Symposium news

We will be hosting a symposium discussion for this project.

Signs of Life: Puppetry, Emotions, Embodiment and Empathy

Friday 20th October, 2017

14.00-17.30, SCR, Arts Two Building, QMUL (Mile End)

A panel discussion with Mervyn Millar, Emily Cross, Matthew Longo, Susanne Quadflieg and Joel Smith

Chaired by Tiffany Watt Smith, QMUL.

What role can puppetry play in unravelling the psychology of empathy and our ability to read each other’s emotions? How is it that we can we so easily feel apparently real emotions when watching bodies that are not even flesh and blood? Do we react differently to puppets, robots, avatars and people? And can we ever really ‘suspend disbelief’?

Puppet designer and director Mervyn Millar has been Artist in Residence at QMUL’s Centre for the History of Emotions during 2017.  He will discuss these themes and related phenomena with academic researchers from a variety of fields: Prof Emily Cross (School of Psychology, Bangor University), Prof Matthew Longo (Cognitive Neuroscience ‘Bodylab’, Birkbeck University of London), Dr Susanne Quadflieg (Senior Lecturer in Experimental Psychology, University of Bristol), Dr Joel Smith (Senior Lecturer in Philosophy, University of Manchester) and some puppeteers will all contribute to round table discussions and debate. Chaired by Dr Tiffany Watt-Smith (Centre for the History of Emotions, QMUL).

Attendance is free but limited: please email Helen Stark ( by 16 October for a place. This event will be recorded.

The page on the Centre for the History of Emotions site is here

We hope to be able to put the recording online at some point.


2nd August 2017

Where I am now: Story and science


I’m working away – a little fruitlessly sometimes – trying to pick my way through how to turn this research into engaging narrative.


I’m focusing on two strands – the perceptual experience and the embodied experience. The idea isn’t to make a show about puppetry – but to allow this research to bring to the surface themes that will be well expressed by puppetry – because of the way it works on us.

My thinking is that when we are watching puppets we are likely to be engaged in embodied empathy with the puppet character and our physical engagement with emotion and action will be quite present. I think we are also involved in perceptual games. The dominant one if course is that something is there in front of us that is acting like a person but is not one – and by extension, in the theatre, a number of things are ‘representing’ with quite loose semiotics. So, in my puppet stage, for example, it’s quite easy for a stick or an object to stand in for a person or a dog.


We see these things and we perceive (without much effort) patterns that allow us to fabulate and ‘see’ the behaviour of people and a story. We watch the puppet differently from an actor because the anomalous aspects of its design and movement keep our attention on it. So familiar movements and experiences that would be ‘tuned out’ as non-meaningful when an actor performs them can seem fresh and engaging.


So what kind of story revolves around these experiences? I made a deal with myself that I wouldn’t make a story about someone who goes mad, or someone who has an unusual condition. I want to play with where these phenomena are at the edge of our everyday experience.

Our unreliable perception – and our tendency to see life when there isn’t any, suggests a story about the reliability of witness. I enjoy Edelman’s contention “every act of perception is to some degree an act of creation, and every act of memory is to some degree an act of imagination”. It’s compelling at the moment to think about fabulation in its wider sense too – how prominent and influential men and media tell stories that make the world they want – or the world they most fear – and how seductive those stories are.

The perception strand expands, if we take in some of the other perceptual errors we fall into, into some profound questions about what we really do see and how our attention and consciousness are formed. This seems hard to talk about but may still be something that the puppets and actors can explore.

The other key strand in the story is about connection – and particularly unspoken connection. A lot of the articles and pieces I’ve been reading have been on how we read emotions, thoughts and intentions in other people – theory of mind, or ‘mind-reading’. Some of this is feeds in to the perceptual story – particularly ideas about how much we predict versus how much we are really seeing. But the side that I find most compelling is all of the sub- (or perhaps pre-) conscious reading of each others’ behaviour that seems to be happening via ‘motor resonance’ and what are being called mirror systems. We know from theatre and literature how powerful verbal imagery can be in conjuring physical sensations – and we know that the physical components of emotions can be contagious. So it seems very plausible that the viewer can read and empathise with the movement qualities of performing objects. How do we communicate when speech is denied to us – how closely can we identify with each other through our gestures and postures? Some of the body image work I have been finding out about suggests that in fact we are having to tell ourselves that each others’ bodies are not our own – and that the body image can extend to include other bodies and elements.


