Autistic adults exhibit highly precise representations of others’ emotions but a reduced influence of emotion representations on emotion recognition accuracy

Abstract

To date, studies have not yet established the mechanisms underpinning differences in autistic and non-autistic emotion recognition. The current study first investigated whether autistic and non-autistic adults differed in terms of the precision and/or differentiation of their visual emotion representations and their general matching abilities, and second, explored whether differences therein were related to challenges in accurately recognizing emotional expressions. To fulfil these aims, 45 autistic and 45 non-autistic individuals completed three tasks employing dynamic point light displays of emotional facial expressions. We identified that autistic individuals had more precise visual emotion representations than their non-autistic counterparts, however, this did not confer any benefit for their emotion recognition. Whilst for non-autistic people, non-verbal reasoning and the interaction between precision of emotion representations and matching ability predicted emotion recognition, no variables contributed to autistic emotion recognition. These findings raise the possibility that autistic individuals are less guided by their emotion representations, thus lending support to Bayesian accounts of autism.

Mismatching expressions: Spatiotemporal and kinematic differences in autistic and non-autistic facial expressions

“…When we explored demographic factors, IQ was the only significant contributor for autistic individuals, explaining 22.6% of the variance in accuracy. These results add to a growing literature suggesting that different psychological mechanisms are involved in autistic and non-autistic emotion recognition (e.g., Keating, Ichijo & Cook, 2023;Keating, Kraaijkamp, & Cook, pre-print;Rump et al, 2009). Within this literature, there is evidence that the precision of visual emotion representations (i.e., the way one pictures others' emotional expression in the mind's eye) contributes to emotion recognition accuracy for non-autistic individuals, but not autistic individuals (e.g., Keating & Cook, pre-print;Keating, Ichijo & Cook, 2023).…”

“…We did not make any formal predictions regarding the magnitude of activation of facial landmarks since this evidence was highly mixed (e.g., Faso et al, 2015;Grossman et al, 2013;Lampi et al, 2023;Legiša, Messinger, Kermol, & Marlier, 2013;Loveland et al, 1994;Mathersul, McDonald, & Rushby, 2013;Oberman, Winkielman, & Ramachandran, 2009;Stagg et al, 2013;Yoshimura et al, 2015), and potentially confounded by alexithymia (see Trevisan, Bowering, &Birmingham, 2016 andKeating &Cook, 2020). Finally, in line with signal detection theory (McNicol, 2005) and previous findings (e.g., Keating, Ichijo & Cook, 2023;Keating & Cook, pre-print;Keating, Kraaijkamp & Cook, pre-print), we predicted that the precision and differentiation of participants' own productions would contribute to their ability to recognise others' facial expressions.…”

“…That is, we found that individuals high in alexithymia displayed elevated activation of the mouth smile blendshapes when posing anger, and the mouth frown blendshape when posing happiness (relative to those low in alexithymia). Thus, to formally test the contribution of autism and alexithymia to the differentiation of angry and happy spoken expressions, we conducted an exploratory random forests analysis (Breiman, 2001), using the Boruta wrapper algorithm (Kursa & Rudnicki, COMPARING AUTISTIC AND NON-AUTISTIC FACIAL EXPRESSIONS 2010; as in Keating, Ichijo & Cook, 2023;Keating & Cook, pre-print;Keating, Kraaijkamp & Cook, pre-print).…”

“…Building on the body movement and emotional experience literatures, we predicted that more jerky, and less precise and/or differentiated, expressions would be associated with reduced emotion recognition accuracy. We explored whether this was the case for both autistic and non-autistic individuals by conducting a random forests analysis (Brieman, 2001) separately for each group, using the Boruta wrapper algorithm (Kursa & Rudnicki, 2010; as in Keating, Ichijo & Cook, 2023;Keating & Cook, pre-print;Keating, Kraaijkamp & Cook, pre-print). To test our predictions, we included emotion recognition accuracy as the outcome variable: feature variables included mean jerk, mean jerk precision, mean jerk distance (i.e., differentiation) for both posed and spoken expressions; mean activation, mean activation precision, mean activation distance (i.e., differentiation) for both posed and spoken expressions; plus AQ and TAS.…”

“…However, a parallel literature, concerning the internal experience of emotion, draws attention to more specific features. This literature reports that the precision and differentiation of one's own emotional experiences and visual representations Keating, Ichijo, & Cook, 2023;Keating & Cook, pre-print;Keating, Kraaijkamp, & Cook, pre-print) may contribute to the ability to recognise the emotions of others. Indeed, this literature has its roots in signal detection theory (see McNicol, 2005), where it is argued that a 'signal' distribution and a 'noise' distribution that are imprecise (i.e., wide) and indistinct (i.e., overlapping) provide a low sensitivity to discriminate between the 'signal' and 'noise'.…”