarging the eyes, which has previously been shown as an effective method to induce the uncanny valley effect in adults. Children with ASD and typically developing (TD) children participated in a two-alternative forced choice task that asked them to choose one they liked more from the two images presented on the screen. We found that TD children showed the effect, i.e., the enlargement of eye size and the approaching realism reduced their preference. In contrast, children with ASD did not show the uncanny valley effect. Our findings in TD children help resolve the controversy in the literature about the existence of the uncanny valley effect among young children. Meanwhile, the absence of the uncanny valley effect in children with ASD might be attributed to their reduced sensitivity to subtle changes of face features and their limited visual experience to faces caused by diminished social motivation. Last, our findings provide practical implications for designing robots and virtual characters for the intervention of children with ASD.

Avoidance of Novelty Contributes to the Uncanny Valley

A hypothesis suggests that objects with a high degree of visual similarity to real humans trigger negative impressions (i.e., the uncanny valley). Previous studies have suggested that difficulty in object categorization elicits negative emotional reactions to enable the avoidance of potential threats. The present study further investigated this categorization-difficulty hypothesis. In an experiment, observers categorized morphed images of photographs and human doll faces as "photograph" or "doll" and evaluated the perceived eeriness of the images. Additionally, we asked the observers to answer questionnaires on behavioral inhibition systems (BIS). The results indicated that individual differences in the BIS score were associated with enhanced eeriness in the objects with a specific human likeness. These findings suggest that the tendency to avoid a potentially threatening novel experience contributes to promoting the perceived eeriness of objects with some degree of visual similarity to real humans.

Robots As Intentional Agents: Using Neuroscientific Methods to Make Robots Appear More Social

Robots are increasingly envisaged as our future cohabitants. However, while considerable progress has been made in recent years in terms of their technological realization, the ability of robots to interact with humans in an intuitive and social way is still quite limited. An important challenge for social robotics is to determine how to design robots that can perceive the user's needs, feelings, and intentions, and adapt to users over a broad range of cognitive abilities. It is conceivable that if robots were able to adequately demonstrate these skills, humans would eventually accept them as social companions. We argue that the best way to achieve this is using a systematic experimental approach based on behavioral and physiological neuroscience methods such as motion/eye-tracking, electroencephalography, or functional near-infrared spectroscopy embedded in interactive human-robot paradigms. This approach requires understanding how humans interact with each other, how they perform tasks together and how they develop feelings of social connection over time, and using these insights to formulate design principles that make social robots attuned to the workings of the human brain. In this review, we put forward the argument that the likelihood of artificial agents being perceived as social companions can be increased by designing them in a way that they are perceived as intentional agents that activate areas in the human brain involved in social-cognitive processing. We first review literature related to social-cognitive processes and mechanisms involved in human-human interactions, and highlight the importance of perceiving others as intentional agents to activate these social brain areas. We then discuss how attribution of intentionality can positively affect human-robot interaction by (a) fostering feelings of social connection, empathy and prosociality, and by (b) enhancing performance on joint human-robot tasks. Lastly, we describe circumstances under which attribution of intentionality to robot agents might be disadvantageous, and discuss challenges associated with designing social robots that are inspired by neuroscientific principles.

You Look Human, But Act Like a Machine: Agent Appearance and Behavior Modulate Different Aspects of Human-Robot Interaction

Gaze following occurs automatically in social interactions, but the degree to which gaze is followed depends on whether an agent is perceived to have a mind, making its behavior socially more relevant for the interaction. Mind perception also modulates the attitudes we have toward others, and determines the degree of empathy, prosociality, and morality invested in social interactions. Seeing mind in others is not exclusive to human agents, but mind can also be ascribed to non-human agents like robots, as long as their appearance and/or behavior allows them to be perceived as intentional beings. Previous studies have shown that human appearance and reliable behavior induce mind perception to robot agents, and positively affect attitudes and performance in human–robot interaction. What has not been investigated so far is whether different triggers of mind perception have an independent or interactive effect on attitudes and performance in human–robot interaction. We examine this question by manipulating agent appearance (human vs. robot) and behavior (reliable vs. random) within the same paradigm and examine how congruent (human/reliable vs. robot/random) versus incongruent (human/random vs. robot/reliable) combinations of these triggers affect performance (i.e., gaze following) and attitudes (i.e., agent ratings) in human–robot interaction. The results show that both appearance and behavior affect human–robot interaction but that the two triggers seem to operate in isolation, with appearance more strongly impacting attitudes, and behavior more strongly affecting performance. The implications of these findings for human–robot interaction are discussed.

