When facial expressions do and do not signal minds: The role of face inversion, expression dynamism, and emotion type.

Abstract

Recent research has linked facial expressions to mind perception. Specifically, Bowling and Banissy (2017) found that ambiguous doll-human morphs were judged as more likely to have a mind when smiling. Herein, we investigate 3 key potential boundary conditions of this "expression-to-mind" effect. First, we demonstrate that face inversion impairs the ability of happy expressions to signal mindful states in static faces; however, inversion does not disrupt this effect for dynamic displays of emotion. Finally, we demonstrate that not all emotions have equivalent effects. Whereas happy faces generate more mind ascription compared to neutral faces, we find that expressions of disgust actually generate less mind ascription than those of happiness. (PsycINFO Database Record

Happy = Human: A Feeling of Belonging Modulates the “Expression-to-Mind” Effect

“…Across Studies 1-3, we replicated the happy-to-human link observed in past work (Bowling & Banissy, 2017;Krumhuber et al, 2019), but extended this in three important ways. First, we demonstrated that the past results replicate with real human faces.…”

“…The present work competitively tested three possible explanations for the previouslyestablished "happiness-to-human" link (Bowling & Banissy, 2017;Krumhuber et al, 2019).…”

Sensitivity to face animacy and inversion in childhood: Evidence from EEG data

“…Compared to animate human faces, inanimate (artificial) faces are remembered more poorly (Balas & Pacella, 2015;Crookes et al, 2015) and discriminated less efficiently (Crookes et al, 2015). Artificial appearance impacts the perception of face gender (Balas, 2013) and trustworthiness (Balas & Pacella, 2017), and inhibits the attribution of psychological attributes (Looser & Wheatley, 2010) in a manner that is sensitive to facial expression (Bowling & Banissy, 2017;Krumhuber, Lai, Rosin, & Hugenberg, 2018), face inversion (Krumhuber et al, 2018), and group affiliation (Hackel, Looser, & Van Bavel, 2014).…”

Benchmarking non-photorealistic rendering of portraits

“…Image-based non-photorealistic rendering (NPR) is at the intersection of computer graphics and computer vision, and has the aim of synthesising new images based on the analysis of existing images 1 . NPR can be applied in many different ways, such as: the rendering of CAD models of furniture and interior designs as watercolour style illustrations to provide more appealing renderings for sales brochures [1], stylising images to different degrees to provide stimuli for perceptual studies investigating the theory of mind [2], stylising images to reduce patients' aversion to otherwise unpleasant pictures of surgical procedures [3], and image enhancement prior to generating 3D bas-reliefs in order to emphasise salient structures and reduce noise and visual clutter [4]. NPR can be applied to images, video, and 3D models, but in this paper we will focus on imagebased NPR, and in particular the rendering of portrait images, which is also known as portrait image stylisation.…”

“…• First, for a typical computer vision or machine learning task such as classification, regression or detection, there is normally assumed to be a correct solution, often referred to as "ground truth". However, for stylisation, ground truth generally does not exist; for instance, if a particular NPR task is to produce stylisations in the manner of the artist Monet it is not possible to acquire ideal images before and after 2. Some of these issues were identified by David Salesin in his NPAR 2002 keynote speech on seven Grand Challenges for NPR, where amongst other things he talked about (1) How can we quantify success, and (5) the Artistic Turing Test: can NPR achieve products indistinguishable from an artist's works?…”