STEM learning with robots
How do students learn computational thinking? These friendly robots are providing the key in research into student learning.
There’s a look of pride on the face of a young Indigenous student at Maitland Lutheran School as she echoes the Narungga word for ‘knee’ that she’s just programmed this NAO robot to learn.
It’s a 21st century breakdown of the ‘heads and shoulders’ song, and it’s involved a group of students actively leading the programming of the robots to dance and sing, while they simultaneously teach and learn the sleeping language of Narungga.
Until now, the world’s only Narungga speaker was just one person, Tania Wanganeen, who learnt the language based on records that were left by the German missionaries who worked on the Yorke Peninsula in South Australia, just over 100km west of Adelaide.
Robots in classrooms
The robots came to the school as part of a collaborative research project to understand the impact of the robots on student learning.
Teachers from Independent Schools in South Australia have played host to the programmable humanoid ‘NAO’ robots in classrooms as part of the project, involving partners from Swinburne University in Melbourne, the University of Queensland, the Queensland University of Technology, and the Association of Independent Schools of South Australia.
The project provided researchers with insights into the ways in which the robot could be integrated into the curriculum – with classes exploring everything from language learning to design and maths using the robots.
STEM learning through student-led teaching
Teachers pitched their idea to gain access to the robots. This access ranged from 2 months to 9 months.
The researchers then gathered information on the STEM learning through journals, surveys and follow up interviews, identifying themes of curiosity, challenge, communication, collaboration and critical thinking, which they have integrated into a model presented at the Australian Computers in Education Conference (ACEC) in 2016.
Few of the teachers had any background in teaching Digital Technologies before the robots arrived in their classrooms, so they needed to learn alongside the students.
While Maitland Lutheran School embedded the robots in both Aboriginal history and digital technologies, another group of Year 9 & 10 students at Immanuel College used the robots in the German classroom. The robots interacted in German to support younger students to learn the correct German possessive pronoun for parts of the body.
“The students were limited in what they could program the robot to do using the drag and drop programming, and realized they would need to program in Python, a general purpose programming language used in industry,” says Monica Williams, an Educational Consultant at Association of Independent Schools of South Australia.
“So they actually went to Google to find out how to program this in Python – and then shared what they had learnt with the teacher.”
“I believe the best outcome is when students treat the robot like a peer student that is learning with them,” says Marie Boden, an academic from the School of Information Technology and Electrical Engineering at the University of Queensland.
“In another project I was telling the designers and engineers not to create a robot that already does everything – but one that can be taught and engineered further,” she adds.
Therese Keane, Deputy Chair of the Department of Education at Swinburne University in Melbourne, says the students wanted to go beyond what they have learned.
“They are thinking laterally and pushing themselves. They are trying to work out how the robot can do what they want them to do. And the process gives them positive feedback by being a 3D object and giving them immediate natural feedback on what they’ve tried to do.”
“Because of the human-like qualities of the NAO robot, the students saw the robot as a little sister, and this made them persist for longer, and more resilient when things did not work out at first the way they expected,” adds Monica Williams.
“There have been lots of surprises. What students have been able to do far surpasses what educators expected them to be able to do. If students have access to cutting edge technology, they can quickly master complex software. It has shown the importance of having sophisticated technology as part of student learning.”
Learning to teach
“Each teacher and user used the robots in very different ways,” says Monica Williams.
“They all get so involved – they feel really successful and they relate to the technology in a very different way than they did when they were interacting with a computer program.
“For example: it’s okay if the robot doesn’t do exactly what they want them to do from the start – they see it as them teaching the robot.
“I think it will have an impact on future pedagogy. They talk about flipped classroom a lot but here it is student-led learning – it’s the perfect learning situation in groups.”
Engagement and deep learning
The researchers were completely open with how the classrooms used the robots, how deeply they embedded their use into the curriculum.
“We wanted to know with our research project what they would do and how they would use it in the curriculum,” says Therese Keane.
“We left it to the teacher to convince our team that they were going to use the robot in a positive educational way in their expressions of interest. We wanted to see how the robot would be used. Our interest was in seeing how the robot could be used to develop their knowledge and understanding of Digital Technologies, in STEM learning and how it could be integrated into other learning areas,” says Chris Chalmers.
The outcomes were surprisingly consistent, whether it was in an early learning centre, a more experienced teacher and also across genders.
“The teachers used the same language – the same words – and we were able to come up with a model that sums up how they use the robots in the classroom. In 2016 we were interested in seeing if this model fit. We found that it was the perfect way of summarising what led to the engagement, what led to the deep learning and it supported and extended the data from 2015.”
As well as extending the research model, the University of Queensland are working on the design of a new humanoid robot that will work as a support for the teacher.
“The humanoid robots that exist now are very cute, but we’d like something that could also support the teachers,” says Marie Boden.
“It has been really surprising that some of the teachers that had no Digital Technologies experience – who just thought they would bring it as show and tell – have gotten pulled in by the students’ interest and how they have changed their pedagogies because they have been learning along with the students.”
– Heather Catchpole