Experiments in Artificial Theory of Mind

Since setting out my initial thoughts on robots with simulation-based internal models about 5 years ago - initially in the context of ethical robots - I've had a larger ambition for these models: that they might provide us with a way of building robots with artificial theory of mind - something I first suggested when I outlined the consequence engine 4 years ago.

Since then we've been busy experimentally applying our consequence engine in the lab, in a range of contexts including ethics, safety and imitation, giving me little time to think about theory of mind. But then, in January 2017 I was contacted by Antonio Chella, inviting me to submit a paper to a special issue on Consciousness in Humanoid Robots. After some hesitation on my part and encouragement on Antonio's I realised that this was a perfect opportunity.

Of course theory of mind is not consciousness but it is for sure deeply implicated. And, as I discovered while researching the paper, the role of theory of mind in consciousness (or, indeed of consciousness in theory of mind) is both unclear and controversial. So, this paper, written in the autumn of 2017, submitted January 2018, and - after tough review and major revisions - accepted in June 2018, is my first (somewhat tentative) contribution to the machine consciousness literature.

Experiments in Artificial Theory of Mind: From Safety to Story-Telling, advances the hypothesis that simulation-based internal models offer a powerful and realisable, theory-driven basis for artificial theory of mind.

Here is the abstract

Theory of mind is the term given by philosophers and psychologists for the ability to form a predictive model of self and others. In this paper we focus on synthetic models of theory of mind. We contend firstly that such models—especially when tested experimentally—can provide useful insights into cognition, and secondly that artificial theory of mind can provide intelligent robots with powerful new capabilities, in particular social intelligence for human-robot interaction. This paper advances the hypothesis that simulation-based internal models offer a powerful and realisable, theory-driven basis for artificial theory of mind. Proposed as a computational model of the simulation theory of mind, our simulation-based internal model equips a robot with an internal model of itself and its environment, including other dynamic actors, which can test (i.e., simulate) the robot’s next possible actions and hence anticipate the likely consequences of those actions both for itself and others. Although it falls far short of a full artificial theory of mind, our model does allow us to test several interesting scenarios: in some of these a robot equipped with the internal model interacts with other robots without an internal model, but acting as proxy humans; in others two robots each with a simulation-based internal model interact with each other. We outline a series of experiments which each demonstrate some aspect of artificial theory of mind.

For an outline of the work of the paper see the slides below, presented first at the SPANNER workshop in York in September 2018, then at a workshop on Social Learning and Cultural Evolution at ALife 2019 in July 2019.

In fact all of the experiments outlined here have been described in some detail in previous blog posts (although not in the context of artificial theory of mind):

1. The Corridor experiment 
2. The Pedestrian experiment
3. The Ethical robot experiments: with e-puck robots and with NAO robots
4. Experiments on rational imitation (the imitation of goals)
5. Story-telling robots**

The thing that ties all of these experiments together is that they all make use of a simulation-based internal model (which we call a consequence engine), which allows our robot to model and hence predict the likely consequences of each of its next possible actions, both for itself and for the other dynamic actors it is interacting with. In some of the experiments those actors are robots acting as proxy humans, so those experiments (in particular the corridor and ethical robot experiments) are really concerned with human-robot interaction.

Theory of mind is the ability to form a predictive model of ourselves and others; it's the thing that allows us to infer the beliefs and intentions of others. Curiously there are two main theories of mind: the 'theory theory' and the 'simulation theory'. The theory theory (TT) holds that one intelligent agent’s understanding of another’s mind is based on innate or learned rules, sometimes known as folk psychology. In TT these hidden rules constitute a 'theory' because they can be used to both explain and make predictions about others’ intentions.  The simulation theory (ST) instead holds that “we use our own mental apparatus to form predictions and explanations of someone by putting ourselves in the shoes of another person and simulating them” (Michlmayr, 2002).

When we hold our simulation-based internal model up against the simulation theory of mind, the two appear to mirror each other remarkably well. If a robot has a simulation of itself inside itself then it can explain and predict the actions of both itself, and others like itself by using its simulation-based internal model to model them. Thus we have an embodied computational model of theory of mind, in short artificial theory of mind.

So, what properties of theory of mind (ToM) are demonstrated in our five experiments?

