LEGO Therapy: The Roles, the Rules and the Evidence
LEGO therapy is a group approach to social communication in which children build a model together in fixed roles, so that asking, describing, waiting and negotiating are built into the task. This guide is for parents, teachers and clinicians: the roles, a workable rule set, what a group needs, and an honest read of the trial evidence.
Reviewed byMichael Callans, MSWThe roles in a collaborative build
A workable rule set, posted where everyone can see it
- If you break it, you fix it. If you cannot fix it, ask someone to help.
- If someone else is using it, do not take it. Ask for it.
- Inside voices. No shouting, and no shouting over anyone.
- Keep hands and feet to yourself, and keep bricks out of mouths.
- Nobody takes apart someone else's model without asking them.
- Stay in your role, even when you can see the answer. Telling the builder what to do is the engineer's job.
- Clean up together at the end, counting the pieces back into the box.
How a session runs (about 60 minutes)
- Open with the rules read aloud by one participant, and name who holds which role for today.
- Warm up with five to ten minutes of freestyle building to a shared theme, in pairs rather than alone.
- Run the collaborative build: one set, three roles, no role swapping mid-model.
- Pause once in the middle for a short check: does anyone need something they have not asked for yet?
- Finish the model, then let the group show it and say one thing that was hard and one that worked.
- Pack up together, and agree who takes which role next time before anyone leaves.
What you need to run a group
Bricks
One or two sets with printed instructions, sized to finish inside a session, plus a tub of loose bricks for freestyle. Sets can be rebuilt for weeks, so the cost is one-off.
Space and time
A table everyone can reach, out of general traffic, for 45 to 60 minutes at a fixed weekly slot. Predictability is part of the method.
Group
Three participants is the standard unit, up to six with a foreman. Mix in children who find the roles easy, and keep anyone who dislikes building out of it.
One adult
A teaching assistant, parent or clinician who can coach the process, redirect requests to the right person, and resist finishing the model themselves.
What the adult does, and does not do
- Prompt the engineer to describe in words rather than point: which brick, what color, how many.
- Send requests back to the right person. When a participant asks you for a brick, point them to the supplier.
- Let short silences and small muddles sit. The repair is the skill being practiced.
- Comment on the process, not the product: who asked clearly, who waited, who noticed a mistake and said so.
- Do not take a role, do not build, and do not tidy the bricks into a better system mid-session.
- Do not use brick time as a reward or a sanction. The moment it becomes contingent, the motivation the method relies on is gone.
A note on the evidence
LeGoff's 2004 study of 47 participants with a waiting-list design found improvements on three social competence measures at 12 and 24 weeks, and a 2006 follow-up reported that the gains held. Both came from the developer with unblinded raters. Independent work is more cautious. A 2021 systematic review of 19 studies in Brain Sciences called the method promising for social interaction while warning that the studies were small and of low quality. The largest trial, I-SOCIALISE, randomized 98 mainstream schools in the United Kingdom and found that teacher-rated social skills at 20 weeks favored the groups without reaching statistical significance, at a standardized effect size of 0.18, with no clear benefit at 52 weeks. The fair summary: low cost, well liked, probably harmless, and not yet shown to work.
When it is not the right fit
Skip it when the participant does not enjoy building, since the whole method rests on wanting to do the activity. Think twice when a group cannot hold the rules safely yet, when bricks are a special interest that is already a point of conflict at home, or when the real need is speech and language therapy, an assessment, or help with sensory load in the classroom. A weekly club is also the wrong answer to a crisis. If distress, school refusal or aggression is the presenting problem, that needs its own attention first, and a group can come later.
Naming and trademark
LEGO is a registered trademark of the LEGO Group. The name is used on this page only to describe the method that LeGoff published and that researchers have studied; Psychology.com is not affiliated with, endorsed by or sponsored by the LEGO Group, and nothing here is official guidance from the company. For that reason the 2023 I-SOCIALISE trial used the neutral label Play Brick Therapy, and programs that charge for training or sessions generally do the same. Any brand of construction brick that fits together reliably will run the same session.
