Attention bias to emotional faces varies by IQ and anxiety in Williams syndrome
Supplementary Methods
Social dot-probe task
In this task, a neutral face and an emotional face (happy or angry) were presented simultaneously on the computer screen for 500 milliseconds (msec). Immediately after the faces disappeared, a gray dot appeared in the same location as either the neutral face or the emotional face. Participants were instructed to press one of the shift keys as quickly as possible on the left or right side of the keyboard corresponding to the location of the dot. Visual cues were installed on the keyboard to aid with key pressing. The probe remained on the screen until a response was made or until 10 seconds had passed. A fixation cross remained on the screen throughout the block of trials. The experiment began with six practice trials. Accuracy and reaction time (RT) were recorded for all trials. Congruent probes are those that appeared on the same side as the emotional face and incongruent probes were on the opposite side. An attention bias is quantified by a faster response to the congruent than the incongruent probes.
The experiment consisted of 288 experimental trials divided into 12 blocks of 24 trials. Images were from 24 different individuals (12 male, 12 female) from the Karolinska Directed Emotional Faces (Lundqvist, Flykt, & Ohman, 1998). The stimuli were adjusted to grayscale, and a gray border was added to cover the background and hair. On each trial two images were presented showing the same actor. Each block included 8 happy/neutral, 8 angry/neutral, and 8 neutral/neutral trials. Each actor was seen once in each block and an equal number of times in every possible condition. Within each block of 8 trials, emotional position and probe position were manipulated such that there were 4 congruent and 4 incongruent trials. The position of the emotion and probe were counter-balanced within the happy/neutral and angry/neutral trials and the probe was counterbalanced in the neutral/neutral trials.
The social dot-probe task took about 25 minutes to complete. After each of the 12 blocks, we offered breaks and behavioral supports, such as checklists for completed blocks, to help the participants maintain attention and motivation.
Social dot-probe data cleaning
For the within-individual data-cleaning, incorrect trials (M=4.7 trials, SD=4.4) and trials with timing errors (i.e., trials with reaction times (RTs) less than 200 milliseconds (ms) (M=.4 trials, SD=.9) were removed. Next, RTs that exceeded the individual’s mean + 2 standard deviations were removed which constituted, on average, 3% of the trials for an individual (M=8.5 trials, SD= 3.9). Analyses were performed with mean RT data calculated for each condition – happy congruent, happy incongruent, angry congruent, angry incongruent, and neutral. Following within-individual cleaning procedures, mean RTs were examined across individuals and conditions. One individual was dropped from further analysis because they were an extreme and consistent outlier across all conditions with greater than 1 sec mean RTs for several conditions. The remaining condition-specific outliers were winsorized to 2 standard deviations (4% of mean RTs). Bias scores were calculated by subtracting the congruent RT from the incongruent RT for happy and angry faces.
References
Lundqvist, D., Flykt, A., & Ohman, A. (1998). The Karolinska Directed Emotional Faces (KDEF) [CD-ROM]. Stockholm, Sweden: Department of Clinical Neuroscience, Psychology Section, Karolinska Institute.