Social robots for English language teaching

Posted: March 21, 2022 in ed tech, investment, vocabulary
Tags: , , , ,

In May of last year, EL Gazette had a story entitled ‘Your new English language teacher is a robot’ that was accompanied by a stock photo of a humanoid robot, Pepper (built by SoftBank Robotics). The story was pure clickbait and the picture had nothing to do with it. The article actually concerned a chatbot (EAP Talk) to practise EAP currently under development at a Chinese university. There’s nothing especially new about chatbots: I last blogged about them in 2016 and interest in them, both research and practical, dates back to the 1970s (Lee et al., 2020). There’s nothing, as far as I can see, especially new about the Chinese EAP chatbot project either. The article concludes by saying that the academic behind the project ‘does not believe that AI can ever replace a human teacher’, but that chatbots might offer some useful benefits.

The benefits are, however, limited – a point that is acknowledged even by chatbot enthusiasts like Lee et al (2020). We are some way from having chatbots that we can actually have meaningful conversations with, but they do appear to have some potential as ‘intelligent tutoring systems’ to provide practice of and feedback on pre-designated bits of language (especially vocabulary and phrases). The main benefit that is usually given, as in the EL Gazette article, is that they are non-judgemental and may, therefore, be appropriate for shy or insecure learners.

Social robots, of the kind used in the illustration for the EL Gazette story, are, of course, not the same as chatbots. Chatbots, like EAP Talk, can be incorporated into all sorts of devices (notably phones, tablets and laptops) and all sorts of applications. If social robots are to be used for language learning, they will clearly need to incorporate chatbots, but in what ways could the other features of robots facilitate language acquisition? Pepper (the robot in the picture) has ‘touch sensors, LEDs and microphones for multimodal interactions’, along with ‘infrared sensors, bumpers, an inertial unit, 2D and 3D cameras, and sonars for omnidirectional and autonomous navigation’. How could these features help language acquisition?

Lee and Lee (2022) attempt to provide an answer to this question. Here’s what they have come up with:

By virtue of their physical embodiment, social robots have been suggested to provide language learners with direct and physical interactions, which is considered one of the basic ingredients for language learning. In addition, as social robots are generally humanoids or anthropomorphized animal shapes, they have been valued for their ability to serve as familiar conversational partners, having potential to lower the affective filter of language learners.

Is there any research evidence to back up these claims? The short answer is no. Motivation and engagement may sometimes be positively impacted, but we can’t say any more than that. As far as learning is concerned, Lee and Lee (2022: 121) write: involving social robots led to statistically similar or even higher [English language learning] outcomes compared with traditional ELT contexts (i.e. no social robot). In other words, social robots did not, on the whole, have a negative impact on learning outcomes. Hardly grounds for wild enthusiasm … Still, Lee and Lee, in the next line, refer to the ‘positive effectiveness of social robots in English teaching’ before proceeding to enumerate the ways in which these robots could be used in English language learning. Doesn’t ELT Journal have editors to pick up on this kind of thing?

So, how could these robots be used? Lee and Lee suggest (for younger learners) one-on-one vocabulary tutoring, dialogue practice, more vocabulary teaching, and personalized feedback. That’s it. It’s worth noting that all of these functions could equally well be carried out by chatbots as by social robots.

Lee and Lee discuss and describe the social robot, NAO6, also built by SoftBank Robotics. It’s a smaller and cheaper cousin of the Pepper robot that illustrates the EL Gazette article. Among Lee and Lee’s reasons for using social robots is that they ‘have become more accessible due to ever-lower costs’: NAO6 costs around £350 a month to rent. Buying it outright is also an option. Eduporium (‘Empowering the future with technology’) has one on offer for $12,990.00. According to the blurb, it helps ‘teach coding, brings literature to life, enhances special education, and allows for training simulations. Plus, its educational solutions include an intuitive interface, remote learning, and various applications for accessibility!’

It’s easy enough to understand why EL Gazette uses clickbait from time to time. I’m less clear about why ELT Journal would print this kind of nonsense. According to Lee and Lee, further research into social robots ‘would initiate a new era of language learning’ in which the robots will become ‘an important addition to the ELT arsenal’. Yeah, right …

References

Lee, H. & Lee, J. H. (2022) Social robots for English language teaching. ELT Journal 76 (1): 119 – 124

Lee, J. H., Yang, H., Shin D. & Kim, H. (2020) Chatbots. ELT Journal 74 (3): 338 – 3444

Comments
  1. Grzegorz says:

    Thank you for yet another great read, Philip! If I may add my twopence:

    [1] the whole craze with teaching critical thinking (along with virtually all other so-called 21st century skills) is predicated upon the assumption (totally unwarranted imho) that teaching professionals (incl. EFL teachers) themselves already have the skill(s) (or disposition(s)) in question – however defined. Your argument makes it painfully clear that this is far from being the case, incl. the very conceptual grasp of the issue (or rather lack thereof), let alone the question of whether a particular skill/ disposition is in principle teachable/ learnable, and if it were to be, how should one go about facilitating its development in one’s learners.

    [2] This brings me to a more general reflection: isn’t it that the menu of things that the EFL teaching community is being pushed (and not too gently) to concern ourselves with is growing at a rather alarming rate …? CT (and other 21st century skills), global citizenship, inclusion, promoting growth mindset, students’ well-being – to name but a few. One starts wondering about where it all leaves the good, old-school teaching of language, its beauty and complexity …?

    Kindest,

  2. […] In May of last year, EL Gazette had a story entitled ‘Your new English language teacher is a robot’ that was accompanied by a stock photo of a humanoid robot, Pepper (built by SoftBank Robotics). The story was pure clickbait and the picture had nothing to do with it.  […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s