Attentional Sensitivity to Features of Angry Faces and Domain-General Cognitive Characteristics

Background:  
An atypical face and emotion processing in ASD have received wide attention in the research of ASD. It is, however, still not clear whether/ how their processing is different from typical people. In the current study, to assess their basic feature regarding emotional responses, we focused on the "anger superiority effect (ASE)". ASE refers to a phenomenon where an angry face is detected more quickly than a happy or neutral face in a crowd of distracters (i.e., face-in-the-crowd task). This is believed to stem from the attention-getting properties in such threatening stimuli. Previous studies have reported that ASE is also observed in individuals with ASD (Ashwin et al., 2006; Krysko & Rutherford, 2009; Isomura et al., 2014), but the underlying mechanisms may be different from those in TD. Although ASE is known to require the configulation of features in faces (Fox et al., 2000; Tipples et al., 2002; Weymer et al., 2011), for some individuals such as individuals with ASD who have local-biased cognitive characteristics, single isolated facial features may be sufficient to activate the systems of quick attention allocation to anger. Furthermore, the mechanisms for that may be attributed to more domain-general cognitive mechanisms depending on each individual’s characteristics.

Objectives:  
In this study, we examined whether ASD and TD children would show feature-based anger superiority effect, and explored the association with their domain-general cognitive characteristics, by focusing on global/local perceptual attentional bias.

Methods:  
Sixteen children with ASD and 16 TD children aged 7- to 13-year-old participated in this study. First they were given the face-in-the-crowd task that included three types of face-conditions: whole, eyebrows, and mouth conditions. In the eyebrows and mouth conditions, stimuli involved only the respective features with a facial contour. Global/local attentional bias was assessed by a Navon task, in which a bigger letter composed of smaller elements were identified at the global or local level. The size of individual global/local interference (namely, the Navon index) was calculated using individuals’ response times based on a previous study (Zhu et al., 2010), and was used for a correlation analysis with their performance on the face-in-the-crowd tasks.

Results:  
The quicker detection of angry faces over happy faces (i.e., Anger Superiority Effect) was observed both for TD and ASD children in the whole and eyebrows conditions. By contrast, in the mouth condition, only the ASD group showed the quicker detection of downward mouth over upward mouth. The results of the correlation analysis revealed that, in ASD, individuals who showed stronger local-biased perceptual characteristics tended to show the stronger single feature-based Anger Superiority Effect. In TD, by contrast, such relationships were not observed.  

Conclusions:
The findings suggested that, for ASD individuals, who have local-biased perceptual characteristics, emotion-related facial features may work as emotion-evoking stimuli. Extracting local features and perceiving emotions may be a foundation of quick attention allocating to angry faces in individuals with ASD