Autism & Developmental

Artificial scaffolding: Augmenting social cognition by means of robot technology.

Ghiglino et al. (2023) · Autism research : official journal of the International Society for Autism Research 2023
★ The Verdict

A humanoid robot raised autistic children’s NEPSY-II Theory of Mind scores more than a plain cube robot did.

✓ Read this if BCBAs running social-skills groups for young autistic clients who can access a humanoid robot.
✗ Skip if Clinicians without robot access or those working solely on daily-living skills.

01Research in Context

01

What this study did

Ghiglino et al. (2023) tested whether a talking humanoid robot could teach Theory of Mind to autistic children better than a boxy machine.

Kids played short games with either a child-size robot that had a face, arms, and voice or with a plain cube-shaped robot.

Before and after the sessions the team gave the NEPSY-II Theory of Mind sub-test to see if scores moved.

02

What they found

Children who worked with the humanoid robot gained more points on the NEPSY-II ToM scale than children who worked with the cube robot.

The study calls the gain “significant,” meaning the robot with a face helped more than the box.

03

How this fits with other research

Soleiman et al. (2023) saw a similar boost: two robots acting out emotions taught kids to read happy, sad, angry, and scared faces.

Zhou et al. (2025) looked at 13 robot trials and say the field still needs clearer proof, even though the average effect is medium.

Kostrubiec et al. (2020) found mixed results: robot rewards lifted engagement but did not beat plain ball play for social growth.

The new ToM study adds a bright spot by showing one clear, measurable gain on a standard test, something earlier work called for.

04

Why it matters

If you have access to a NAO or similar humanoid robot, you can now try short ToM games and track NEPSY-II scores before and after.

Keep sessions brief, use simple scripts, and graph the child’s score change to see if the robot gives you an extra teaching tool.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Run a five-minute false-belief game with your NAO, note the child’s NEPSY-II score this week, and plan to re-test after ten sessions.

02At a glance

Intervention
other
Design
quasi experimental
Sample size
43
Population
autism spectrum disorder
Finding
positive

03Original abstract

The concept of scaffolding refers to the support that the environment provides in the acquisition and consolidation of new abilities. Technological advancements allow for support in the acquisition of cognitive capabilities, such as second language acquisition using simple smartphone applications There is, however, one domain of cognition that has been scarcely addressed in the context of technologically assisted scaffolding: social cognition. We explored the possibility of supporting the acquisition of social competencies of a group of children with autism spectrum disorder engaged in a rehabilitation program (age = 5.8 ± 1.14, 10 females, 33 males) by designing two robot-assisted training protocols tailored to Theory of Mind competencies. One protocol was performed with a humanoid robot and the other (control) with a non-anthropomorphic robot. We analyzed changes in NEPSY-II scores before and after the training using mixed effects models. Our results showed that activities with the humanoid significantly improved NEPSY-II scores on the ToM scale. We claim that the motor repertoire of humanoids makes them ideal platforms for artificial scaffolding of social skills in individuals with autism, as they can evoke similar social mechanisms to those elicited in human-human interaction, without providing the same social pressure that another human might exert.

Autism research : official journal of the International Society for Autism Research, 2023 · doi:10.1002/aur.2906