Robots in the classroom

Sofia Serholt works in the Computer Science & Engineering department at Chalmers University of Technology and is a member of the LIT community in Gothenburg. She is one of the leading European researchers in the area of robots and education. In this interview, Sofia talks with Neil Selwyn about her recent work, and the growing interest being shown in AI and education.

NEIL: So, first off, for the uninitiated, exactly how are robots being used in schools? What sort of technologies are we talking about here?

SOFIA: A big issue in education right now is programming robots – using robots as tools to learn different programming languages. However, my research is mostly focused on robots that actually teach children things and/or learn from them, and perhaps interact with them socially. So, these are humanoid robots that are similar to humans in different ways. Then, we also see robots that can be remotely controlled in classrooms – like Skype on Wheels – and these can be used by children who can’t be in the classroom for different reasons such as if they’re currently undergoing cancer treatment. We have some very rare situations where teachers have remotely controlled robots to be able to teach a class as well. So, that’s basically how the field is right now.

NEIL: Now, this all sounds really kind of futuristic and interesting in theory, but in practice, one of the main issues is how people actually accept these machines, and how they gain trust in them. What have you found out about people’s reactions to these technologies in the classroom?

SOFIA: When it comes to the research that I’ve done in the classroom I have put actual robots in the classroom for several months and studied interactions with them. However, I haven’t done any rigorous research about acceptance in those cases. However, I do see that children are generally optimistic and positive about it … outwardly anyway. And teachers, of course, appreciate having a researcher there and being part of a project, and it’s fun and interesting. But, teachers haven’t been very involved in the actual robot and the interactions. Instead it’s been on the sidelines of what they’re doing. When I talk to teachers and students that are not part of any study, I can see that students are concerned about certain things that robots can do or what they shouldn’t do. And, the privacy issue is one thing that they are very concerned about. For example, they don’t want to be recorded by a robot, which might be necessary for it to be able to interpret different things about the person it’s interacting with. Also children don’t want to be graded by a robot. And, this is something that also teachers resonate with, so they don’t want to give away their authority in that sense. So, there are a lot of open questions right now – for example, how this technology might impact children in the long run. We don’t know very much about this. We know about our children have toys and grow up with toys. But these robots are a new thing – kind of like a social interaction partner that is not really human, but does kind of mirror human behaviour. And also, it has certain restrictions, like it might not be able to speak with a humanlike intonation. So how does that affect children if they were to grow up with these robots? In reality that’s a kind of study we can’t conduct.

NEIL: So, can you just tell us a bit about the research you’ve done yourself on robots in the classroom? 

SOFIA: I’ve looked at interaction breakdowns. In one study I selected situations or instances where children either became notably upset or they became inactive when interacting with the robot. Not because they were bored with the actual game … that didn’t actually happen because it was very engaging and fun. But these were instances when students couldn’t do anything – they couldn’t proceed – and also when they started doing other stuff in the room, or began to talk to their friends instead of working on the topic. And, what I saw was that when the robot doesn’t understand what the child is saying, it generates a situation where on the one hand you have a robot who can express a lot of stuff and tell you what to do. But, if the child can’t ask the robot a question or show uncertainty, it creates a very difficult situation when they’re alone with this robot. And this is what leads to these breakdowns – when students need help from the outside. I looked at six randomly picked students over the course of the three-and-a-half months that I was in the school. And out of those sessions, I spotted 41 breakdowns. Some children were very upset about not understanding what to do, and not being able to move on. Some got really angry at the robot for disrupting and destroying their strategy that they were using.  But, I think the worst case was when the students who’ve actually felt that they were not good enough and they put the blame on themselves. You know, because robots and computers have a lot of authority in the sense. You know, you don’t think that your calculator lies to you?

NEIL: Yeah, yeah.

SOFIA: You trust in your calculator more than your own calculations. And, I spotted this similar tendency here. If the robot broke down when I was with it then it must not like me. And, having to deal with those kinds of situation is ethically problematic, I think.

NEIL: So, is that a design problem? Can we design robots to suddenly be a bit more imprecise, or as you say, not to kind of break down the magic between the student and the machine?

