Empathy Machines: Love, Guilt and Paracosmics in Interactive Characters
Hello everyone. This beside me is the largest fraction of my PhD project so far. I am deliberately showing you a rough, early prototype, and I do not want to say too much else about my work at this point. However, I believe with quite unacademic confidence that I can predict what you are seeing, and what is going through your minds, when you look at this collection of shapes and lines moving on a 2D field. No matter how many people I show this to, and no matter who those people are (whether academics, lay friends or random people on the internet), I can comfortably predict that this paltry, primitive abstract will coalesce in your minds into the impression of something greater; something alive, animate, and with purpose.
This inherent human tendency and compulsion is very useful to my work, and to the work of all artists, working in all disciplines. This entity that you see on this screen represents ten small .png files and about 200 lines of code; however, by combining them in an authorial, intentional manner I can reliably induce in you what amounts to a hallucinatory experience. Here, I mean ‘hallucinatory’ in the extremely abstract sense of seeing something which isn’t actually there. In this case, you are hallucinating a face, the features of an animal, a being, without explicitly being told to.
This is a well-known, primeval feature of the human brain, with its roots in the developmental stages of childhood; academics from the fields of literary theory, psychology, anthropology, paediatrics and many others have been fascinated for a long time by these tendencies to see life, and its attendant narrative, where there is none. For my purposes I have grouped these traits, and terms such as ‘animism’, ‘anthropomorphism’, ‘zoomorphism’, ‘personification’ and the more everyday ‘imagination’, under the general concept of paracosmics.
A paracosm is generally defined as an imaginary world held in the mind of an individual. Though it tends to refer to geographic hallucinations such as Tolkien’s Middle Earth or the detailed fantasies of certain outsider artists such as Henry Darger here, I believe that we can usefully apply it to any hallucinatory, ascriptive experience; when we imagine the characteristics of someone we have never met, picture a fictional character eating breakfast, see a face in a rock, daydream about flying over a fantastical land or ascribe animal characteristics to people (or indeed vice versa), we are creating a ‘cosm’; a hypothetical, hallucinatory space in which to model something that isn’t real.
I still find it amazing how little this ability needs in order for the imagination to be stimulated. This is an extract from a three-minute animation made in 1944, depicting three abstract shapes moving around a field; no different in composition and content to the ‘being’ which sits beside me here. This animation was shown to thirty study participants (all women, oddly), who were simply asked to describe what they were seeing. All but one participant spontaneously produced a narrative of the events unfolding, characterisation of each shape and its relationship to the others, even the emotions that they imagined them to be feeling. Many professed to feeling real emotions themselves; sorrow for the small triangle, anger at the large, and so on.
Moving on from these psychological and artistic studies, recently I have been branching out into a study of the paracosmic manifestation right across human culture; high to low, everyday to sacred. This will, I hope, allow me to explore other ways of thinking about my prototype here as a character, or as the suggestion of a character, onto which an audience will project their manifestations.
The famous computer scientist Joseph Weizenbaum, in his book Computer Power & Human Reason, takes a moment about halfway through to compare the modern computer with one of these manifestations. He describes an ancient technology which has reliably and consistently performed important tasks through interaction with human beings in the material world. At the end of this extremely long sentence, he reveals that he is talking about…
…the teddy bear.
Indeed, toys are an ancient technology, as old as our species; they are artefacts that induce a paracosmic effect in both children and adults, often through their incorporation of some deliberate design elements that are suggestive of character, individuality and agency. They are, in a phrase, empathy machines; built things designed, in their building, to reliably induce attachment and paracosmic engagement, almost to a ritual degree. Heider & Simmel’s animated shapes were exactly the same, as was this toy horse from Ancient Greece. Even today, beneath museum glass, it is an invitation to imagine the animal it represents, both in its shape and in the metaphor of movement produced by those spinning wheels.
The influential child psychologist Jean Piaget famously believed that children engage in this “widespread animism” throughout childhood; attributing thoughts, feelings, intent and life to inanimate objects, whether characterised or not. Whilst modern psychologists such as Nathalia Gjersoe have shown that children are actually quite able to distinguish between inanimate objects and living beings under laboratory conditions, even at very early ages, it is still evident that whilst children are playing with them, toys encourage them to create vivid paracosms, emotions and imaginings spontaneously. This is something which the play theorist Johan Huizinga and others have called the magic circle; the extrasocial space in which play, imagination and belief in things unreal is permitted, and outside which logic and utility might instead take precedent. We cannot deny the enormous emotional, creative and imaginative effect that comes from interacting with artefacts inside this magic circle; Pamela McCorduck writes about this in relation to another sort of cultural artefact, the clockwork automata popular in 17th century Europe: “we [were] not fooled for an instant, just enchanted.”
Toys have undergone a technological advancement in the last twenty or thirty years; the introduction of microelectronics has led to toys whose personalities and responses are not only imagined by the child, but are real and hard-coded. This was perhaps most culturally noticeable in the 1990s and early 2000s with the introduction of electronic pets such as the Furby and the Tamagotchi; toys which reacted back to children, which changed over time, which acted as if they had a life of their own; which were, in the words of a rather utopian article in Time magazine from the year 2000, “artificial life-forms”.
And it is in the power of such artefacts to induce empathy and paracosms that we start to wade into murkier ethical waters. It comes down to this: if our empathetic drive can be manipulated by almost any artefact, with only the barest suggestion of life or animacy or character, how easily might this drive be abused? How might that ‘magic circle’, in which we can be influenced by such artefacts, be grown or shrunk without us realising it, or moved into new arenas of our everyday lives? I think this question comes down to the purpose of that manipulation: why are we trying to convince people to ‘believe’ in something, to see life where there is none?
