‘Private money will really push the field forward’

Neuralink Monkey MindPong, deconstructed. Jon Sutton reports.

You may have seen the video Neuralink, Elon Musk’s computer to brain interface company, released recently. It claims to show a macaque monkey playing the video game pong with its mind

When my 16-year-old son excitedly told me to watch it, I scoffed ‘Psychologists have been doing that for years. Adrian Owen told me all about it in 2017.' 

To quote an extract from Adrian’s book, Into the Grey Zone:

'The scientific story begins with experiments in the late 1990s, scanning apparently vegetative patients to find that their brains still 'recognised' the faces of loved ones, or the meaning of words. Could it be that, at least in some cases, patients were still fully conscious, aware of their surroundings, understanding conversations, and simply unable to move or communicate in any way? To answer the question, the experiments moved on to using brain imaging for actual communication. The patient might be told: 'We will ask you some questions. If the answer is yes, imagine you are playing tennis. If the answer is no, imagine you are walking around your house.' Brain imaging can tell the difference between what happens when a person imagines playing tennis versus imagining walking round the home… might apparently vegetative patients be able to use this method to answer questions put to them?

Incredibly, sometimes, it worked… using brain imaging, the group had found a way to unlock the trap of a conscious mind in an uncontrollable body. Over two decades, many nuances have emerged. Consciousness and ability to communicate are certainly not present in all patients – probably, indeed, in a minority. For this minority, however, the importance of the findings is staggering. The methods have been developed to show that a patient can follow the plot of a movie, showing a normal, complex response to moments of suspense. They can be used to ask the deepest questions, sometimes showing that a trapped patient can still say that they are happy.  They can also sometimes predict who it is that will finally come out of their coma and return to their families and friends.' 

But is Neuralink the same? I reached out to Adrian (Professor of Cognitive Neuroscience and Imaging, Department of Physiology and Pharmacology and Department of Psychology, Western University), and to his colleague Dr Andrew Pruszynski, Canada Research Chair in Sensorimotor Neuroscience Associate Professor, Physiology and Pharmacology, Psychology, and Principal Investigator, Brain and Mind Institute).

Adrian Owen: ‘There are a few interesting things about this story. The simple take (on my side) is that we have never been able to achieve this sort of thing in human patients (which is what we would really like to do) because surface electrodes (EEG) do so much more poorly than implanted electrodes and ethically it’s tough to contemplate putting implanted electrodes into unresponsive humans (although we are working through some of the ethical issues and may get there eventually). The closest we’ve got is in this video. It’s a healthy person (a member of my team) basically playing pong with an EEG. But there are loads of caveats… it’s a (reasonably) high density EEG system (so expensive) and crucially this took many hours of practise, both by the participant and by the computer that learned to decode his thoughts. And even then he was only imagining squeezing his hands (to go up) and imagining squeezing his feet (to go down). 

That’s a good indication of just how hard it is to achieve this stuff with EEG, regardless of how much money you throw at the problem. And in patients who can’t communicate, training is almost impossible – hence, lack of progress. What complicates matters is that there are many commercial players who make consumer devices that claim to do the same for $100 – they don’t. I regularly challenge these companies to a game of Pong and none have been able to do it under laboratory conditions despite displaying wonderful videos of such on YouTube etc. A lot of people are prepared to lie about how easy ‘mind control’ is to make a buck. 

Then there’s a whole different angle to this story… what Musk et al achieved has been done before (10 years ago) and a lot of people think his team are just recycling old science for commercial glory. But I think it’s more complicated than that… what they have achieved is remarkable in that it signals what is likely to come next and future possibilities (patients!). So there’s a lot of science/political mud-slinging about that this week. 

Andrew Pruszynksi: ‘If you watch this video Paul Nuyujukian, Director of the Brain Interfacing Laboratory (BIL) at Stanford University, breaks down of the capabilities of the Neuralink BMI. In some ways the results are similar to what labs have shown in the past, but at the same time there is no doubt they are on par with state of the art in animal lab setting and further advanced in several important areas, including the electrode technology and the surgical approach and the data streaming. I think it's becoming pretty clear that this in an exciting time where lab ideas are moving into the corporate realm (not just Neuralink, there are many companies in this space now) and the focus / resources that private money brings to bear will really push the field forward.’

BPS Members can discuss this article

Already a member? Or Create an account

Not a member? Find out about becoming a member or subscriber