Related tools & tests
Want support around social communication?
A clinician who works with autistic children can help you pick what is worth the family's time. Free to search our directory.
Therapists Who Work With Autism
A worksheet is a start. These licensed therapists work with the same problem; every member is matched to the federal NPI Registry and board checks are dated on the profile.
-
View profile
Ginny BrundeenLCSW, LICSW
Identity matched to the NPI Registry
“I seek to help clients who feel stuck to find a path forward by empowering them to make meaningful changes in their lives.”
Lake Stevens, Washington Telehealth $165 per session, sliding scale
-
View profile
Karmelyn LeeBCBA
Identity matched to the NPI Registry
“BCBA and counseling intern helping kids, teens, and parents through behavior, big feelings, and school struggles with practical, evidence-based tools.”
Tampa, Florida Telehealth and in person $100 to $120 per session
- View profile
About this tool
LEGO therapy, also called LEGO based therapy or brick-based therapy, is a group method for working on social communication in which children build a model together while holding fixed roles. It was developed by Daniel LeGoff, a clinical neuropsychologist in the United States, who noticed in the late 1990s that children on his caseload who struggled to interact in conversation did so readily when a brick build gave them something to interact about. He published the first study in the Journal of Autism and Developmental Disorders in 2004, and the method was set out as a manual in 2014 by LeGoff with Gina Gomez de la Cuesta, G.W. Krauss and Simon Baron-Cohen.
The idea behind LEGO therapy is that the medium does a lot of the therapeutic work. The build is predictable and rule-governed, it has a clear end point, success is visible, and for many children it is intrinsically motivating in a way that a social skills lesson is not. Split the job across three people and communication stops being the exercise and becomes the means to an end: you cannot finish the model without asking for a piece, describing what you need, waiting your turn, and sorting out a disagreement about where the wheels go. The social practice is embedded in the task rather than taught on a worksheet.
The collaborative build is the core. In a group of three, one child takes the engineer role and holds the instructions, describing what is needed next. A second is the supplier, who keeps the bricks and hands over what is asked for, and nothing else. The third is the builder, who assembles the model from what the engineer describes and the supplier provides. Nobody can do their job without talking to the other two, and nobody can quietly take over. Roles rotate between sessions or between models so each child gets practice in all three positions, and larger groups split into pairs or add a foreman who checks that everyone has what they need.
Read the diagram as text
Three figures stand in a row, each holding one thing: the engineer holds a ruled sheet of instructions, the supplier holds a tray of bricks, the builder stands beside a part-built model. A neutral arc runs from the engineer to the supplier, the request; an accented arc runs from the supplier to the builder carrying one accented brick, the piece asked for and handed over. A faint dashed arc skips both and runs straight from the tray to the model: an adult supplies the brick, and the asking never happens.
Three roles, one brick at a time
Conceptual sequence, not an outcome
The engineer: Holds the instructions and describes what is needed next, without pointing
The supplier: Holds the bricks and hands over what is asked for, and nothing else
The builder: Assembles the model from what the engineer describes and the supplier provides
The model: Grows one piece at a time, and only through the handoffs
The engineer → asks for the piece and describes it → The supplier
The supplier → hands over exactly what was asked for → The builder
The builder → the model grows by one brick → The model
The supplier → an adult supplies the brick and the asking never happens → The model (error transition)
LEGO therapy is used mainly with autistic children and children with social communication difficulties, typically from about six to the early teens, and runs most often as a weekly school club of around 12 weeks. It also appears in clinic groups, in special education settings, and in mixed groups that include children who are not autistic. The target is not a diagnosis but a set of skills: joint attention, turn taking, requesting and clarifying, handling a plan that changes, and tolerating someone else's idea about the build. If you are at the stage of wondering whether social difficulties have a developmental explanation at all, Child & Teen Autism Test is a structured starting point rather than an answer.
