Learning programming is a difficult task that relies on a wide range of other skills, such as reading, writing, and memory, which can all be affected in children with SEND. Researchers have been trialling different methods of teaching programming through robotics to overcome some of these issues.
Robots and SEND
In recent years, robots have been developed as assistants or ‘teachers’ in the classroom to support children with SEND. However, robots can also be used as a highly motivating tool to develop children’s programming abilities. Three recent studies have taken different approaches to using robots in terms of the hardware, software, and teaching methods they employed. All three provide promising avenues for future research, but also highlight specific strategies for making programming lessons more inclusive.
Reducing the text required to produce programs
A key barrier to engaging with programming is a student’s literacy level. Studies by Lahav, Talis, Shelkovitz, and Horen (2019) and González-González et al (2019) addressed this issue by using symbols to aid children with SEND to program their robots.
González-González et al used the KIBO robot and software with children with Down syndrome (DS). This system involves the robot moving along interlocking wooden blocks on which there are symbols corresponding to different movements. Children put the blocks together to make a program and then scan the barcodes on the blocks using the scanner embedded in the robot. The program is sent to the robot which then moves as directed along the blocks. Of the seven participants with DS involved in the study (cognitive age 3-6 years), five of them understood the KIBO blocks and could program basic sequences by the end of the training. Four of them could produce the program independently.
Lahav et al used the EV3 robot built from the LEGO Mindstorms EV3 flexible kit, alongside a free visual programming language app called Kinderbot, which was developed especially for young children. They recorded the programming abilities and descriptions of the robot behaviour in two primary school children with high functioning Autism Spectrum Disorder (ASD). While there were some difficulties in producing more complex programs involving subroutines, the children did begin to use more technological descriptions of the robot and its movements after the training.
Using ‘unplugged’ activities and modelling
Another key feature of the study by González-González et al was the use of an ‘unplugged’ activity, in which the students put together programs using cards with symbols before engaging with the wooden blocks and robot. This allowed them to concentrate on the programs without the distraction of the robot and scanner. Using these types of tasks has also been encouraged when teaching all children programming, showing that pedagogical strategies used to assist those with lower abilities can be beneficial to all.
In a study by Knight, Wright, Wilson and Hooper (2019), a ‘Model-Lead-Test’ strategy was used with children with ASD and challenging behaviour aged 15-18 years. This involved first modelling block-based programming using an app (OzoBlockly) to control an Ozobot Evo robot. They then worked with the students to produce the same program, before allowing the students to do the task independently. Using this approach, they found that the three participants were all able to produce explicitly-taught programs. Most importantly, they were able to generalise their learning to produce different programs that were directed by the teacher but not explicitly taught, and also programs that were self-directed.
Developing research and further practice
The studies are all based on work with a small number of children who were taught individually. While this is beneficial for testing principles in research, it will be necessary to scale up the approaches to make them useful for teachers working with larger groups. Using strategies that can benefit all children, not just those with the lowest ability, will aid teachers in delivering the computing curriculum effectively for all learners.
There is also a challenge when it comes to dealing with text-based coding, which is important for teaching computing later in the curriculum. Catherine Elliot, an eLearning consultant with Sheffield County Council’s eLearning service, has some great ideas about how to approach this. You can see an interview with her on the Computing at School website or through #include on CAS TV. There are plenty of other resources for teaching computing to students with SEND on the CAS website. Other good banks of resources can be found through SEND computing and Barefoot Computing, including lesson plans for working with robots and other hands-on activities.
CS González-González, E Herrera-González, L Moreno-Ruiz, N Reyes-Alonso, S Hernández-Morales, MD Guzmán-Franco, and A Infante-Moro (June 2019). Computational Thinking and Down Syndrome: An Exploratory Study Using the KIBO Robot. In Informatics (Vol. 6, No. 2, p. 25). Multidisciplinary Digital Publishing Institute.
VF Knight, J Wright, K Wilson, and A Hooper (2019). Teaching Digital, Block-Based Coding of Robots to High School Students with Autism Spectrum Disorder and Challenging Behavior. Journal of autism and developmental disorders, pp.1-14.
O Lahav, V Talis, and R Shekovitz (June 2019). High-Functioning Autistic Children Programming Robotic Behavior – A Case Study. In EdMedia+ Innovate Learning (pp. 1374-1379). Association for the Advancement of Computing in Education (AACE).