Since the 1990s, 26 randomized studies have been conducted. Some show improvements in social communication—but the results are mixed.

Knowledge gap

What is the effect of music therapy for children and young people with ASD?

What did we still lack knowledge about?

• Does music therapy have effects beyond the treatment period?
• How does it affect the brain—not just behaviour?
• What is the impact on quality of life, participation, and mental health?

Where does the project fit in?
M4A is among the first studies to link clinical outcomes with brain imaging, in an effort to understand not just whether music therapy works, but how.

Executive summary

The study examined the effects of music therapy compared to play therapy in children with autism. The aim was to gain clearer insight into what actually works—and why. The study was designed as a randomized crossover trial with 68 children in Norway and Austria, where participants received both music therapy and play therapy. Outcomes were measured across several domains, including social communication, quality of life, mental health, and brain structure and function.

The results showed no clear differences between the two interventions on the main outcomes in the short term. At the same time, both interventions led to positive development over time. The study also found that music therapy influenced brain networks in ways that may help explain improvements in functioning and interaction.

Overall, the findings offer a deeper understanding of how interventions may work—not just whether they do—and add nuance to the evidence base for treating children with autism.

Methods

The study used a quantitative design and was conducted as a randomized crossover trial. The sample included 68 children aged 6–12 with an autism diagnosis—50 in Vienna and 18 in Bergen.

Data collection included a range of measures to capture a broad picture of outcomes:

• teacher-rated social communication
• questionnaires on quality of life and participation
• MRI scans of brain structure and function
• biological measures such as cortisol and microbiome

The children took part in both music therapy and play therapy, with play therapy serving as the comparison condition. Analyses included comparing the effects of the two treatments, examining changes over time, and exploring links between brain changes and behaviour.

The study also had some limitations. The sample size was smaller than planned, and there was variation between therapists and locations. In addition, movement during MRI scans affected data quality. Overall, this means the results should be interpreted with some caution.

Results

Main findings
No significant difference was found between music therapy and play therapy on the primary outcomes after three months. However, both interventions led to positive development over time, with improvements in both groups also seen at longer follow-up (up to 9 months).

Secondary findings
Some smaller but interesting differences emerged. There was a tendency toward better quality of life in the play therapy group, while music therapy was associated with fewer emotional difficulties. The differences were modest, and neither approach clearly outperformed the other.

Brain data provided additional insight. Music therapy led to changes in brain connectivity, particularly a reduction in “overconnectivity” between regions linked to sound, movement, and language. This may indicate less “noise” and more efficient communication in the brain. These changes were also associated with improvements in communication.

The findings point to two possible mechanisms:

• more efficient organisation of brain networks (less unnecessary activity)
• better integration between senses such as sound, movement, and emotion

There was considerable individual variation in response. Many children preferred music therapy, but the effects were not the same for everyone. What works varies from child to child.

Conclusion

The findings suggest that music therapy is not necessarily more effective than other interventions, but that it may produce different kinds of effects—particularly in how the brain functions. The study helps bridge therapy and neuroscience, and highlights the need to measure more than just symptoms. Participation, quality of life, and lived experience should play a larger role as key outcomes.

For practitioners, this means interventions need to be tailored to the individual child—there is no one-size-fits-all solution. Music therapy may be especially relevant for some children, and underscores the importance of relationship, engagement, and experience alongside symptom reduction.

For decision-makers, the findings highlight the need for a broader understanding of impact, including functioning and quality of life. There is also a clear case for flexible, individualised services.

Further research

Future research should focus on identifying which children benefit most from different interventions, exploring alternative ways of measuring brain function, and strengthening user involvement in deciding what outcomes to measure and prioritise.