The original evidence is LeGoff's own. His 2004 study used a repeated-measures waiting-list design with 47 participants aged six to sixteen and reported significant improvement on three measures of social competence at both 12 and 24 weeks, with no gains during the waiting period. A retrospective follow-up by LeGoff and Michael Sherman in 2006 reported that gains held up over three years. Both studies came from the developer, outcome raters were not independent, and one of the measures used has been criticized as a weak instrument for the purpose, so the early results are better read as promising than as proof.
The best independent read is a 2021 systematic review in Brain Sciences by Antonio Narzisi and colleagues, which covered 19 studies and concluded that LEGO based therapy looks promising for social interaction in autism while urging caution because of small samples and low study quality. The largest trial to date, I-SOCIALISE, was published in the journal Autism in 2023 by Barry Wright and colleagues: a cluster randomized trial in 98 mainstream schools in the United Kingdom with 250 autistic pupils aged seven to fifteen, comparing 12 weeks of the method plus usual support against usual support alone. Teacher-rated social skills at 20 weeks favored the intervention but did not reach statistical significance, with a standardized effect size of 0.18, and the difference at 52 weeks was smaller still.
Taken together, the honest position is that LEGO therapy is cheap, popular with families and staff, very unlikely to do harm, and not established as an effective treatment. The I-SOCIALISE authors reported that parents and teachers thought it was good for the children taking part and that schools found it inexpensive to run, which matters when the alternative is often nothing at all. It has not been included in the major evidence reviews of autism interventions, and nobody should be told it will change a child's social development. Choosing it as a pleasant, low-risk weekly group with plausible social benefit is defensible; choosing it instead of speech and language therapy or other recommended support is not.
A word on aims. The method works best when the goal is practice and enjoyment rather than making a child look less autistic. Pressure to perform eye contact, scripted phrases or constant smiling turns a liked activity into masking, which autistic adults consistently describe as exhausting and costly. The useful targets are practical: asking for what you need, saying when an instruction was unclear, waiting without the wait becoming unbearable, finishing something with other people. A child who does those things in their own style has succeeded. Our broader Social Skills Activities for Kids collection covers other activities with the same low-pressure framing.
Running a group needs less than people expect: one or two sets with instructions, a loose brick supply, a table, three to six participants and one adult who can stay out of the way. The adult's job is to coach the process, not the build. Prompt the engineer to describe rather than point, send requests back to the right person when a child appeals to the adult instead, let short silences sit, and comment on what the group did rather than how the model looks. Avoid two common errors: becoming the supplier yourself, and using brick time as a reward to be withdrawn, which breaks the one thing the method depends on.
At home the same structure works with a sibling, a cousin or a visiting friend, though two people means the engineer and builder roles with the parent on bricks, and the group dynamics that do most of the work are thinner. Many families find the printable rule set more useful than the roles: a short, visible, consistent list of expectations for shared play. Keep sessions short, keep them voluntary, and stop while everyone still wants more. LEGO is a trademark of the LEGO Group, used here only to name the method; this page is not affiliated with or endorsed by the company, and the 2023 trial used the neutral name Play Brick Therapy for the same reason.
References
- LeGoff DB. Use of LEGO as a therapeutic medium for improving social competence. Journal of Autism and Developmental Disorders. 2004;34(5):557-571. doi:10.1007/s10803-004-2550-0
- LeGoff DB, Gomez de la Cuesta G, Krauss GW, Baron-Cohen S. LEGO-Based Therapy: How to Build Social Competence Through LEGO-Based Clubs for Children with Autism and Related Conditions. London: Jessica Kingsley Publishers; 2014.
- LeGoff DB, Sherman M. Long-term outcome of social skills intervention based on interactive LEGO play. Autism. 2006;10(4):317-329.
