npj Summary – An artificial intelligence approach for selecting effective teacher communication strategies in autism education

Like Comment
Read the paper

Autism spectrum condition (ASC) describes a broad number of neurodevelopmental conditions that can be expressed as an obsession with detail, compromised social abilities, and communication challenges. Individuals with ASC are more common in todays society, with 1 in 100 identifying with the condition, compared to 1 in 10,000 in the 1960s. Autism spectrum conditions have long term effects on an individual’s social life, relationships, education, and employment prospects.

Traditional strategies like verbal communication are not always as effective in autism education, so teachers employ a variety of interventions such as, pictures and symbols, gestures, physical prompts, or objects to communicate with students. A research team based at University College London sought to improve the educational outcomes for students in the spectrum, by exploring whether artificial intelligence might determine the most effective communication strategy in the classroom.

For this experiment, the researchers focused on the student’s responsiveness to instruction and included additional data about the participating students, for instance, age, gender, communication and cognitive development. The team deployed machine learning techniques to model teacher-student interactions and captured a total of 5460 interactions between teachers and a cohort of seven students. Based on the data recorded, a classifier or algorithm was developed that predicted students’ responses to the teacher’s communication strategies.

The researchers believe the model created has the potential to help teachers make the right decisions about communication strategies and contribute to improving the learning outcomes for children in autism education.

Please read An artificial intelligence approach for selecting effective teacher communication strategies in autism education by Vasileios Lampos, Joseph Mintz and Xiao Qu, published in npj Science of Learning.


Reference

Lampos, V., Mintz, J. & Qu, X. An artificial intelligence approach for selecting effective teacher communication strategies in autism education. npj Sci. Learn. 6, 25 (2021). https://doi.org/10.1038/s41539-021-00102-x

Poster Image by Polina Kovaleva from Pexels.

Gabrielle Ahern

Former Managing Editor, npj Science of Learning Community

I am based in Queensland, Australia, and have an interesting portfolio of skills, experience and qualifications in content creation, and science research.