Well, the first thing to note is that not all experiments implement full ST. In the corridor, pedestrian and ethical robot experiments robots predict their own actions using the simulation-based internal model, i.e. ST, but use a much simpler TT to model the other robots; we use a simple ballistic model for those other robots (i.e. by assuming the robot will continue to move at the speed and direction it is currently moving). Thus I describe these experiments as ST (self) + TT (other), or just ST+TT for short. I argue that this hybrid form of artificial ToM is perfectly valid, since you and I clearly don't model strangers we are trying to avoid in a crowded corridor as anything other than people moving in a particular direction and speed. We don't need to try and intuit their state of mind, only where they are going.

The rational imitation and story-telling experiments do however, use ST for both self and other, since a simple TT will not allow an imitating robot to infer the goals of the demonstrating robot, nor is it sufficient to allow a listener robot to 'imagine' the story told by the storytelling robot.

The table below summarises these differences and highlights the different aspects of theory of mind demonstrated in each of the five experiments.

*Theory Mode: ST (self) + TT/ST (other)

An unexpected real-world use for the approach set out in this paper, is to allow robots to explain themselves. I believe explainability will be especially important for social robots, i.e. robots designed to interact with people. Let me explain by quoting two paragraphs from the paper.

A major problem with human-robot interaction is the serious asymmetry of theory of mind. Consider an elderly person and her care robot. It is likely that a reasonably sophisticated near-future care robot will have a built-in (TT) model of an elderly human (or even of a particular human). This places the robot at an advantage because the elderly person has no theory of mind at all for the robot, whereas the robot has a (likely limited) theory of mind for her. Actually the situation may be worse than this, since our elderly person may have a completely incorrect theory of mind for the robot, perhaps based on preconceptions or misunderstandings of how the robot should behave and why. Thus, when the robot actually behaves in a way that doesn’t make sense to the elderly person, her trust in the robot will be damaged and its effectiveness diminished.

The storytelling model proposed here provides us with a powerful mechanism for the robot to be able to generate explanations for its actual or possible actions. Especially important is that the robot’s user should be able to ask (or press a button to ask) the robot to explain “why did you just do that?” Or, pre-emptively, to ask the robot questions such as “what would you do if I fell down?” Assuming that the care robot is equipped with an autobiographical memory, the first of these questions would require it to re-run and narrate the most recent action sequence to be able to explain why it acted as it did, i.e., “I turned left because I didn’t want to bump into you.” The second kind of pre-emptive query requires the robot to interpret the question in such a way it can first initialize its internal model to match the situation described, run that model, then narrate the actions it predicts it would take in that situation. In this case the robot acts first as a listener, then as the narrator (see slide 18 above). In this way the robot would actively assist its human user to build a theory-of-mind for the robot.

---

**This one remains, for the time-being, a thought experiment.

See also: When Robots Tell Each Other Stories: The Emergence of Artificial Fiction

Reference:

Michlmayr, M. (2002). Simulation Theory Versus Theory Theory: Theories Concerning the Ability to Read Minds. Master’s thesis, Leopold-Franzens- Universität Innsbruck.

The feeling of what it is like to be a Robot

Philosopher Thomas Nagel famously characterised subjective experience as “something that it is like to be…” and suggested that for a bat, for instance, there must be something that it is like to be a bat [1]. Nagel also argued that, since we humans differ so much from bats in the way we perceive and interact with the world, then it is impossible for us to know what it is like for a bat to be a bat. I am fascinated, intrigued and perplexed by Nagel’s ideas in equal measure. And, since I think about robots, I have assumed that if a robot were ever to have conscious subjective experience then there must be something that it is like to be a robot that – even though we had designed that robot – we could not know.

But I now believe it may eventually be just possible for a human to experience something approaching what it is like to be a robot. To do this would require two advances: one in immersive robot tele-operation, the other in the neuroscience of body self-image manipulation.

Consider first, tele-operation. Tele-operated robots are, basically, remotely controlled robots. They are the unloved poor relations of intelligent autonomous robots. Neither intelligent nor autonomous, they are nevertheless successful and important first wave robots; think of remotely operated vehicles (ROVs) engaged in undersea exploration or oil-well repair and maintenance. Think also of off-world exploration: the Mars rovers are hugely successful; the rock-stars of tele-operated robots.

