Familiarity, orientation, and realism increase face uncanniness  by  sensitizing  to  facial distortions

Abstract

The uncanny valley predicts aversive reactions toward near-humanlike entities. Greater uncanniness is elicited by distortions in realistic than unrealistic faces, possibly due to familiarity. Experiment 1 investigated how familiarity and inversion affect uncanniness of facial distortions and the ability to detect differences between the distorted variants of the same face (distortion sensitivity). Familiar or unfamiliar celebrity faces were incrementally distorted and presented either upright or inverted. Uncanniness ratings increased across the distortion levels, and were stronger for familiar and upright faces. Distortion sensitivity increased with increasing distortion difference levels, again stronger for familiar and upright faces. Experiment 2 investigated how face realism, familiarity, and face orientation interacted for the increase of uncanniness across distortions. Realism increased the increase of uncanniness across the distortion levels, further enhanced by upright orientation and familiarity. The findings show that familiarity, upright orientation, and high face realism increase the sensitivity of uncanniness, likely by increasing distortion sensitivity. Finally, a moderated linear function of face realism and deviation level could explain the uncanniness of stimuli better than a quadratic function. A re-interpretation of the uncanny valley as sensitivity toward deviations from familiarized patterns is discussed.

The uncanniness of written text is explained by configural deviation and not by processing disfluency

“…As sufficient experience with a written language allows holistic processing of words (Björnström et al, 2014; Wong et al, 2010) and sensitivity to configural distortions (Wong et al, 2019), the moderating effect of familiarity on uncanniness can be explained by an intrinsic negative evaluation of stimuli that deviate from learned configural patterns. Familiarity has been shown to moderate the effect of configural deviation on uncanniness in faces (Diel & Lewis, 2022) and novel stimuli (Diel & Lewis, in review). Here, the effect is replicated with text stimuli.…”

“…Uncanniness and Deviation From Specialized Categories. Other researchers proposed that heightened sensitivity to deviations in specialized categories, especially faces (Diel & Lewis, 2022;MacDorman et al, 2009a;MacDorman & Chattopadhyay, 2016;Matsuda et al, 2012), amplifies the uncanniness of atypical stimuli. Diel and Lewis (2022) found that participants' sensitivity to uncanniness in deviating faces was increased when faces were familiar compared with novel, and upright compared with inverted, indicating an effect of deviation from familiar stimuli driven by perceptual experience with the stimulus type, driven by an increased ability to detect deviations in specialized categories.…”

“…The effect of perceptual disfluency depends on the expectations towards typical appearance, which may be driven by experience (Wänke & Hansen, 2015). Given that people are more aware of deviations or changes in more familiar or specialized stimuli (Diel & Lewis, 2022), potential deviations may be more readily processed disfluently in those categories. Thus, the same type of deviation may appear more aesthetically unappealing in more, compared with less, specialized categories due to increased processing disfluency.…”

“…Thus, the same type of deviation may appear more aesthetically unappealing in more, compared with less, specialized categories due to increased processing disfluency. In other words, the degree of familiarity or specialization would increase the sensitivity to deviations by increasing disfluency, and this effect would be more relevant for perceptual rather than for conceptual disfluency, given the specificity hypothesis (Diel & Lewis, 2022;Vogel et al, 2020).…”

“…These deviations could be detected especially with specialized (e.g., configural) processing. A higher sensitivity to deviations or errors for specialized categories could then lead to slightly deviating stimuli appearing unappealing, eerie or strange (Diel & Lewis, 2022; Diel & MacDorman, 2021). One way to investigate the extent to which the perception of uncanniness is relevant to domains beyond human or animal likeness is to test the effect of deviation on uncanniness in non-human stimulus categories, moderated by familiarity or specialization.…”