The lay notion of how human communication works is, basically, that we exchange packages of information that are encoded and then decoded in our heads using language. But it just takes a little observation to see that this “computer metaphor” just doesn’t apply to the human way of communicating. We can’t reduce communication to a transfer of abstract information, and the same happens to be true of human-machine communication if we expect it to be of any use.

Think of turn-taking during a conversation. Speakers and listeners switch roles very quickly, often faster than typical reaction times. Silences between turns last less than 300 milliseconds, leading us to believe that a lot more is going on than simply waiting for the information to arrive in order to decide the next step. In fact, a lot of anticipatory planning is happening in both the speaker and listener. There’s a continuous monitoring of the other in search of subtle cues as to when the turn will be yielded. Such cues include changes of prosody, gestures, gaze shifts toward the listener, implicit verbal cues such as filler utterances…. Turn-taking is just one example of embodiment, a critical dimension of communication that most classical models ignore.

In fact it was the the study of human language that led the way toward the revolution of embodiment in the cognitive sciences. Cognition, in this view, unfolds in time while we perceive and act. Cognitive processes are continuously linked to sensory information and continuously coupled to motor systems that impact on the sensed world.

Communication is embodied in the obvious sense that natural communication occurs between people with physical and physiological bodies, and their cognitive abilities originate from processes in their nervous systems. But the embodiment stance to human communication is deeper than that. It extends the scientific approach to human language by adding a new emphasis on the grounding of language in the ongoing perception of the scene.

We begin to understand the emergence of language during development as a consequence of the immersion of the child in an environment shared with speakers. Development helps us explain how gesture and other non-linguistic cues are tightly coupled into language processing in the here and now of human communication. Embodied communication is also essential for researchers building artificial cognitive systems that autonomously generate language to communicate with a human and, conversely, make sense of human language and gesture by tying it to the system’s own scene perception.

Researchers exploring these different approaches to the study of embodied communication will share their ideas and insights during the Fifth EUCogIII Members Conference in Bochum 19-20 March 2014. Peter Dominey from the INSERM Stem Cell and Brain Research Institute in Lyon, France will show, for instance, how the iCub robot acquires “meaning” by establishing a link between physical action and language, and Yiannis Aloimonos from the University of Maryland in the USA will demonstrate how an “action grammar” is able to parse human action into meaningful sequences. Psychologists like Larissa Samuelson and Susan Wagner Cook from the University of Iowa will present the developmental perspective on embodied communication, focusing on how children learn words and how gesture and language co-develop. Srini Narayana will show in theoretical models how meaning and grammar are intimately tied together in the learning process.

Our aim is to advance the interdisciplinary interchanges between artificial and human cognition research fields. Both technical and natural aspects of communication will therefore be represented at the conference. To develop cognitive systems we need to understand the variety of ways that speech and gesture are coordinated, and develop technical aspects like creating automatic semantic descriptions of complex human activity from perceptual data. Knowing more about how humans really communicate and make sense of human communication is important if we want to create machines that are smart like us.

EUCog is the European Network for the Advancement of Artificial Cognitive Systems, Interaction and Robotics.

Here are the slides of my keynote at last week’s excellent EUCog meeting: Social and Ethical Aspects of Cognitive Systems. And the talk itself is here, on YouTube.

The talk was in three parts.

Part 1: here I outline why and how roboticists must be ethical. This is essentially a recap of previous talks. I start with the societal context: the frustrating reality that even when we meet to discuss robot ethics this can be misinterpreted as scientists fear a revolt of killer robots. This kind of media reaction is just one part of three linked expectation gaps, in what I characterise as a crisis of expectations. I then outline a few ethical problems in robotics – just as examples. Here I argue it’s important to link safe and ethical behaviour – something that I return to later. Then I recap the five draft principles of robotics.

Part 2: here I ask the question: what if we could make ethical robots? I outline new thinking which brings together the idea of robots with internal models, with Dennett’s Tower of Generate and Test, as a way of making robots that can predict the consequences of their own actions. I then outline a generic control architecture for robot safety, even in unpredictable environments. The important thing about this approach is that the robot can generate next possible actions, test them in its internal model, and evaluate the safety consequences of each possible action. The unsafe actions are then inhibited – and the robot controller determines which of the remaining safe actions is chosen, using its usual action-selection mechanism. Then I argue that it is surprisingly easy to extend this architecture for ethical behaviour, to allow the robot to predict the robot actions that would minimise harm for a human in its environment. This appears to represent an implementation of Asimov’s 1st and 3rd laws. I outline the significant technical challenges that would need to be overcome to make this work.

But, assuming such a robot could be built, how ethical would it be? I suggest that with a subset of Asimovian ethics it probably wouldn’t satisfy an ethicist or moral philosopher. But, nevertheless – I argue there’s a good chance that such a minimally ethical robot could help to increase trust, in the robot, from its users.

Part 3: in the final part of the talk I conclude with some ethical questions. The first is: if we could build an ethical robot, are we ethically compelled to do so? Some argue that we have an ethical duty to try and build moral machines. I agree. But the counter argument, my second ethical question, is are there ethical hazards? Are we opening a kind of ethical Pandora’s box, by building robots that might have an implicit claim to rights, or responsibilities. I don’t mean that such a robot would ask for rights, but instead that, because it is has some moral agency, then we might think it should be accorded rights. I conclude that we should try and build ethical robots. The benefits I think far outweigh any ethical hazards, which in any event can, I think, be minimised.

*It was not so much an epiphany, as a slow conversion from sceptic to believer. I have long term collaborator Michael Fisher to thank for doggedly arguing with me that it was worth thinking deeply about how to build ethical robots.