The neuropsychology of first impressions: Evidence from Huntington's disease

Abstract

Impairments of emotion recognition have been widely documented in Huntington's disease (HD), but little is known concerning how these relate to other aspects of social cognition, including first impressions of traits such as trustworthiness and dominance. Here, we introduce a novel and sensitive method to investigate the ability to evaluate trustworthiness and dominance from facial appearance, with control tasks measuring ability to perceive differences between comparable stimuli. We used this new method together with standard tests of face perception to investigate social cognition in HD. We found that a subgroup of people with HD was impaired at perceiving trustworthiness and dominance, and that perceiving trustworthiness and dominance were correlated with impaired facial expression recognition. In addition, we used diffusion tensor imaging (DTI) to provisionally identify candidate brain regions associated with social cognition by contrasting regional functional anisotropy (FA) measures between subgroups of HD participants showing normal or impaired perception of trustworthiness and dominance, and by correlating these regional brain abnormalities with behavioural performance on tests of emotion recognition. In this way we show for the first time alterations in perception of trustworthiness and dominance in people with HD and link these to regions which may map the boundaries of the social brain. The pattern of breakdown seen in this neurodegenerative disease can thus be used to explore potential inter-relationships between different components of social cognition.

Understanding facial impressions between and within identities

“…That a PCA-based approach can work so well shows that much of the information used to create consensual trait impressions can be found in the images themselves. Theoretical approaches to facial impression formation often emphasise inferences based on stereotypes involving gender or age (Macrae & Bodenhausen, 2000;Oldmeadow et al, 2013;Quinn & Macrae, 2011;Sutherland, Young, Mootz & Oldmeadow, 2015) and the importance of facial expression (Montepare & Dobish, 2003Oosterhof & Todorov, 2008;Said, Sebe, & Todorov, 2009;Sprengelmeyer et al, 2016;Zebrowitz, Kikuchi, & Fellous, 2007). While such higher-order inferential factors undoubtedly play a role in impression formation, our data show that much of the variance in impressions can be modelled without needing to use them as explicit mediators.…”

Faces, people and the brain: The 45th Sir Frederic Bartlett Lecture

“…Of course, these are not mutually exclusive possibilities, but one promising way to tease apart their contributions may again be through neuropsychological studies. For example, Sprengelmeyer et al (2016) found that atypical first impressions of faces by patients with Huntington’s disease were correlated with impairments affecting recognition of facial expressions—a finding that is very much in line with Secord’s (1958) and Zebrowitz’s (2017) ideas.…”

Pre-print

“…Popular applications include morphing faces between the average faces of many females and males to induce a sexual dimorphism transform to individual faces, to test the relative preferences of observers for these facial qualities (Glassenberg, Feinberg, Jones, Little, & DeBruine, 2010;Perrett et al, 1994). Other applications involve warping faces between individuals with relatively more or less symptoms of depression (Scott, Kramer, Jones, & Ward, 2013), transforming faces between average and very attractive appearances to alter their distinctiveness or typicality (Sofer, Dotsch, Wigboldus, & Todorov, 2015), manipulating apparent cues to height in faces (Re, DeBruine, Jones, & Perrett, 2013), increasing or decreasing qualities associated with perceived health (Re, DeBruine, Jones, & Perrett, 2013), altering the appearance of faces perceived as high or low on the Big 5 (Sutherland et al, 2015), and manipulating perceived trustworthiness and dominance to examine how clinical samples respond to these cues (Sprengelmeyer et al, 2016). This approachproducing composite images, and using them to warp individual faces between the composites in a gradated fashion -has recently been referred to data-driven, as it requires sampling a range of faces to create the composites before applying the FACE REGRESSION 7 changes across a subset of individual faces (Sutherland et al, 2017), and allows testing of observations that emerge from composite faces.…”