I’d love it if you can think of any good examples of the sort of phenomena here that spring to mind that might be vivid examples to help me compose events and images.

11th July 2017


Theory and Practice

It has been a while since I last blogged. I had a very productive time at the O’Neill Puppetry Conference in Connecticut in June, taking this project out of the academy and dropping it into the laps of performers and puppeteers. I had a brilliantly open, resourceful, ingenious and creative group of puppeteers from a range of backgrounds to work with and they fielded this material with aplomb. We looked at how some of these themes relate to the way we learn about puppetry and theatre, and played with how some of these phenomena might be realised on stage. Before I left I had a great meeting with Dr Tiffany Watt-Smith, my main contact at the QMUL Centre for the History of Emotions. She talked about how it was time to get away from reading research papers and into the mindset of an artist again. The O’Neill experience has certainly helped do that.


Three Strands

In preparing for the O’Neill I ended up thinking of the work in terms of three strands that might be productive for exploration. One, which I called Embodiment, is about the way the puppet or object might be able to communicate emotion (or feeling) through movement. It’s a non-scientific way of looking at the ideas of motor resonance as a perceptual system, and at empathy or even mindreading triggered by movement. The second strand was Perception – we used a traditional puppetry format, the ‘tabletop puppet’ (in this case about 70cm high) to dramatise some ideas relating to our unreliable perception – seeing things that seem to be alive, not seeing things that are there, and so on.

These are both really rich strands of exploration that a re familiar to puppeteers. My own style of teaching, which emphasises the breath, explicitly relates thought to the body. The substitution of the puppet body for a ‘real’ body already opens up the audience’s relationship to perceptual tricks and happy mistakes.


Invisible Puppetry

The third strand was about Imagination. I’ve been really interested in how memory, imagination and perception (and perhaps dream too) seem to be processes that overlap. And as the aesthetic philosophers reminded me, there is plenty of evidence that we have emotional and empathic responses to wholly imaginary characters. So this strand was about storytelling – conjuring character into being and inviting empathy with them. Puppetry without the puppets, or as it became, working with invisible puppets. We looked at using the puppeteers’ skill at focusing space very specifically and sensitively to allow an audience to imagine characters in a specific location in space – and even to begin to imagine them moving and behaving without the assistance of a puppet or object to act as a proxy.



Now I’m working on pulling some of these ideas into a story that might provide the plot for a show. The intention is to find a narrative framework that will allow puppetry to explore the themes as it is told. The focus of the story isn’t on the specific experience of watching a puppet – it’s on the phenomena and processes that contribute to that particular perceptual miracle. Hopefully this means that puppetry as a medium adds to the audience’s engagement with the subject matter. If you’re interested in helping me figure out what this story might be, let me know.


That doesn’t mean I’ve stopped reading research articles or stopped trying to understand how it is that we do this amazing thing.

19th May 2017


I enjoyed this video from TEDx Guildford – Russell Dean exploring many of the ideas that have prompted this project. I worked with Russell on his show Dissonance last year. It was a Wellcome Trust funded production in which Russell collaborated with a neuroscientist to make a theatre piece about cognitive dissonance.  Russell, a prodigious theatrical polymath, wrote and directed the show, made the puppet (he’s also a brilliant mask maker) and designed the set.

I’m not sure I agree with all of Russell’s conclusions – but this is a hugely engaging 17 minutes on how these phenomena might engage our perceptual systems.


Susan and the banana – three models

The Action Observation Network – ‘mirror neurons’ – is really interesting for theatre people. Here we have been, banging on for years about how you ‘identify with’ the character on stage and how theatre can give you a ‘visceral’ experience – and, finally, along comes science to say that, yes, when you watch Hamlet struggling to contain his fury and vengeance while questioning his motives and logic, if the actor is successfully embodying that very detailed conflict, everyone in the audience (or at least those who are watching), as they see and mirror all those tiny physical changes in rhythm, posture, inflection and movement, are ‘feeling it in their gut’ too. Or in the motor planning area for feeling it in their gut. There is, as you might expect, much interest and anticipation to see what comes out of this field: what kind of actions produce an effect on an audience member, and which don’t?

