The avatars are coming! But is it innovation in teacher education?

It was bound to happen here ….

For a couple of years now I’ve been studying the ‘innovation’ of game-like, avatar simulations in teacher preparation. The spark that lit my interest was a discussion at Teachers College during the Educating the Teacher Educators research project in which colleagues left me dumbfounded when they showed me some of the US programmes exploiting – to varying degrees of sophistication – avatars and simulation technologies in pre-service teacher education programmes. As with most things, it was only a matter of time before one turned up here in England and – hey presto – this term an email came my way advertising an ‘innovative’, avatar-based simulation designed to prepare student teachers to teach, talk to parents and win a top score on Candy Crush. OK, I made that last bit up. Anyway, here is one of their promotional videos from YouTube so you get the idea:

The email and the company’s website (which you can find here) boast that they have conducted a pilot with a large Education Faculty at an English university whose students and Dean seem to heap fulsome praise on it.

The first thing that struck me about the product as it was shown in the video was how crude it seemed technically. The avatars are pretty poorly rendered; they reminded me of the 1990s video games that I didn’t play. Moreover, the avatar seemed to be operated in real-time by someone the company calls a ‘simulation specialist’ – a person with a fairly limited range of voice acting skills, on the evidence of the video – who operated and voiced the figures in response to the live webcam feed of the student teacher teaching (via a connection that looked like it had some bandwidth problems). I might be wrong but I don’t think we’re talking cutting edge Artificial Intelligence here. Indeed, you have to book a time slot with the company so that the simulation specialist is online, presumably, and not auditioning for Emmerdale or a Shake ‘n’ Vac commercial. So in many ways, this product in the video reminded me of some of the early and technically unsophisticated American avatar simulations such as the one reported on in this news video from 2009:

Now these examples – eight years apart – are at the relatively low-tech end of the spectrum. There are some more technically advanced versions that do employ AI in producing avatar interactions with human subjects. They are often associated with inter-disciplinary research programmes that promote a ‘safe to teach’ approach to pre-service teacher education – an approach that seeks to eliminate all risk from teacher preparation whose advocates talk about not ‘letting loose’ new teachers on classes until they are ‘perfect’.

And there lies one of the problems with the approach in general (whether high or low-tech) which is that any form of teacher preparation that seeks to have ‘perfect’ new teachers is doomed to fail. So it’s an extremely good idea to plan; an extremely good idea to ‘practice’ and ‘rehearse’; but any kind of initial teacher education that believes it’s possible to produce ‘perfect’ new teachers who are guaranteed to be 100% safe (or even proportionately more safe than teachers from programmes that don’t use avatar simulations) don’t understand human beings never mind pre-service – or initial – teacher education.

That’s not to say that seeking to make teachers as ‘classroom ready’ as possible is a bad thing. However, promoting a technology on the basis that it will allow prospective teachers to ‘perfect’ their practice before they have entered the practice field of a school pre-supposes that the children, parents and complex situations they will encounter will be as controlled and predictable as, say, flying a passenger aircraft. Indeed, advocates of this general approach to teacher development often make the comparison with flight simulator training in the aviation industry where pilots learn how to fly an aircraft in the simulator before taking control in the cockpit of a real aircraft.

There are at least two critical responses to this analogy. First, while pilots can rely on sophisticated instruments that monitor wind, weather, global positioning and just about every aspect of the plane in flight, at the time of writing this blog, teachers cannot rely on similar monitoring of individual student functioning (e.g. blood pressure, eye movement, galvanic skin response, etc., etc. never mind cognition), parent functioning and the ‘weather’ of human relations in and out of school. Second, even with all the sophisticated monitoring systems and the high-tech simulation training that is available in the aviation industry, there are still situations that cannot be planned for and aren’t simulated: for example, when both engines on your aircraft fail within minutes of take-off and you have to land in the Hudson River or when your co-pilot locks you out of the cabin when you’ve gone for a toilet break. These situations may be the extremes that a pilot may encounter in their professional life but my point is that these situations arise out of either the unmonitored and unmonitorable aspects of the job or the entirely unpredicted and unpredictable. And it is these aspects that characterise teaching as public-facing (rather than technical instrument-facing), professional work.

