Brief report: feasibility of social cognition and interaction training for adults with high functioning autism.

Geurts et al. (2008) ran a small group class for adults with high-functioning autism. The class taught Social Cognition and Interaction Training, or SCIT-A for short.

Each week the adults met in person. They practiced reading faces, guessing thoughts, and talking with others. The team wanted to know if the lessons were doable and helpful.

The program worked. Adults got better at theory-of-mind tasks. They also showed small gains in real-life social talk.

No one dropped out, so the plan was easy to run.

Sawyer et al. (2014) later moved the same idea into a 3-D virtual classroom for teens. Their online version also raised social skills, showing the lessons can travel beyond a face-to-face adult group.

Koegel et al. (2025) swapped the group format for an AI chatbot. The bot gave short daily empathy drills to autistic teens and adults. Scores still shot up, proving social-cognition training keeps working even when a machine leads the drill.

Shum et al. (2019) tested PEERS, another group social-skills class, but with Chinese teens and parent coaches. Their strong RCT results line up with the positive trend seen in SCIT-A, backing the wider point that structured peer learning helps autistic speakers.

You now have choices. If you serve verbal adults with ASD, you can pull ready-made SCIT-A lessons off the shelf. If your client is a rural teen, try the virtual version. If travel is tough, an AI chatbot can pinch-hit for daily practice. Pick the mode that fits your setting, but keep the core: teach mind-reading, give peer reps, and track change.