Roboticists are good at appropriating technologies or devices developed for other applications and putting them to good use in robots: examples are WiFi, mobile phone cameras and the Microsoft Kinnect. With the high profile launch of the Oculus Rift headset, and their acquisition by Facebook, and with competing devices from Sony and others, there are encouraging signs that immersive Virtual Reality (VR) is on the verge of becoming a practical, workable proposition. Of course VR’s big market is video games – but VR can and, I believe, will revolutionise tele-operated robotics.

Imagine a tele-operated robot with a camera linked to the remote operator’s VR headset, so that every time she moves her head to look in a new direction the robot’s camera moves in sync; so she sees and hears what the robot sees and hears in immersive high definition stereo. Of course the reality experienced by the robot’s operator is real, not virtual, but the head mounted VR technology is the key to making it work. Add haptic gloves for control and the robot’s operator has an intuitive and immersive interface with the robot.

Now consider body self-image modification. Using mirror visual feedback researchers have discovered that it is surprisingly easy to (temporarily) modify anyone’s body self-image. In the famous rubber hand illusion a small screen is positioned to hide a subject’s real hand. A rubber hand is positioned where her hand could be, in full view, then a researcher simultaneously strokes both the real and rubber hands with a soft brush. Within a minute or so she begins to feel the rubber hand is hers, and flinches when the researcher suddenly tries to hit it with a hammer.

Remarkably H.H. Ehrsson and his colleagues extended the technique to the whole body, in a study called ‘If I Were You: Perceptual Illusion of Body Swapping’ [2]. Here the human subject wears a headset and looks down at his own body. However, what he actually sees is a mannequin, viewed from a camera mounted on the mannequin’s head. Simultaneous tactile and visual feedback triggers the illusion that the mannequin’s body is his own. It seems to me that if this technique works for mannequins then it should also work for robots. Of course it would need to be developed to the point that elaborate illusions involving mirrors, cameras and other researchers providing tactile feedback are not needed.

Now imagine such a body self-image modification technology combined with fully immersive robot tele-operation based on advanced Virtual Reality technology. I think this might lead to the robot's human operator experiencing the illusion of being one with the robot, complete with a body self-image that matches the robot's possibly non-humanoid body. This experience may be so convincing that the robot's operator experiences, at least partially, something like what it is to be a robot. Philosophers of mind would disagree - and rightly so; after all, this robot has no independent subjective experience of the world, so there is no something that it is like to be. The human operator could not experience what it is like to think like a robot, but she could experience what it is like to sense and act in the world like a robot.

The experience may be so compelling that humans become addicted to the feeling of being a robot fish, or robot dragon or some other fantasy creature, that they prefer this to the quotidian experience of their own bodies.

---

[1] Nagel, Thomas. What is it like to be a bat?, Mortal Questions, Cambridge University Press, 1979.

[2] Petkova VI, Ehrsson HH (2008) If I Were You: Perceptual Illusion of Body Swapping. PLoS ONE 3(12): e3832. doi:10.1371/journal.pone.0003832

Why robots will not be smarter than humans by 2029

In the last few days we've seen a spate of headlines like 2029: the year when robots will have the power to outsmart their makers, all occasioned by an Observer interview with Google's newest director of engineering Ray Kurzweil.

Much as I respect Kurzweil's achievements as an inventor, I think he is profoundly wrong. Of course I can understand why he would like it to be so - he would like to live long enough to see this particular prediction come to pass. But optimism doesn't make for sound predictions. Here are several reasons that robots will not be smarter than humans by 2029.

• What exactly does as-smart-as-humans mean? Intelligence is very hard to pin down. One thing we do know about intelligence is that it is not one thing that humans or animals have more or less of. Humans have several different kinds of intelligence - all of which combine to make us human. Analytical or logical intelligence of course - the sort that makes you good at IQ tests. But emotional intelligence is just as important, especially (and oddly) for decision making. So is social intelligence - the ability to intuit others' beliefs, and to empathise. 

• Human intelligence is embodied. As Rolf Pfeifer and Josh Bongard explain in their outstanding book you can't have one without the other. The old Cartesian dualism - the dogma that robot bodies (the hardware) and mind (the software) are distinct and separable - is wrong and deeply unhelpful. We now understand that the hardware and software have to be co-designed. But we really don't understand how to do this - none of our engineering paradigms fit. A whole new approach needs to be invented.