SOFIA: I think we could have obviously accomplished a lot more than what we did in that study.  We used a teaching robot and there were a lot of outside things that affected it. You know, sunlight affects it, heat in the room affects it, how long it’s been going on affects the ability of the robot to work like it should. But, nevertheless we have to kind of ask ourselves how far along that road can we go to uphold this illusion? The illusion that this is a sentient being in the eyes of the student. I think that’s a question for philosophy and ethics really.

NEIL: So, you’ve – you’ve moved very quickly from looking at these things as teaching and learning technologies to ethical questions. These are big kind of issues to be grappling with. So, what are the main ethical questions that we need to be asked here? We’ve got privacy …

SOFIA: Yeah, and we have issues of responsibility.  For example, who is responsible for the robot? We see a lot of companies developing these technologies and selling them. However, where does their responsibility end and where does the teacher’s responsibility begin? And, according to teachers, they want to be the responsible party in terms of what’s going on in the classroom. However, they do feel at the same time that they can’t have this responsibility if they can’t monitor what’s going on. So, the idea with a teaching robot is that it’s supposed to work autonomously, and it’s not supposed to be under the control of teachers. And, often times the teachers don’t know even how it works, right, so they can’t control it.

But one teacher asked how this benefits them … because if they have to walk around and keep an eye on the robot all the time, then what kind of sacrifice is this for their roles as teachers? And, we also have the inevitable fact that robots do break and robots don’t support physical interaction as much as we are led to believe. Because they look humanoid, it is tempting to think you can shake its hand, you can give it a high five, you can give it a hug … but this usually doesn’t work unless the robot is programmed to go along with this.

NEIL: So, this issue of is the robot going along with it leads me to think about questions of deception. If the robot is mimicking certain behaviour, is that an ethical issue as well?

SOFIA: Well, I think it might become an issue if robots have these kind of social interaction features. Of course, there is a level of deception in that, because they’re not social. You can erase the program, you can accidentally erase a log about one child and then the robot won’t remember that child anymore. That’s a big issue that I think we’re going to be seeing a lot more of. 

NEIL: So, just backtracking from the ethics for a second, one of the things that spring into mind is why on earth should we be using these machines if there are all these issues. Presumably there are kind of very strong learning and teaching rationales for using robots in the classroom. What sort of things do we know about the learning can take place around a robot? 

SOFIA: We have certain indications that robots are preferred by students over virtual agents –  for example, intelligent tutoring systems that have a virtual agent with different levels of animation in the agent. So, the more humanlike the embodiment of the robot is, the more physical it is, the better the learning outcomes. However, these are not long-term field studies that I’m talking about here.  These findings come from very controlled experiments, often not even with children. And, so we honestly don’t know too much about the learning outcomes. In my study, I had a robot that taught geography and map-reading to children, and also sustainability issues. The learning goal was that the children should be able to reason about sustainability and the economic issues involved, the social issues involved, and it’s a complex interaction. So, we didn’t see any learning outcomes in that regard. In map-reading there was a slight learning improvement, but not as much as one would hope after a month’s worth of interactions with this robot.

 NEIL: So, this is very future focused research, it’s a very future focused area of education, and doing research in this area must be really, really tricky. And also, there’s a lot of hype in this area as well. So, looking forward in the future, what do you realistically think we’ll see in 20 years’ time? And, what is actually hype?

SOFIA: I think as soon as you talk about the social aspects of interaction, then we have a problem. And if we talk about AI as being very clever at certain specific tasks, then yes, we have this already – this is technology that is coming. But, if we talk about a general social intelligence that’s supposed to make its own decisions and deductions based on how you are, and it gets to know you in human terms, and it gets to reason and think, then I’m not sure I believe this is going to happen at all. This would require a very complex form of programming and machine learning, so I guess we’ll see … I’m a bit reluctant to answer that question, because ‘who knows’?

NEIL: Now, I just wanted to finish on a nice easy question. It’s often said that robots and AI actually raise this existential question of ‘what does it mean to be human in a digital age’? I was wondering if your work has led you to any such insights? What will it mean to be human in the 21stcentury? And, what implications might this have for education?

SOFIA: I think we’re going to start to see that there is something else to human nature that technology might not be able to fill. The question is how we want to proceed knowing this. And children are what we define as a vulnerable group in society that we have some sort of duty of care towards. And, if we see all these problems with technology, if we see problems and potential suffering, then maybe we should talk about those issues and not just sweep them under the carpet. I don’t think there’s going to be any revolutionary situation where you see that robots somehow make us question our own sense of being in the world. But, I do think that if we interact with them too much, then we’re going to have problems knowing what we are. So it’s important that we don’t put this technology in the hands of children who are too young to be able to critically assess what’s going on.

