Comparing Social Skills Between Adults with ASD and Schizophrenia

Background: Although Autism Spectrum Disorder (ASD) and Schizophrenia (SCZ) are both characterized by social dysfunction, their non-shared features and different developmental trajectories suggest that the pathways and mechanisms underlying this impairment may differ. Systematically-matched comparisons of the two groups can help identify disorder-specific features that are missed when simply comparing each clinical group to typically-developing (TD) controls.

Objectives: While an increasing number of studies have compared ASD and SCZ on social cognitive abilities and their neural correlates (for a review, see Sasson et al., 2011), no work has yet compared the two on social behavior. This project sought to examine how social skills are similar and different between ASD and SCZ relative to TD adults.

Methods: 164 individuals (54 ASD; 54 SCZ; 56 TD) participated. The groups were matched on gender (all 87% male), and comparable on age (group means between 25.69-28.68 years), ethnicity (80-89% Caucasian), educational attainment (group means between 13.45-13.89 years), and IQ (group means between: 103.32-106.02). Participants completed the Social Skills Performance Assessment (SSPA; Patterson et al., 2001), a structured three minute role-play involving a social interaction scenario (meeting a new neighbor). Eleven social skills were coded on a one-to-nine scale (Pinkham & Penn, 2006): verbal clarity, verbal fluency, meshing, involvement, asks questions, gaze, appropriate affect, flat affect, appropriate content, social anxiety, and overall social skill. Additionally, three codes were added for their relevance to ASD: repetitive verbal content, repetitive nonverbal content, and verbosity. Two coders, blind to the diagnostic category of the participants, were trained to adequate reliability (ICC>.7 for all codes).

Results: As measured by the PANSS (Kay et al., 1987), the SCZ group had greater positive, negative, and general symptoms relative to the ASD group (ps <.001). Only negative symptoms correlated with overall social skill (r=-.56, p<.001), and thus were co-varied in analyses. Both clinical groups were rated lower on each of the 14 social skills relative to the TD group (all ps<.05), with the exception of SCZ not differing from controls on two ASD-specific items (repetitive verbal and nonverbal content), ASD not differing on verbal clarity, and both groups not differing from controls on meshing. When directly compared, the SCZ group was rated as higher on overall social skill than the ASD group (p=.001). The SCZ group also asked more questions (p=.004), were rated as more involved (p=.008), demonstrated less repetitive nonverbal (p=.002) and verbal (p=.050) content, and trended towards more appropriate affect (p=.064) and content, and less anxiety (p=.077). IQ correlated with overall social skill in SCZ (r=.35, p=.011) but not in ASD (r=.15, p=.271).

Conclusions: Both ASD and SCZ are characterized by significant social skill impairments relative to controls. When directly comparing the clinical groups, ASD demonstrated poorer social skills, both overall and within several specific areas (reciprocity, appropriate content and affect, and repetitive behaviors). These results underscore the significant social impairment experienced by intellectually-capable adults with ASD. Future work will examine how these patterns of social skills relate to social cognition and other areas of social functioning.