Human Perception of Animacy in Light of the Uncanny Valley Phenomenon

The uncanny valley hypothesis by Japanese roboticist Masahiro Mori posits a nonlinear relation between human replicas’ human likeness and the emotional responses they elicit. In three studies, we corroborated the uncanny valley hypothesis, using the uncanny phenomenon as a vehicle to shed a new light on human animacy perception. In Study 1, 62 participants rated emotional responses and human likeness of 89 artificial and human faces. In Study 2, another 62 participants conducted a visual looming task with the same 89 faces allowing for the measurement of perceived threat. Results support the uncanny valley hypothesis, suggesting that the uncanny feeling may serve a function to wary humans of the potential danger of entities crossing the animate–inanimate boundary. In Study 3, 36 participants sorted faces as either real or unreal as quickly as possible in a reaction time sorting task allowing for the measurement of categorical uncertainty associated with animacy perception. Faces associated with longer sorting reaction times were also those associated with the highest ratings of negative emotions, suggesting that categorical uncertainty in animacy detection is related to the uncanny feeling. Results are discussed in light of human animacy perception and new directions for future research are suggested.

Circling Around the Uncanny Valley: Design Principles for Research Into the Relation Between Human Likeness and Eeriness

The uncanny valley effect (UVE) is a negative emotional response experienced when encountering entities that appear almost human. Research on the UVE typically investigates individual, or collections of, near human entities but may be prone to methodological circularity unless the properties that give rise to the emotional response are appropriately defined and quantified. In addition, many studies do not sufficiently control the variation in human likeness portrayed in stimulus images, meaning that the nature of stimuli that elicit the UVE is also not well defined or quantified. This article describes design criteria for UVE research to overcome the above problems by measuring three variables (human likeness, eeriness, and emotional response) and by using stimuli spanning the artificial to human continuum. These criteria allow results to be plotted and compared with the hypothesized uncanny valley curve and any effect observed can be quantified. The above criteria were applied to the methods used in a subset of existing UVE studies. Although many studies made use of some of the necessary measurements and controls, few used them all. The UVE is discussed in relation to this result and research methodology more broadly.

When categorization-based stranger avoidance explains the uncanny valley: A comment on MacDorman and Chattopadhyay (2016)

Artificial objects often subjectively look eerie when their appearance to some extent resembles a human, which is known as the uncanny valley phenomenon. From a cognitive psychology perspective, several explanations of the phenomenon have been put forth, two of which are object categorization and realism inconsistency. Recently, MacDorman and Chattopadhyay (2016) reported experimental data as evidence in support of the latter. In our estimation, however, their results are still consistent with categorization-based stranger avoidance. In this Discussions paper, we try to describe why categorization-based stranger avoidance remains a viable explanation, despite the evidence of MacDorman and Chattopadhyay, and how it offers a more inclusive explanation of the impression of eeriness in the uncanny valley phenomenon.

Familiar faces rendered strange: Why inconsistent realism drives characters into the uncanny valley

Computer-modeled characters resembling real people sometimes elicit cold, eerie feelings. This effect, called the uncanny valley, has been attributed to uncertainty about whether the character is human or living or real. Uncertainty, however, neither explains why anthropomorphic characters lie in the uncanny valley nor their characteristic eeriness. We propose that realism inconsistency causes anthropomorphic characters to appear unfamiliar, despite their physical similarity to real people, owing to perceptual narrowing. We further propose that their unfamiliar, fake appearance elicits cold, eerie feelings, motivating threat avoidance. In our experiment, 365 participants categorized and rated objects, animals, and humans whose realism was manipulated along consistency-reduced and control transitions. These data were used to quantify a Bayesian model of categorical perception. In hypothesis testing, we found reducing realism consistency did not make objects appear less familiar, but only animals and humans, thereby eliciting cold, eerie feelings. Next, structural equation models elucidated the relation among realism inconsistency (measured objectively in a two-dimensional Morlet wavelet domain inspired by the primary visual cortex), realism, familiarity, eeriness, and warmth. The fact that reducing realism consistency only elicited cold, eerie feelings toward anthropomorphic characters, and only when it lessened familiarity, indicates the role of perceptual narrowing in the uncanny valley.