• As-smart-as-humans probably doesn't mean as-smart-as newborn babies, or even two year old infants. They probably mean somehow-comparable-in-intelligence-to adult humans. But an awful lot happens between birth and adulthood. And the Kurzweilians probably also mean as-smart-as-well-educated-humans. But of course this requires both development - a lot of which somehow happens automatically - and a great deal of nurture. Again we are only just beginning to understand the problem, and developmental robotics - if you'll forgive the pun - is still in its infancy.

• Moore's Law will not help. Building human-equivalent robot intelligence needs far more than just lots of computing power. It will certainly need computing power, but that's not all. It's like saying that all you need to build a cathedral is loads of marble. You certainly do need large quantities of marble - the raw material - but without (at least) two other things: the design for a cathedral, and/or the knowhow of how to realise that design - there will be no cathedral. The same is true for human-equivalent robot intelligence. 

• The hard problem of learning and the even harder problem of consciousness. (I'll concede that a robot as smart as a human doesn't have to be conscious - a philosophers-zombie-bot would do just fine.) But the human ability to learn, then generalise that learning and apply it to completely different problems is fundamental, and remains an elusive goal for robotics and AI. In general this is called Artificial General Intelligence, which remains as controversial as it is unsolved.

These are the reasons I can be confident in asserting that robots will not be smarter than humans within 15 years. It's not just that building robots as smart as humans is a very hard problem. We have only recently started to understand how hard it is well enough to know that whole new theories (of intelligence, emergence, embodied cognition and development, for instance) will be needed, as well as new engineering paradigms. Even if we had solved these problems and a present day Noonian Soong had already built a robot with the potential for human equivalent intelligence - it still might not have enough time to develop adult-equivalent intelligence by 2029.

That thought leads me to another reason that it's unlikely to happen so soon. There is - to the best of my knowledge - no very-large-scale multidisciplinary research project addressing, in a coordinated way, all of the difficult problems I have outlined here. The irony is that there might have been. The project was called Robot Companions, it made it to the EU FET 10-year Flagship project shortlist but was not funded.

Search Results

Could we experience the workings of our own brains?

One of the oft quoted paradoxes of consciousness is that we are unable to observe or experience our own conscious minds at work; that we cannot be conscious of the workings of consciousness. I've always been puzzled about why this is a puzzle. After all, we don't think it odd that word processors have no insight into their inner workings (although that's a bad example because we might conceivably code a future self-aware WP and arrange for it to access its inner machinery).

Perhaps a better example is this. The act of picking up a cup of hot coffee and bringing it to your lips appears, on the face of it, to be perfectly observable. No mystery at all. We can see the joints and muscles at work, 'feel' the tactile sensing of the coffee cup, and its weight as we begin to lift it. We can even build mathematical models of the kinetics and dynamics, and (with somewhat more difficulty) make robot arms to pick up cups of coffee. But - I contend - we are kidding ourselves if we think we know what's going on in the complex sensory and neurological processes that appear so effortless to perform. The fact we can observe and even feel ourselves lifting a coffee cup gives very little real insight. And the mathematical models - and robots - are not really models of the human neurological and physiological processes at all, they are models of idealised abstractions of limbs, joints and hand.

I would argue that we have no greater insight into the workings of this (apparently straightforward) physical act, than we do of thinking itself. But again this is not surprising. The additional cognitive machinery to be able to access or experience the inner workings of any process, whether mental or physical, would be huge and (biologically) expensive. And with no apparent survival value (except perhaps for philosophers of mind), it's not surprising that such mechanisms have not evolved. They would of course require not just extra grey matter, but sensing too. It's interesting that there are no pain receptors within our brains - that's why it's perfectly possible to have brain surgery while wide awake.

But this got me thinking. Imagine that at some future time we have nanoscale sensors capable of positioning themselves throughout our brains in order to provide a very large sensor network. If each sensor is monitoring the activity of key neurons, or axons, and able to transmit its readings in real-time to an external device, then we would have the data to provide ourselves with a real-time activity image of our own brains. It could be presented visually, or perhaps sonically (or via multi-media). It might be fun for awhile, but this personal brain imaging technology (let's call it iBrain) probably wouldn't provide us with much more insight or experience of our own thought processes.

