Attentional Processing of Emotion FACES in THOSE High and Low in Autistic Behaviors and Social Anxiety

Background: Attention to or away from emotional faces may explain later deficits in emotion identification in individuals with autistic traits and social anxiety. Because these traits are significantly correlated yet separate constructs, it is important to examine the independent role they each play in the attentional processing of emotions.

Objectives: This study examined preferential attention to faces in a sub-clinical sample with different levels of autistic traits and social anxiety using a dot-probe paradigm.

Methods: The participants included undergraduate students (122 males, 133 females, 11 unspecified; M_age = 19.30 [SD = 1.89] years) from a medium-sized university in the southeastern United States. The pictures used in the dot probe paradigm included the same models displaying either a basic emotion (happy, fear, surprise, disgust) or a neutral expression (Tottenham et al., 2009). On each trial, two face stimuli were displayed simultaneously on the left and the right side of the screen; one face was an emotional face while the other was a neutral face depicting the same person. The presentation time of the faces was 100 ms, 200 ms, or 500 ms. The faces were followed by the presentation of a dot on one side of the screen. Reaction time to a button press indicating which side of the screen the dot was displayed was recorded as a measure of relative attention to the emotional face versus the neutral face. Participants also completed a self-report measure of autistic behaviors, the Broad Autism Phenotype Questionnaire (BAPQ; Hurley, Losh, Parlier, Reznick, & Piven, 2007), and the Social Phobia and Anxiety Inventory (SPAI-23), a self-report measure of social phobia (Roberson-Nay, Strong, Nay, & Beidel, 2007).

Results: Consistent with previous studies, BAPQ scores were correlated with SPAI scores, r = .51, p < .001. SPAI was included as a covariate in analyses involving BAPQ and BAPQ was used as a covariate in analyses including SPAI as a factor. A 3 (Block; 100, 200, 500ms) X 4 (Emotion: disgust, fear, happy, surprise) X 3 (BAPQ group: high, middle, low) ANOVA was conducted. Of most theoretical interest, this analysis revealed a main effect of BAPQ, F(2, 247) = 3.77, p = .024. Those who scored highest on the BAPQ were significantly slower than those who scored the lowest on the BAPQ. To assess the effects of social anxiety on dot probe reaction time, a 3 (Block; 100, 200, 500ms) X 4 (Emotion: disgust, fear, happy, surprise) X 3 (SPAI group: high, middle, low) was also conducted. Unlike what was observed for BAPQ scores, there was not a main effect for SPAI.

Conclusions: The present findings add to growing evidence that, despite being relatively strongly correlated, those who report high levels of autistic behaviors and those who report high levels of social anxiety demonstrate distinctive task performance, suggesting dissociable underlying processing mechanisms. Future research should assess the performance of the four potential subcategories (high or low in autistic behaviors and social anxiety or those who are only high in autistic behaviors or social anxiety).