- Narzisi A, Sesso G, Berloffa S, et al. Could you give me the blue brick? LEGO-based therapy as a social development program for children with autism spectrum disorder: a systematic review. Brain Sciences. 2021;11(6):702. doi:10.3390/brainsci11060702
- Wright B, Kingsley E, Cooper C, et al. I-SOCIALISE: results from a cluster randomised controlled trial investigating the social competence and isolation of children with autism taking part in LEGO based therapy clubs in school environments. Autism. 2023;27(8):2281-2294. doi:10.1177/13623613231159699
Cite this source
Psychology.com. (2026, November 6). LEGO Therapy. Psychology.com. https://psychology.com/tools/lego-therapy
LEGO Therapy FAQ
What is LEGO therapy?
LEGO therapy is a group method for working on social communication, usually with autistic children, in which a model is built collaboratively in fixed roles. One participant is the engineer and holds the instructions, one is the supplier and hands over bricks on request, and one is the builder. Because no role can finish the model alone, asking, describing, waiting and negotiating are built into the activity rather than taught separately. Daniel LeGoff developed it in the United States and published the first study in 2004, and it now runs mostly as a weekly school club of around 12 weeks.
Does LEGO therapy work for autism?
The evidence is encouraging but not conclusive. LeGoff's own 2004 and 2006 studies reported gains in social competence that held over time, but they were run by the developer with unblinded raters. A 2021 systematic review of 19 studies called the approach promising while cautioning about small samples and low study quality. The biggest trial, I-SOCIALISE in 2023, randomized 98 schools in the United Kingdom and found teacher-rated social skills slightly better in the groups at 20 weeks, short of statistical significance, with little difference a year later. Treat it as a low-cost, well-liked activity with plausible benefit, not a proven treatment.
How do I run LEGO therapy at home?
Start with the rule set rather than the roles, since consistency is what makes shared building go well. With two children, use the engineer and builder roles and keep the brick supply yourself; with three, run the full structure and rotate roles each session. Pick a set that can be finished in one sitting, keep sessions to 45 minutes at a fixed time, and coach the process instead of the model. The one rule for adults: do not build, and do not make brick time conditional on behavior elsewhere, because the motivation to do the activity is the method.
What age is LEGO therapy for?
Most LEGO therapy groups run with participants from about six to the early teens, which matches the trial populations: LeGoff's 2004 study covered ages six to sixteen, and I-SOCIALISE recruited pupils aged seven to fifteen. Below six, the roles are usually too demanding and simpler turn-taking play fits better. In the mid-teens, interest in the activity matters more than age, and some groups run with older participants who like building. The practical test is whether the person wants to build and can hold a role for a few minutes, not the number on their birth certificate.
How is LEGO therapy different from play therapy?
LEGO therapy is a structured social skills group that happens to use bricks. Play therapy is a broader therapeutic approach in which play is the medium for expressing and working through feelings, usually led by a licensed therapist and often one to one. LEGO therapy has specific targets, a fixed structure, assigned roles and a visible end product, and it can be run by a trained teaching assistant in a school. Neither replaces the other, and neither replaces speech and language therapy or an assessment where those are what is actually needed.
Is LEGO therapy affiliated with the LEGO company?
No. LEGO is a registered trademark of the LEGO Group, and the method's name simply describes what it uses. Neither the published manual's approach as described here nor this page is endorsed by or connected to the company, which is also why the 2023 trial used the neutral name Play Brick Therapy. Practically, any reliable construction brick works, and training providers often prefer brick-based language for the same reason. If a service implies official backing from the toy company, that claim is worth questioning before you pay.
What are the goals of a LEGO therapy group?
Useful goals are practical and observable: asking another person for what you need, saying when an instruction was unclear, waiting through someone else's turn, accepting a build decision you disagree with, and finishing a shared project. Goals that aim at appearing less autistic, such as enforced eye contact or scripted social phrases, tend to produce masking, which autistic adults describe as draining and counterproductive. The group should also be enjoyable, since an activity nobody wants to attend is not doing anything. Reviewing goals every few weeks with the adult who runs the group keeps them honest.