Looking to toys, especially in the last one hundred years, that purpose is commerce. The term ’empathy machine’ has never been more appropriate, and the introduction of reactiveness, of interactivity, in these latter-day ‘digital toys’ allow companies to even more tightly focus their efforts to invest children’s ideas, their emotions, and thus the buying power of their parents.
And it is thus inevitable that the emotions and ideas that are generated by interacting with a Furby or a Tamagotchi are, in the main, exploitative. In their design, they often only produce two emotions; love and guilt. Love come from the toy’s design, its appealing appearance, as well as the personality portrayed by the simple digital technologies at its heart. The guilt comes from the tight feedback loop in a child’s or adult’s interaction with it; the toys cry and whinge if they are not fed or played with, they grow grumpy or sad, and even die. The Tamagotchi manual is full of warnings and admonitions for people who do not clean up after their little pet, or feed them daily.
It is in this matrix of guilt and love, combined with the toys’ novelty and marketing narrative, that the Furby and Tamagotchi found both huge commercial success at the point of sale and ongoing engagement within the homes of their customers. The mania that surrounded the Furby when it was launched is well-known, and the phrase ‘Tamagotchi Effect’ was coined after research into a widely-reported phenomenon of obsessive and intense emotional attachment to the toy by children and adults alike. Many schools and workplaces banned them; I remember my own primary school doing so. Such was the power of the paracosmic, personifying tendency in those who played with them. This tendency is not, as many reported at time, a problem in and of itself; but it is a problem when that tendency is exploited for profit.
Such cultural artefacts are not merely economically influential; they can have political and social influence as well. It is not widely reported that the Tamagotchi was originally developed as a sort of training tool for young Japanese girls to be doting mothers. There have been many controversies (too many to list here) involving the enforcement of arbitrary gender divisions by toys, a real concern considering how influential these artefacts are in shaping perceptions of the world.
We cannot take the disingenuous position, either, that these effects only influence children. As I believe I have shown, our knowledge that something is not actually real does not mean that we cannot be influenced by it seeming real or alive. As Heider & Simmel’s study shows, even a few shapes can make us feel something fundamental; and when we feel something fundamental, we are vulnerable to influence.
In advertising there are countless studies that show how important emotional attachment is to brand loyalty; and personification, anthropomorphism and narrative technique are some of the most effective tools that we have for generating paracosms, inference and emotions. Giving your brand or product an identity, the hints of a personality, an invitation to embody, is a popular way to do this.
In a recent paper with the excellent title Deconstructing the Meerkat, researchers from Queen Mary University looked at the ridiculously successful use of cute, funny, anthropomorphised meerkat characters to sell car insurance, and went on to propose a set of principles for creating such characters in the future. This manipulation and its purpose must be noted in terms of advertising ethics. When you can induce people to choose your website over all others with the promise of a free doll, based on a character fantasy that has nothing to do with what you are selling, we cannot deny that the paracosmic artefact can influence big kids as well as small ones.
We might extrapolate out from commerce and move into the more global, political effects of inducing the paracosmic tendency in human beings, with the purpose, perhaps, of sidestepping logic and rational debate and promoting in their place patriotism, jingoism and non-rational narratives, in order to influence voting patterns or law. Here is an example of the magic circle growing from the size of child’s bedroom to the size of a nation; Uncle Sam is a national hallucination. As I have already noted, the fact that Uncle Sam is not real does not affect the complex and emotive myth that he represents, and in deploying him as narratively logical, just another person, and inducing personification in the population, what he represents suddenly becomes simpler, more appealing and more human. In the narrative and character of Uncle Sam, as elsewhere, purpose is everything.
Of course, purposes can be positive as well as negative. While commonly a dirty word when it comes to nature writing or natural science, the anthropomorphism that is at the centre of the advertising campaigns of animal charities are important drivers of revenue for those charities. In imagining and anthropomorphising the internal states of those animals we are driven to sympathy, empathy, and the giving of charity. Some may argue that this is commercial manipulation like any other, and there’s a debate there, certainly.
In the medical sphere, both animals and anthropomorphised robots are used with the purpose of providing succour, companionship and mental engagement for the very old and the very ill; acting as stimuli for the atrophied paracosmic abilities of these people. At a recent event I was lucky enough to meet Robin here, a project by roboticists from the University of Hertfordshire to simulate the symptoms of diabetes in a robot, with the (now-proven-successful) hypothesis that caring for a ‘sickly’ robot might help young diabetes sufferers manage their own condition without supervision.
So, if we accept the incredible associative and paracosmic powers of the human mind, then the ethical ramifications of manipulating those powers comes down to purpose. So what is my purpose, my agenda, with my character? Well, my purpose is I hope, artistic, subversive and self-reflexive.
I am interested in exploring these paracosmic tendencies in human beings, in our unstoppable urge to see agency wherever we look, and determining how this might relate to the construction of fictional characters in any medium. I want to use the power that characterised artefacts have over us to critique the practice, and push it to its very limits. Though I have only just begun my project, already my creature, my collection of shapes, is influencing you in subtle yet fundamental ways! This is fascinating to me, and I plan to take it as far as I can. However, I hope that I have shown that there are far worse purposes for manipulating the inbuilt, inescapable tendencies of my audience.
Thank you very much, and I leave you with Jeff Winger, putting it far more pithily than I have here.