The ethical dilemma of the robot teacher

The rise of automated teaching technologies

We need to talk about robots.

Specifically, we need to talk about the new generation of AI-driven teaching technologies now entering our schools. These include various ‘autonomous interactive robots’ developed for classroom use in Japan, Taiwan and South Korea. Alongside these physical robots, are the software-based ‘pedagogical agents’ that now provide millions of students withbespoke advice, support and guidance about their learning. Also popular are ‘recommender’ platforms, intelligent tutoring systems and other AI-driven adaptive tutoring – all designed to provide students with personalised planning, tracking, feedback and ‘nudges’. Capturing thousands of data-points for each of its students on a daily basis, vendors such as Knewton can now make a plausible claim to know more about any individual’s learning than their ‘real-life’ teacher ever could.

One of the obvious challenges thrown up by these innovations is the altered role of the human teacher. Such technologies are usually justified as a source of support for teachers, delivering insights that “will empower teachers to decide how best to marshal the various resources at their disposal”. Indeed, these systems, platforms and agents are designed to give learners their undivided attention, spending indefinitely more time interacting with an individual than a human teacher would be able. As a result, it is argued that these technologies can provide classroom teachers with detailed performance indictors and specific insights about their students. AI-driven technology can therefore direct teachers’ attention toward the most needy groups of students – acting as an ‘early warning system’ by pointing out students in most need of personal attention.

On one hand, this might sound like welcome assistance for over-worked teachers. After all, who would not welcome an extra pair of eyes and expert second opinion? Yet rearranging classroom dynamics along these lines prompt a number of questions about the ethics, values and morals of allowing decisions to be made by machines rather than humans. As has been made evident by recent AI-related controversies in healthcare, criminal justice and national elections, the algorithms that power these technologies are not neutral value-free confections. Any algorithm is the result of somebody deciding on a set of complex coded instructions and protocols to be repeatedly followed. Yet in an era of proprietary platforms and impenetrable coding, this logic typically remains imperceptible to most non-specialists. This is why non-specialist commentators sometimes apply the euphemism of ‘secret sauce’ when talking about the algorithms that drive popular search engines, news feeds and content recommendations. Something in these coded recipes seems to hit the spot, but only very few people are ‘in the know’ over the exact nature of these calculations.

This brings us to a crucial point in any consideration of how AI should be used in education.

If implementing an automated system entails following someone else’s logic then, by extension, this also means being subject to their values and politics.

Even the most innocuous logic of [IF X THEN Y] is not a neutral, value-free calculation. Any programmed action along these lines is based on pre-determined understandings of what X and Y is, and what their relation to each other might be. These understandings are shaped by the ideas, ideals and intentions of programmers, as well as the cultures and contexts that these programmers are situated within. So key questions to ask of any AI-driven teaching system include who is now being trusted to program the teaching? Most importantly, what are their values and ideas about education? In implementing any technological system, what choices and decisions are now being pre-programmed into our classrooms?

The ethical dilemma of robot teachers

The complexity of attempting to construct a computational model of any classroom context is echoed in the ‘Ethical Dilemma of the Self-Driving Car’. This test updates a 1960s’ thought experiment known as ‘the Trolley Dilemma’ which posed a simple question: would you deliberately divert a runaway tram to kill one person rather than the five unsuspecting people it is currently hurtling toward? The updated test – popularised by MIT’s ‘Moral Machine’ project – explores human perspectives on the moral judgements made by the machine intelligence underpinning self-driving cars. These hypothetic scenarios involve a self-driving car that is imminently going to crash through a pedestrian crossing. The car can decide to carry on the same side of the road or veer onto an adjacent lane and plough into a different group of pedestrians. Sometimes another option allows the car to self-abort by deciding to swerve into a barrier and sacrifice its passengers.