However… not all of the papers on the AON agree on how it works. After a recent chat with Dr Caroline Catmur at Kings College, I sent her an email asking to clarify some of the interpretations that I had made. To my surprise, the question wasn’t hopelessly naive. So here are three ideas for how the mirror neuron system might work – with apologies to the scientists who work in the filed for any misinterpretations I might have made.

Model A (a reactive model)

I see Susan and a banana. As Susan reaches for the banana, my matching motor planning areas respond by firing synchronously.

This should allow me to accurately identify what she is doing, and may possibly indirectly contribute to my understanding her intention (to take the banana, to push the banana over towards me, to crush the banana) via improved interoception of the movement if that helps with this. (Other brain systems also contribute to interpret her intention.)

Model B (a goal-directed model)

I see Susan and a banana. As Susan reaches towards the banana, I perceive or guess her intention (in non-motor areas) – ‘to take the banana’ – and whatever motor planning areas I would normally use to reach for the banana fire.

This may allow me to better track the flow of her intentions rather than matching only the actions.

Model C (a predictive model)

I see Susan and a banana. My motor planning areas fire in the areas which I would use to reach for the banana – because from the look of the banana (ripe, tasty) and what I know of Susan (likes bananas) this is the most likely outcome.

This may allow me to be ‘ahead’ of Susan. The interoceptive experience might influence my predictions. Deviations from my prediction (in what she actually does) tell me more about her intention and might increase my attention.

In a sense the only difference between B and C is that in C I am simulating before I see any action from Susan (this can surely be experimentally tested). Caroline observes also that if C is correct, in ambiguous situations the observer’s motor system should show reduced activation – as if there is no prediction, there is no engagement. C (and to an extent B) turn the system from an action observation network into an action modelling network. It returns us to some of the eggy theory of mind questions – do I model what I would do in that situation, or what I think Susan would do? And how would I do that without imagining a complete Susan in my head?

Both work with different implications for the theatre context. When I see Hamlet paused at a moment of suspension, Claudius vulnerable and at prayer – am I feeling what I see in the actor’s (or puppet’s) body? Or am I imagining how I would act in that situation and somehow ‘living it out’ in parallel with the action on stage, the performance of the actor constantly compared with my instincts?

5 May 2017

Perception – Creation – Imagination – Memory

I think I am finally working my way through some of the perceptual-mechanical questions towards some bigger ones (and as we know there are some notoriously hard problems on this safari..). I love this line in Edelman and Tononi’s “Consciousness” book:

“every act of perception is, to some degree, an act of creation, and every act of memory is, to some degree, an act of imagination.” 

I must admit that I find the prose in that book hard going – it’s the hardest read out of my shelf of pop-neuro paperbacks – and I don’t completely follow the theoretical work that they put into the idea of non-representational memory that underpins it. But I love their basic ideas – that consciousness is a process not a thing, that it’s a sort of product of the ongoing feedback and re-feedback processes in the brain – and that therefore it can’t be pinned down into one area. I like the idea that a memory (and by extension an imaginary phenomenon) isn’t encoded information but a sort of state of brain that might recruit as many specialist areas as are appropriate. And I like that his idea of how the brain works is rooted in embodiment. (Also I know that ‘what I like’ isn’t an acceptable criterion for scientific acceptance.. but I work on the level of poetic association).

Please correct this if it’s wrong, scientists – or point me at counterarguments!

It works well for the experience of watching the puppet. We are creating a sensation – or a phenomenon – from the information available and our predictions. We receive more information than normal from our mirror systems because some other data might be partial or absent. We use these systems to construct our ‘memory’ of the event. Even a fraction of a second later when we are relating a new  moment to the previous one, fresh perception has become (re)constructed memory. Both are still embodied and both exist inside us and with physical (which is to say emotional) engagement.

Or something like that? And what does it imply in terms of how we relate to (our ideas of) each other?