Two further issues arise for me with the use of avatar simulations in teacher education one of which is a matter of principle. Becoming a teacher is – and should be – a humane process and one where we accept that teachers make mistakes (as human beings tend to, even of the professional variety). There is something for students as well as teachers (and the profession) to learn when we allow teachers to be fallible. We don’t plan on teachers making mistakes that damage people but we should plan on teachers making the kind of mistakes that allow them to improve both their practice and their values and beliefs. To quote Deborah Britzman from an early Harvard Education Review article, it is a ‘cultural myth’ (and a malign one) to persist in the view that there is a state of personal grace and technical perfection that is the final destination for the teacher. Do we wish to see situations where opportunities for teacher development have to be confined to virtual environments because the classroom must be preserved as a 100% safe space?

And secondly, if there are (and there are) those relatively mundane but important aspects of a teacher’s job that can be rehearsed in advance of professional practice (e.g. taking a register, handing out books, telling a child to stop playing with much more sophisticated avatar simulations on their phone), do teacher educators really have to book a time slot for their student teachers with a simulation specialist, rely on a good broadband connection for the webcam to work properly, and then try to interact with a crude avatar voiced by someone who is channelling South Park? Or do they simply ask the student teachers to form small groups and role play their own simulations that focus on these small aspects of practice? With the teacher educator perhaps also intervening in role? And then follow this up with supervised practice in school? In about the second week of the programme. The former option relies on a contract with the avatar simulation company; the latter doesn’t. In terms of their status as an approximation of practice, which is the more reliable?

In the US, these teacher preparation avatar simulations (especially the early ones) have also been criticised for perpetuating crude racial stereotypes. The avatars who have needed ‘behaviour management’ seem to have been disproportionately from African American and Latino backgrounds. And to compound this stereotyping, as you can see in the news video above from 2009, you then have a young white woman operating these ‘avatars of colour’ and opening themselves up to accusations of ‘performing black-face’. Surely what would be more useful for the young white women who still form the majority of the teaching profession in the US and the UK would be opportunities to learn about the different communities they will teach and some training in racial self-awareness. (Incidentally, both of these are offered in an outstanding doctoral-level course for prospective teacher educators at Teachers College led by Professor Felicia Moore Mensah).

So is this teacher education avatar simulation product an innovation?

I would say no. It might claim to be novel but in reality these technologies are almost ten years old in the teacher education setting and much older in digital media generally. Any anyway, in a public sector profession such as teaching, innovation has to mean more than fetishizing novelty or creating a new market for entrepreneurs selling us something we didn’t realise we needed. To use the OECD definition of public sector innovation (Daglio et al, 2015), some new public value has to be created; the way we prepare teachers has to be significantly enhanced for any justifiable extra cost. I would say at this stage, avatar simulations in teacher education don’t meet this test.

Is that always likely to be true? Probably not. A King’s colleague has been working on haptic simulation technologies for dental education for several years now so that dentists can be taught specialised procedures in touch-sensitive ways before working on patients. It has met a need and has likely reduced pain and discomfort. It is an innovation in dental education that is technology-driven but is patient-centred and humane in approach. But I’m not convinced student teachers need to make an appointment with a simulation specialist via a webcam (or even against a green screen background) to practise telling children to stop playing with their pencil cases.

Anyway, the response from the north American, critical teacher education community has been characteristically creative and, occasionally, as in this video made by Juan Carlos Castro, Chair of the Art Education department at Concordia University in Montreal, very funny (with thanks to my TC colleague Mary Hafeli for sending it my way):

Keep it REAL. Happy holidays!