The Uncanny Valley: Existence and Explanations

More than 40 years ago, Japanese roboticist Masahiro Mori (1970/2005) proposed the “uncanny valley” hypothesis, which predicted a nonlinear relation between robots’ perceived human likeness and their likability. Although some studies have corroborated this hypothesis and proposed explanations for its existence, the evidence on both fronts has been mixed and open to debate. We first review the literature to ascertain whether the uncanny valley exists. We then try to explain the uncanny phenomenon by reviewing hypotheses derived from diverse theoretical and methodological perspectives within psychology and allied fields, including evolutionary, social, cognitive, and psychodynamic approaches. Next, we provide an evaluation and critique of these studies by focusing on their methodological limitations, leading us to question the accepted definition of the uncanny valley. We examine the definitions of human likeness and likability, and propose a statistical test to preliminarily quantify their nonlinear relation. We argue that the uncanny valley hypothesis is ultimately an engineering problem that bears on the possibility of building androids that may some day become indistinguishable from humans. In closing, we propose a dehumanization hypothesis to explain the uncanny phenomenon.

Reducing consistency in human realism increases the uncanny valley effect; increasing category uncertainty does not

Human replicas may elicit unintended cold, eerie feelings in viewers, an effect known as the uncanny valley. Masahiro Mori, who proposed the effect in 1970, attributed it to inconsistencies in the replica's realism with some of its features perceived as human and others as nonhuman. This study aims to determine whether reducing realism consistency in visual features increases the uncanny valley effect. In three rounds of experiments, 548 participants categorized and rated humans, animals, and objects that varied from computer animated to real. Two sets of features were manipulated to reduce realism consistency. (For humans, the sets were eyes-eyelashes-mouth and skin-nose-eyebrows.) Reducing realism consistency caused humans and animals, but not objects, to appear eerier and colder. However, the predictions of a competing theory, proposed by Ernst Jentsch in 1906, were not supported: The most ambiguous representations-those eliciting the greatest category uncertainty-were neither the eeriest nor the coldest.

Navigating a social world with robot partners: A quantitative cartography of the Uncanny Valley

Android robots are entering human social life. However, human-robot interactions may be complicated by a hypothetical Uncanny Valley (UV) in which imperfect human-likeness provokes dislike. Previous investigations using unnaturally blended images reported inconsistent UV effects. We demonstrate an UV in subjects' explicit ratings of likability for a large, objectively chosen sample of 80 real-world robot faces and a complementary controlled set of edited faces. An "investment game" showed that the UV penetrated even more deeply to influence subjects' implicit decisions concerning robots' social trustworthiness, and that these fundamental social decisions depend on subtle cues of facial expression that are also used to judge humans. Preliminary evidence suggests category confusion may occur in the UV but does not mediate the likability effect. These findings suggest that while classic elements of human social psychology govern human-robot social interaction, robust UV effects pose a formidable android-specific problem.

Arousal, valence, and the uncanny valley: psychophysiological and self-report findings

The main prediction of the Uncanny Valley Hypothesis (UVH) is that observation of humanlike characters that are difficult to distinguish from the human counterpart will evoke a state of negative affect. Well-established electrophysiological [late positive potential (LPP) and facial electromyography (EMG)] and self-report [Self-Assessment Manikin (SAM)] indices of valence and arousal, i.e., the primary orthogonal dimensions of affective experience, were used to test this prediction by examining affective experience in response to categorically ambiguous compared with unambiguous avatar and human faces (N = 30). LPP and EMG provided direct psychophysiological indices of affective state during passive observation and the SAM provided self-reported indices of affective state during explicit cognitive evaluation of static facial stimuli. The faces were drawn from well-controlled morph continua representing the UVH’ dimension of human likeness (DHL). The results provide no support for the notion that category ambiguity along the DHL is specifically associated with enhanced experience of negative affect. On the contrary, the LPP and SAM-based measures of arousal and valence indicated a general increase in negative affective state (i.e., enhanced arousal and negative valence) with greater morph distance from the human end of the DHL. A second sample (N = 30) produced the same finding, using an ad hoc self-rating scale of feelings of familiarity, i.e., an oft-used measure of affective experience along the UVH’ familiarity dimension. In conclusion, this multi-method approach using well-validated psychophysiological and self-rating indices of arousal and valence rejects – for passive observation and for explicit affective evaluation of static faces – the main prediction of the UVH.

