Learning will increasingly be adjusted to individual learner characteristics.

– Pushing the Frontiers with Artificial Intelligence, Blockchain and Robots, OECD Digital Outlook 2021

In an education context, robots could take on a number of social roles, such as tutor or peer, each of which gives rise to certain behavioural expectations.

 – Robots Education Peers in a Situated Primary School Study: Personalisation Promotes Child Learning

 Learning is going to get a lot more efficient and effective with developments in educational technology. Enabling technology is strengthening and making possible the fulfilment of long-time desires to create bespoke learning experiences at scale. Though cost and accuracy of such technology are issues which still need working out, the journey towards mass customization is well underway.

The OECD Digital Outlook 2021 discusses exciting developments in educational technology, along with its limitations and potential. It is now becoming possible to get real-time data on individual learners for the purposes of giving them what they need when they need it so they learn better.

According to the aforementioned report, technology can be used to enhance learning in the following ways. Teachers can understand themselves better. Teachers can understand students better. Students can understand themselves better. Both teachers and students would be able to circumvent their own limitations to make the learning experience richer. Technological developments show a lot of promise for special needs students.

Technology or as the report terms it, “teachology” can be classified in three main groups. These are, artificial intelligence, robotics and learning analytics.

Artificial intelligence (AI) would allow the creation of personalised learning profiles in order to make personalized recommendations to students. Such AI consists of the following three stages; detect, diagnose and act. At its highest level, AI would be able to, without a human-in-the-loop, through use of advanced sensors detect or measure various indicators of learning. AI would be able to monitor how engaged students are, through monitoring, among other things, heart rate, eye movement and facial expression. Such AI would be able to calibrate the learning process to increase or maintain engagement. It would also be able to give feedback as students progress in their task. In theory, this should free up teacher time and effect maximum personalisation for students.

Inge Molenaar who wrote the chapter on how AI will complement teachers adapted the six levels of automation towards the self-driving car formulated by the Society of Automated Engineers for use of AI in an education context. The education version consists of the following 6 levels: Teacher Only (no technology), Teacher in Full Control (Technology provides supportive information), Partial Automation (Teacher monitors as technology controls specific tasks), Conditional Automation (Technology controls broader set of tasks and signals when teacher control is needed), High Automation (Teacher Control and Monitoring not required for specific tasks) and Full Automation (no human-in-the-loop).

She suggests that existing technologies “mostly fall under the first three levels of automation” and that they are quite effective in teaching “foundational skills” across subjects. She distinguishes foundational skills from “more complex skills such as problem solving, self-regulation and creativity”.

She adds that at present, though there are technologies which are able, for the large part, to perform on their own with teachers only monitoring, these are more for “structured domains such as Math and Science” and “few solutions exist for unstructured domains”.

We cannot say with certainty how much AI would be able to achieve even in the near future but we do know that stakeholders would be engaged in the process of development. During such discussions, one critical question would be what exactly students absolutely need which can only be gotten from humans. In an ideal situation, there would be perfect complementarity with supercharged learning as a result.

Robotics refers to the use of robots in learning. Such robots can teach or be taught by students so that students are able to reinforce their own learning. They are called social robots because they look, sound and behave like humans and are suited for social settings. There have been experiments which tested the use of robots in learning. For example, Baxter and his colleagues (2017) found that when a social robot was used in the classroom as a peer to students, 7-8-year-olds were able to better learn a task with which they were unfamiliar. The OECD report notes that social robots are especially useful for second-language learning because students may feel it is fun to engage in conversation with robots. It also cites studies which found that social robots were more effective than anonymous reporting to surface cases of bullying in schools because students were comfortable sharing such details with robots.

A cutting-edge application of robots in the trial stage, is what is known as “telepresence robots”. This is when the robot is present with learners and the teacher is at a remote location. The teacher controls the robot and the robot informs the teacher of goings-on in the classroom through sensors and other fittings.

Learning analytics refers to the processing of data generated by students while they are learning. It can be used in two main ways. The first is in early warning systems. Processed data on students could highlight students who are at risk of underperforming or other undesirable outcomes. This could then allow for measures to help these students get back on track. The second way analytics is used, is to personalise learning for students. Such data would be displayed on dashboards for easy access by students and teachers.

While learning is an individual and personal process, it is also at the same time, very much a social process. While students would feel very empowered and excited to witness their own progress because of real-time customization, they would eventually want to share the joy of that progress with friends. Some things are better caught than taught.

Also, we learn for the purposes of flourishing in the societies we live in. At present, given limitations of human and technological ability, collective needs may at times, take precedence over individual needs and this is not a bad thing because, we feel good knowing that we are considering others in our actions. This is part and parcel of belonging.

With technological advancements, there may be fewer trade-offs between individual and collective needs. To the extent that the meeting of individual needs allows for greater and richer participation in the collective, personalisation for the individual would also be good for the group.

The greatest good can now be for everyone.

The Brain Dojo