Differences Between Autistic and Non-autistic Adults in the Recognition of Anger From Dynamic Expressions Remain After Controlling for Alexithymia.

Abstract

A burgeoning literature suggests that alexithymia, and not autism, is responsible for the difficulties with static emotion recognition that are documented in the autistic population. Here we investigate whether alexithymia can also account for difficulties with dynamic facial expressions. Autistic and control adults (N=60) matched on age, gender, non-verbal reasoning ability and alexithymia, completed an emotion recognition task, which employed dynamic point light displays of emotional facial expressions that varied in speed and spatial exaggeration. The ASD group exhibited significantly lower recognition accuracy for angry, but not happy or sad, expressions with normal speed and spatial exaggeration. The level of autistic, and not alexithymic, traits was a significant predictor of accuracy for angry expressions with normal speed and spatial exaggeration.

The inside out model of emotion recognition: how the shape of one’s internal emotional landscape influences the recognition of others’ emotions

“…Building on the face identity literature [7][8][9] and principles of signal detection theory 10 , our a priori hypothesis was that emotion recognition accuracy would be positively predicted by the consistency of, and distance between, emotion representations. To test this, we constructed a linear mixed effects model with accuracy as the outcome variable, representational consistency, distance between emotion representations, the interaction between representational consistency and distance, AQ score, TAS score, and non-verbal reasoning as predictors (clinically www.nature.com/scientificreports/ relevant participant characteristics known to be involved in the experience and perception of emotion; e.g., [22][23][24][25][26][27][28] Since it is likely that emotion recognition is contingent not only on the clarity of emotion representations but also on the ability to match a displayed expression to one's visualization, we also included a visual matching task in our battery. This task assesses how well participants can visually match the speed of one expression to another displayed expression.…”

“…Although this paradigm has several methodological advantages-it can be completed online in just 25 to 35 minutes and does not require participants to translate their emotional experiences into words (see 68 )there are disadvantages of using such computer-based assessments. For example, by employing images to elicit emotional reactions (as is common in the literature e.g., 6,24,26,69,70 ), participants may respond based on how they think they should feel, rather than how they truly feel. Whilst this is a possibility, we specifically address this issue in our task instructions, thus minimizing the likelihood of participants responding in this way: when describing EmoMap, we told participants that "this isn't about what the image represents, or how you think other people, on average, respond to the images.…”

“…That is, we ask whether those with modular internal emotion maps, who have consistent and distinct experiences of anger, happiness and sadness, also tend to have consistent and distinct visual representations of angry, happy and sad facial expressions (note that these emotions were selected as they correspond to different regions in the circumplex model of emotion 21 , varying in both and valence). Throughout these analyses, we control for clinically relevant demographic factors known to be associated with the experience and perception of emotion (e.g., the level of autistic traits, the level of alexithymic traits, and non-verbal reasoning ability [22][23][24][25][26][27][28] ) to ensure that any relationships we discover exist even after accounting for these variables. Finally, we assess the contribution of the consistency and differentiation (i.e., distinctness) of emotional experiences and representations to the recognition of anger, happiness and sadness, and use structural equation modelling to construct the 'Inside Out Model' of emotion recognition; a model which provides insight into the psychological mechanisms by which one's experience of emotions "on the inside" influences one's ability to identify emotions in the "outside world".…”

“…EmoMap was validated by illustrating that individuals high in alexithymic traits, who by definition have difficulties differentiating their own emotions 31 , tended to have emotional landscape maps with a less modular, more overlapping structure, whereas those low in alexithymic traits had modular emotional landscapes. That is, linear mixed effects models predicting mean distance between clusters and mean distance within clusters with TAS score, AQ score, and non-verbal reasoning ability (clinically relevant demographic variables known to be associated with the experience and perception of emotion; e.g., [22][23][24][25][26][27][28], and with subject number as a random intercept revealed alexithymic traits as a significant negative predictor of distance between emotion clusters [F(1,267) = − 5.92, p < 0.05] and distance within emotion clusters [F(1,267) = − 6.16, p < 0.05]. In general, greater overlap was seen between anger and sadness [mean distance (SEM) = 14.39 (0.21)], than happiness and anger [mean distance (SEM) = 20.79 (0.29)], and happiness and sadness [mean distance (SEM) = 20.70 (0.29)] in participants' internal emotional landscapes (see Supplementary Information A for a full discussion).…”

The Inside Out Model of Emotion Recognition: How the Shape of One’s Internal Emotional Landscape Influences the Recognition of Others’ Emotions

“…Although this paradigm has several methodological advantages-it can be completed online in just 25 to 35 minutes and does not require participants to translate their emotional experiences into words (see 68 )there are disadvantages of using such computer-based assessments. For example, by employing images to elicit emotional reactions (as is common in the literature e.g., 6,24,26,69,70 ), participants may respond based on how they think they should feel, rather than how they truly feel. Whilst this is a possibility, we specifically address this issue in our task instructions, thus minimizing the likelihood of participants responding in this way: when describing EmoMap, we told participants that "this isn't about what the image represents, or how you think other people, on average, respond to the images.…”