But let's assume that by the time we have the nanotechnology for harmlessly inserting millions of brain nanosensors we will have also figured out the major architectural structures of the brain - crucially linking the neural scale to the macro scale. Actually, if we believe that the recently announced European and US human brain Grand Challenges will achieve what they are promising in terms of modelling and mapping human brain activity, then such an understanding should only be a few decades away. So now build those maps and structures into the personal iBrain, and we will be presented not with a vast and bewildering cloud of colours, as in the beautiful image above, but a simpler image with major highways and structures highlighted. Still complex of course, but then so are street maps of cities or countries. So the iBrain would allow you to zoom into certain regions and really see what's going on while you (say) listen to Bach (the very thing I'm doing right now).

Then we really would be able to observe our own brains at work and, just perhaps, experience the connection between brain and thought.

Human consciousness could be immortal

Our subjective experience of the 'continuity' of consciousness is surely an illusion. But what makes that illusion and why is it so compelling? That's a deep question but here are I think two fundamental reasons:

1. Embodiment. You are an embodied intelligence. It is a mistake to think of mind and body as somehow separate. Our conscious experience and its awakening as a developing child is surely deeply rooted in our physical experience of the world as mediated through our senses.

2. Environmental continuity. Our experience of the world changes 'relatively' slowly. The word relative is important here since I mean relative to the rate at which our conscious experience updates itself. Of course we do experience the discontinuity of going to sleep then waking to the changed world of a new sunrise, but this is both deeply familiar and predictable.

A word that encompasses both of these is situatedness. Our intelligence, and hence also conscious experience is inextricably situated in our bodies and in the world. Let me illustrate what this might mean with a thought experiment.

Imagine a brain transplant. Your brain complete with its memories and life's experience, together with as much of your central nervous system as might be needed for it to function properly were to be transplanted into a different body. You would wake from the procedure into this new body. I strongly suspect that you would experience a profound and traumatic discontinuity of consciousness and, well, go mad. Indeed it's entirely possible that you simply couldn't (and perhaps mercifully) regain or experience any sort of consciousness at all. Why? Because the conditions for the emergence of consciousness and the illusion of its continuity have been irreversibly broken.

However, if what I have said above is true, there's a flip side to the story that could have extraordinary consequences.

If the continuity of consciousness is an illusion then, in principle at least, it might be possible to artificially perpetuate that illusion.

Imagine that at some future time we have a sufficiently deep understanding of the human brain that we can scan its internal structures for memories, acquired skills, and all of those (at present dimly understood) attributes that make you you. It's surely safe to assume that if we're able to decode a brain in this way, then we would also be able to scan the body structures (dynamics, musculature and deep wiring of the nervous system). It would then be a simple matter to scan, at or just before the point of death, and transfer those structures into a virtual avatar within a virtual world. The simulated brain structures would be 'wired' to the avatar's virtual body in exactly the same way as the real brain was wired to its real body, thus satisfying the requirement for embodiment. If the virtual world is also a high fidelity replica of the real world then we would also satisfy the second requirement, environmental continuity.

I would argue that, under these circumstances, the illusion of the continuity of consciousness could be maintained. Thus, on dying you would awake (in e-heaven), almost as if nothing had happened. Except, of course, that you could be greeted by the avatars of your dead relatives. Even better, because e-heaven is just a virtual environment in the real-world, then you could just as easily be visited by your living friends and relatives. Could this be the retirement home of the far future?

In this way human consciousness could, I believe, be immortal.

On what brains are for

Just started reading Sue Blackmore's new book: Conversations on Consciousness. Great introduction - but I have stalled already on Q2 of the 1st conversation - Baars. He says "the primary function of the nervous system is to encode knowledge...".

Surely not.

Brains are control systems for bodies. Pretty amazing control systems of course, but control systems all the same. Bodies have sensors (senses) and actuators (muscles). In very simple animals the outputs from the senses are almost directly connected to the muscles, so that the animal always reacts reflexively to stimuli. More complex animals have more brain in between senses and muscles and so may deliberate before reacting. Of course even very complex animals still have reflexes - think of the classic reflex test on your knee.

Of course I accept entirely that complex brains do encode knowledge, probably in a multiplicity of ways some of which we can discern - like the apparent spatial mapping of images into the visual cortex - but many ways that are not (yet) at all understandable.

But that is not what they're for.