Unsurprisingly, this third option is very rarely selected by respondents. Few people seem prepared to ride in a driverless car that is programmed to value the lives of others above their own. Instead, people usually prefer to choose one group of bystanders over the other. Contrasting choices in the test might include hitting a homeless man as opposed to a pregnant woman, an overweight teenager or a healthy older couple. These scenarios are complicated further by considering which of these pedestrians is crossing on a green light or jaywalking. These are extreme scenarios, yet neatly illustrate the value-laden nature of any ‘autonomous’ decision. Every machine-based action has consequences and side-effects for sets of ‘users’ and ‘non-users’ alike. Some people gets to benefit from automated decision-making more than others, even when the dilemma relates to more mundane decisions implicit in the day-to-day life of the classroom.

So what might an educational equivalent of this dilemma be? What might the ‘Ethical Dilemma of the Robot Teacher’ look like? Here we might imagine a number of scenarios addressing the question: ‘Which students does the automated system direct the classroom teacher to help?’. For example,

who does the automated system tell the teacher to help first – the struggling girl who rarely attends school and is predicted to fail, or a high-flying ‘top of the class’ boy?

Alternately, what logic should lie behind deciding whether to direct the teacher toward a group of students who are clearly coasting on a particular task, or else a solitary student who seems to be excelling. What if this latter student is in floods of tears? Perhaps there needs to be a third option focused on the well-being of the teacher. For example, what if the teacher decides to ignore her students for once, and instead grab a moment to summon some extra energy?

#1 Who should the robot help next?

#2 Who should the robot help next?

The limits of automated calculations in education

Even these over-simplified scenarios involve deceptively challenging choices, quickly pointing to the complexity of classroom work. Tellingly, most teachers quickly get frustrated when asked to engage in educational versions of the dilemma. Teachers complain that these scenarios seem insultingly simplistic. There are a range of other factors that one needs to know in order to make an informed decision. These might include students’ personalities and home lives, the sort of day that everyone has had so far, the nature of the learning task, the time of academic year, assessment priorities, and so on. In short, teachers quickly complain that their working lives are not this black-and-white, and that their professional decisions are actually based on a wealth of considerations.

This ethical dilemma is a good illustration of the skills and sensitivities that human teachers bring to the classroom setting. Conversely, all the factors that are not included in the dilemma point to the complexity of devising algorithms that might be considered appropriate for a real-life classroom. Of course, many system developers consider themselves well-capable of being able to provide sufficient measurement of thousands (if not millions) of different data-points to capture this complexity. Yet such confidence of quantification quickly diminishes in light of the intangible, ephemeral factors that teachers will often insist should be included in these hypothetical dilemmas. The specific student that a teacher opts to help at any one moment in a classroom can be a split-decision based on intuition, broader contextual knowledge about the individual, as well as a general ‘feel’ for what is going on in the class. There can be a host of counter-intuitive factors that prompt a teacher to go with their gut-feeling rather than what is considered to be professional ‘best practice’.

So, how much of this is it possible (let alone preferable) to attempt to measure and feed into any automated teaching process? A human teacher’s decision to act (or not) is based on professional knowledge and experience, as well as personal empathy and social awareness. Much of this might be intangible, unexplainable and spur-of-the-moment, leaving good teachers trusting their own judgement over what a training manual might suggest that they are ‘supposed’ to do. The ‘dilemmas’ just outlined reflect situations that any human teacher will encounter hundreds of time each day, with each response dependent on the nature of the immediate situation. What other teachers ‘should do’ in similar predicaments is unlikely to be something that can be written down, let alone codified into a set of rules for teaching technologies to follow. What a teacher decides to do in a classroom is often a matter of conscience rather than a matter of computation. These are very significant but incredibly difficult issues to be attempting to ‘engineer’. Developers of AI-driven education need to tread with care. Moreover, teachers need to be more confident in telling technologists what their products are not capable of doing.

The two ‘dilemma’ images were illustrated using graphics designed by Katemangostar / Freepik

Author: Neil Selwyn

Neil Selwyn is a Professor in the Faculty of Education, Monash University and previously Guest Professor at the University of Gothenburg. Neil’s research and teaching focuses on the place of digital media in everyday life, and the sociology of technology (non)use in educational settings.

@neil_selwyn is currently writing a book on the topic of robots, AI and the automation of teaching. Over the next six months he will be posting writing on the topic in various education blogs … hopefully resulting in:  Selwyn, N. (2019)  Should Robots Replace Teachers? Cambridge, Polity