Persistence of the uncanny valley: the influence of repeated interactions and a robot's attitude on its perception

The uncanny valley theory proposed by Mori has been heavily investigated in the recent years by researchers from various fields. However, the videos and images used in these studies did not permit any human interaction with the uncanny objects. Therefore, in the field of human-robot interaction it is still unclear what, if any, impact an uncanny-looking robot will have in the context of an interaction. In this paper we describe an exploratory empirical study using a live interaction paradigm that involved repeated interactions with robots that differed in embodiment and their attitude toward a human. We found that both investigated components of the uncanniness (likeability and eeriness) can be affected by an interaction with a robot. Likeability of a robot was mainly affected by its attitude and this effect was especially prominent for a machine-like robot. On the other hand, merely repeating interactions was sufficient to reduce eeriness irrespective of a robot's embodiment. As a result we urge other researchers to investigate Mori's theory in studies that involve actual human-robot interaction in order to fully understand the changing nature of this phenomenon.

Impact of stimulus uncanniness on speeded response

In the uncanny valley phenomenon, the causes of the feeling of uncanniness as well as the impact of the uncanniness on behavioral performances still remain open. The present study investigated the behavioral effects of stimulus uncanniness, particularly with respect to speeded response. Pictures of fish were used as visual stimuli. Participants engaged in direction discrimination, spatial cueing, and dot-probe tasks. The results showed that pictures rated as strongly uncanny delayed speeded response in the discrimination of the direction of the fish. In the cueing experiment, where a fish served as a task-irrelevant and unpredictable cue for a peripheral target, we again observed that the detection of a target was slowed when the cue was an uncanny fish. Conversely, the dot-probe task suggested that uncanny fish, unlike threatening stimulus, did not capture visual spatial attention. These results suggested that stimulus uncanniness resulted in the delayed response, and importantly this modulation was not mediated by the feelings of threat.

Is it the real deal? Perception of virtual characters versus humans: an affective cognitive neuroscience perspective

Recent developments in neuroimaging research support the increased use of naturalistic stimulus material such as film, avatars, or androids. These stimuli allow for a better understanding of how the brain processes information in complex situations while maintaining experimental control. While avatars and androids are well suited to study human cognition, they should not be equated to human stimuli. For example, the uncanny valley hypothesis theorizes that artificial agents with high human-likeness may evoke feelings of eeriness in the human observer. Here we review if, when, and how the perception of human-like avatars and androids differs from the perception of humans and consider how this influences their utilization as stimulus material in social and affective neuroimaging studies. First, we discuss how the appearance of virtual characters affects perception. When stimuli are morphed across categories from non-human to human, the most ambiguous stimuli, rather than the most human-like stimuli, show prolonged classification times and increased eeriness. Human-like to human stimuli show a positive linear relationship with familiarity. Secondly, we show that expressions of emotions in human-like avatars can be perceived similarly to human emotions, with corresponding behavioral, physiological and neuronal activations, with exception of physical dissimilarities. Subsequently, we consider if and when one perceives differences in action representation by artificial agents versus humans. Motor resonance and predictive coding models may account for empirical findings, such as an interference effect on action for observed human-like, natural moving characters. However, the expansion of these models to explain more complex behavior, such as empathy, still needs to be investigated in more detail. Finally, we broaden our outlook to social interaction, where virtual reality stimuli can be utilized to imitate complex social situations.