27 April 2017


It’s been a while since i put anything on here – I have been away with the serious businesses of puppetry and family, along with feeling further daunted by the volume of information I’ve been taking in. I’m very close to the next phase of the reading which will be to get stuck in to writing on puppetry, emotions and aesthetics, now with the added information that I’ve picked up from the psychological,  neurological and  philosophical papers.

The Girl on the Train

As I was on my way to the airport during the Easter break to go and check on some puppets, a family boarded the tube. The elder daughter, perhaps twelve or thirteen, stood a little apart from the others, leaning against the frame of the doors. She had her iphone headphones in and was holding the phone in one hand while she listened. She wasn’t watching the screen, but her whole body was enacting a very specific dance routine in miniature. Her focus was internal; but something in how specific the movements were made it clear that she wasn’t creating a dance to the music – she was remembering and copying a choreography that she was visualising.  (Can I be sure? No – but I was sure). One of the things that seemed striking was that she wasn’t really reproducing the movements she was imagining, but generating reduced versions of them – accurate in intention and accent, but smaller in scale. Whatever our  interpretations of the ‘action observation’ element of perception, it seemed striking to me that her mirror neurons seem to be firing in response to a completely imagined (in this case, remembered) stimulus – a dancer performing only in her mind’s eye was defined enough to activate the full system. “What’s she mirroring?” I asked myself. I’ve a bit more to understand about the role of imagination and how it can hijack our perceptual equipment.

What We Know

I’m getting on to the stage of reading through all my notes and trying to filter them into a manageable format. I have a wonderful opportunity to share these ideas with puppeteers at the O’Neill Conference this year and would like to make the participants the first guinea pigs in determining what about all of this might be interesting to a wider audience. For me,  an inquiry into perception opens up a  very enjoyable hall of mirrors – but for my show I will need something more like a mirror maze, that has an appealing entrance gate, an enjoyably designed route and a satisfying destination (perhaps an ice cream parlour) at its centre or exit. So far I am trying to make some documents that I vainly hope will be single pages: ‘What We Know’; ‘What We Think’; ‘What We Don’t Know’; ‘What Does The Puppet Do?’. I’m hoping that it will help me communicate (and determine for myself) what seems relevant, what seems interesting, and will perhaps suggest some ways forward for the workshop.

8 March 2017



The bulk of my reading and discussion to date has been to do with the processes of perception – of how we distinguish people from things, and how we interpret information about what those people are feeling or thinking. I’ve been struck how the research indicates how well these processes deal with gaps in that information – seeing and interpreting agency, action and emotion from often very sparse stimuli. This is something that we often exploit as puppet designers by simply leaving out whole parts of the puppet’s body (like here and here)

One of my original questions and provocations was to do with something else – what happens when aspects of the information are in conflict with one another?

In my example the most prevalent conflict is between the apparent agency of a character and its status as an object (say, in the case of an animated stick). Does anyone know who researches this kind of situation or phenomenon – or any papers on it? I think an understanding of how we resolve these perceptual conflicts would be very useful to me. Answers on a postcard. (or in the comments below)



How do we know about attention? In a way, my main premise is about quality of attention – that the audience watch puppets differently from the way they watch actors. My first thought had been that it might be the perceptual cues, absences and mismatches (like the conflicts above) that prompt a difference of attention. But it may also be the case that the novel nature of the puppet’s form prompts a shift in attention, regardless of the quality of the movement or how well it stimulates ‘agent recognition’ systems. I’m not sure what I mean by this, and yet I can feel the subtle gear change when I go from watching actors to watching puppets or animated characters. How might we measure or assess this?



The other big question is about the impact of the context . We know that we are watching a play, we know that it is appropriate here for characters to be represented in narrative – and this predicts our response to the animation of the puppet. I come across references to how knowledge cues influence animacy perception. Cross et al (2015) reference fMRI studies from 2001 and 2002 showing that participants who believe that an action is being executed by a human (rather than a program, robot or AI) show increased responses in social brain regions. This suggests that the presence of the puppeteer (or the knowledge that the puppeteer is there) is enormously important to how we interpret the movement of the puppet (as opposed to, say a robot with a limited, programmed set of responses). The audience know that the puppet’s responses are being selected by a human. Part of the act of belief in the puppet character’s agency is tied in absolutely with keeping in mind that it is a puppet under human control.