Brain-Computer Interface Training of mu EEG Rhythms in Intellectually Impaired Children with Autism: A Feasibility Case Series.

Researchers tested brain training using special EEG equipment in 7 children with autism who also had intellectual disabilities. Four children learned well during the training and showed improvements in behavior and brain activity patterns. Three children didn't learn as well and showed little improvement. Children who moved around too much during brain testing were less likely to benefit.

The study shows this type of brain training might help some children with autism and intellectual disabilities, but more work is needed to figure out which children would benefit most.

This case series examined the feasibility of brain-computer interface training targeting mu EEG rhythms in seven intellectually impaired children with autism (ages 6-8, mean IQ 70.6). Children received neurofeedback training after behavioral preparation with conditioned auditory reinforcers. Four children showed positive learning trends during training ('learners') while three did not ('non-learners'). Learners demonstrated behavioral improvements and increased mu rhythm suppression compared to non-learners, who showed minimal EEG or behavioral changes.

Artifact-creating behaviors during EEG testing were more frequent in non-learners. The study suggests mu-rhythm neurofeedback may benefit some intellectually impaired children with autism when appropriate behavioral preparations are implemented, though identifying ideal candidates remains challenging.

Mu-rhythm neurofeedback may be feasible for some intellectually impaired children with autism when appropriate behavioral preparation is provided. However, careful candidate selection is crucial, as children with greater behavioral challenges during EEG testing appear less likely to benefit. Further research is needed to develop better selection criteria and optimize protocols.

Very small sample size (n=7) limits generalizability. Case series design without control group prevents causal conclusions. High variability in individual responses. Artifact contamination in EEG data, particularly for non-learners, may affect result validity.

Prior studies show that neurofeedback training (NFT) of mu rhythms improves behavior and EEG mu rhythm suppression during action observation in children with autism spectrum disorder (ASD). However, intellectually impaired persons were excluded because of their behavioral challenges. We aimed to determine if intellectually impaired children with ASD, who were behaviorally prepared to take part in a mu-NFT study using conditioned auditory reinforcers, would show improvements in symptoms and mu suppression following mu-NFT. Seven children with ASD (ages 6-8; mean IQ 70.6 ± 7.5) successfully took part in mu-NFT.

Four cases demonstrated positive learning trends (hit rates) during mu-NFT (learners), and three cases did not (non-learners). Artifact-creating behaviors were present during tests of mu suppression for all cases, but were more frequent in non-learners. Following NFT, learners showed behavioral improvements and were more likely to show evidence of a short-term increase in mu suppression relative to non-learners who showed little to no EEG or behavior improvements. Results support mu-NFT's application in some children who otherwise may not have been able to take part without enhanced behavioral preparations.

Children who have more limitations in demonstrating learning during NFT, or in providing data with relatively low artifact during task-dependent EEG tests, may have less chance of benefiting from mu-NFT. Improving the identification of ideal mu-NFT candidates, mu-NFT learning rates, source analyses, EEG outcome task performance, population-specific artifact-rejection methods, and the theoretical bases of NFT protocols, could aid future BCI-based, neurorehabilitation efforts.