A review of empirical evidence on different uncanny valley hypotheses: support for perceptual mismatch as one road to the valley of eeriness

The uncanny valley hypothesis, proposed already in the 1970s, suggests that almost but not fully humanlike artificial characters will trigger a profound sense of unease. This hypothesis has become widely acknowledged both in the popular media and scientific research. Surprisingly, empirical evidence for the hypothesis has remained inconsistent. In the present article, we reinterpret the original uncanny valley hypothesis and review empirical evidence for different theoretically motivated uncanny valley hypotheses. The uncanny valley could be understood as the naïve claim that any kind of human-likeness manipulation will lead to experienced negative affinity at close-to-realistic levels. More recent hypotheses have suggested that the uncanny valley would be caused by artificial–human categorization difficulty or by a perceptual mismatch between artificial and human features. Original formulation also suggested that movement would modulate the uncanny valley. The reviewed empirical literature failed to provide consistent support for the naïve uncanny valley hypothesis or the modulatory effects of movement. Results on the categorization difficulty hypothesis were still too scarce to allow drawing firm conclusions. In contrast, good support was found for the perceptual mismatch hypothesis. Taken together, the present review findings suggest that the uncanny valley exists only under specific conditions. More research is still needed to pinpoint the exact conditions under which the uncanny valley phenomenon manifests itself.

Stimulus-category competition, inhibition, and affective devaluation: a novel account of the uncanny valley

Stimuli that resemble humans, but are not perfectly human-like, are disliked compared to distinctly human and non-human stimuli. Accounts of this "Uncanny Valley" effect often focus on how changes in human resemblance can evoke different emotional responses. We present an alternate account based on the novel hypothesis that the Uncanny Valley is not directly related to 'human-likeness' per se, but instead reflects a more general form of stimulus devaluation that occurs when inhibition is triggered to resolve conflict between competing stimulus-related representations. We consider existing support for this inhibitory-devaluation hypothesis and further assess its feasibility through tests of two corresponding predictions that arise from the link between conflict-resolving inhibition and aversive response: (1) that the pronounced disliking of Uncanny-type stimuli will occur for any image that strongly activates multiple competing stimulus representations, even in the absence of any human-likeness, and (2) that the negative peak of an 'Uncanny Valley' should occur at the point of greatest stimulus-related conflict and not (in the presence of human-likeness) always closer to the 'human' end of a perceptual continuum. We measured affective responses to a set of line drawings representing non-human animal-animal morphs, in which each continuum midpoint was a bistable image (Experiment 1), as well as to sets of human-robot and human-animal computer-generated morphs (Experiment 2). Affective trends depicting classic Uncanny Valley functions occurred for all continua, including the non-human stimuli. Images at continua midpoints elicited significantly more negative affect than images at endpoints, even when the continua included a human endpoint. This illustrates the feasibility of the inhibitory-devaluation hypothesis and the need for further research into the possibility that the strong dislike of Uncanny-type stimuli reflects the negative affective consequences of cognitive inhibition.

A reappraisal of the uncanny valley: categorical perception or frequency-based sensitization?

The uncanny valley (UCV) hypothesis describes a non-linear relationship between perceived human-likeness and affective response. The “uncanny valley” refers to an intermediate level of human-likeness that is associated with strong negative affect. Recent studies have suggested that the uncanny valley might result from the categorical perception of human-like stimuli during identification. When presented with stimuli sharing human-like traits, participants attempt to segment the continuum in “human” and “non-human” categories. Due to the ambiguity of stimuli located at a category boundary, categorization difficulty gives rise to a strong, negative affective response. Importantly, researchers who have studied the UCV in terms of categorical perception have focused on categorization responses rather than affective ratings. In the present study, we examined whether the negative affect associated with the UCV might be explained in terms of an individual's degree of exposure to stimuli. In two experiments, we tested a frequency-based model against a categorical perception model using a category-learning paradigm. We manipulated the frequency of exemplars that were presented to participants from two categories during a training phase. We then examined categorization and affective responses functions, as well as the relationship between categorization and affective responses. Supporting previous findings, categorization responses suggested that participants acquired novel category structures that reflected a category boundary. These category structures appeared to influence affective ratings of eeriness. Crucially, participants' ratings of eeriness were additionally affected by exemplar frequency. Taken together, these findings suggest that the UCV is determined by both categorical properties as well as the frequency of individual exemplars retained in memory.