Our storytelling nature

Jonathan Gottschall on the Heider-Simmel experiment as an illustration of the story paradox.

The human world is made of stories, not people. The people the stories use to tell themselves are not to be blamed. - David Mitchell, Ghostwritten

In 1944, the psychologist Fritz Heider and his research assistant Marianne Simmel produced a very short, very crude animated film. They cut geometrical figures out of cardboard and moved them around on transparent glass in the technique of stop-motion animation. The resulting silent film shows a small triangle, a big triangle, and a small circle moving busily around a rectangle. One side of the rectangle flaps open and closed, and sometimes the geometrical figures slide inside. At the end of the film, the small circle and the small triangle disappear offscreen, and the big triangle butts against the big rectangle until it breaks. Heider and Simmel showed the film to 114 research participants, who were simply asked to describe what they saw. (Before reading further, please take 90 seconds to watch the Heider-Simmel film on YouTube.)

I first saw the short film about 65 years after it was made and watched in delight as the simple geometry resolved itself into a classic three-act love story. Act One: The two lovers, Small Circle and Small Triangle, move side by side onto the screen. Act Two: Big Triangle decides he fancies Small Circle as well. He uses his pointy nose to wedge the lovers apart, then chases Small Circle into his house (the big flapping rectangle), where he tries to trap her in a corner. Act Three: To my relief, Small Circle slips past the lecherous Big Triangle and reunites with her mate outside. They race side by side around the house, with Big Triangle in hot pursuit. Finally, the lovers escape offscreen. Furious and frustrated, Big Triangle slams against the walls of his house until it collapses.

When I showed the film to my students, their reactions surprised and confused me. While many saw a love story like me, others were just as convinced that they’d seen a sordid family drama or a slapstick comedy along the lines of The Three Stooges. I should have been fascinated by the Rorschach element of the film, but I wasn’t – not at first. I was too frustrated by my students’ inability to see the ‘real’ story that I knew Heider and Simmel were trying to tell.

At the time, I’d only heard enough about the Heider-Simmel film to look it up on YouTube, but I’d never read the original scientific paper. When I did, I realised to my embarrassment that my students hadn’t been getting the film wrong; I had been. And it wasn’t until years later that I recognized the film as a powerful illustration of the story paradox: it reveals something wonderful about our nature as storytelling animals while also hinting at something profoundly scary – something like the root of humanity’s largest, deepest evils.

Although this wasn’t divulged in the paper of 1944, Heider later explained in his 1983 autobiography that he did have a hazy story situation in mind when he dreamt up his famous film. ‘As I planned the action of the film,’ Heider wrote, ‘I thought of the small triangle and the circle as a pair of lovers or friends and I thought of the big triangle as a bully who intruded on them’ (p.148). So rather than imagining a definite plot, Heider imagined an open framework consisting of a setup, a conflict, and a resolution, which could support a variety of basic plots, including Heider’s visions of either a buddy flick or a romance.

Despite its built-in ambiguity, the film nonetheless produces impressive convergence in interpretation. In the original experiment, for example, 97 per cent of the 114 people who viewed the film saw a story. Moreover, there were strong regularities in the stories people saw. First, even though the geometrical shapes look and move more like scurrying beetles, almost everyone automatically saw them not as bugs but as people in conflict. And most viewers automatically gendered the shapes in the same way – the circular female, the spear-headed males. On top of that, in a strong majority of cases people agreed on the basic protagonist–antagonist split: Big Triangle was the bad guy, and the two smaller figures were good. They also tended to give the shapes similar personality traits: Big Triangle was a bully, Small Circle was timid.

But there are equally impressive divergences in interpretation. For instance, although the most common story people see is indeed a love triangle, it’s not always my love triangle. Some viewers were convinced that Big Triangle was the aggrieved party, having been cuckolded by Small Triangle. Others thought that Small Circle wanted to be with Big Triangle. He, not she, was the unwilling party. But more times than not, viewers saw no love story at all. Some viewers saw a tale of domestic violence in which Big Triangle was abusing his family. Others saw Big Triangle as a harmless oaf being harassed by Lilliputian invaders. One viewer thought Big Triangle was a witch, trying to catch two children.

Over the years, I’ve conducted scores of informal replications of the Heider-Simmel experiment in my classes and in public lectures, with total numbers of participants stretching to many thousands. And I came to love the way viewers project idiosyncratic stories and meanings onto such simple geometry. The diversity of response shows that when we see the film we’re not experiencing story but are actually creating the story through a sequence of unstoppable cerebral reflexes.

Even when the film is run backward, and there can be no possibility that the psychologists had a definite plot or meaning in mind, people still see stories. When I watched the film backward, I expected to see my love triangle story running in reverse – like the bygone experience of hitting rewind on a VCR tape. Instead, I watched slack-jawed as my love story dissolved into a scene out of A Clockwork Orange in which the formerly predatory Big Triangle (now my stolid protagonist) is harried by a feral pair of anarcho-terrorists.

The point here isn’t limited to how people interpret simple animation. The point is that this is what storytelling animals are up to all the time: we’re trying to impose the meaningful and comforting order of story structure on the ambiguities of existence. And thanks to differences in our minds and experiences, we will reliably fail to see the same story – just like in the famous film.

Moreover, though our interpretations of a primitive cartoon couldn’t matter much less, this effect runs riot through domains of experience where it couldn’t matter much more. Especially because the stories we see can be divisive. When we experience chaotic events, we naturally construct stories to bring order to the chaos. And, as with all of the most typical responses to the Heider-Simmel film, we’re apt to resolve the chaos into a moralistic trinary of victims, villains, and heroes.

The story psychology revealed by the Heider-Simmel effect isn’t the root of all evil. But this humble experiment digs to the root of the most tragic type: the kind otherwise good people get swept up in. I’m referring to the human tendency to glom on to a story often for no good reason at all, to cling to it tenaciously, to let it structure our worldview, and to allow it to project patterns on the world that aren’t really there.
In all my screenings of the Heider-Simmel film, debates over the ‘true story’ have often been raucous but always friendly and light-hearted. People don’t get invested in their interpretations, not intellectually and not emotionally. But if you take the same narrative psychology laid bare by the film and you translate it to situations where (1) there’s a truth that can conceivably be determined, and (2) the stakes run higher than cartoon interpretation, people will dig in for their fictions and fight.

Sometimes it seems to me that what I call the Heider-Simmel effect – our tendency to all watch the same film and see different stories – explains everything about the roiling anger and confusion of modern life. And this effect, souped up by technological and cultural upheaval, helps explain why we now find it so hard to converge on consensus narratives about the basic shape of reality.

- This is an adapted extract from The Story Paradox: How our love of storytelling builds societies and tears them down, with kind permission from the publisher Basic Books.

- Jonathan Gottschall is distinguished research fellow at Washington & Jefferson College

Key sources
Heider, F. (1983). The Life of a Psychologist: An Autobiography. University Press of Kansas.
Heider, F. & Simmel, M. (1944). An Experimental Study of Apparent Behavior. The American Journal of Psychology, 57(2), 243-259.
Klin A. (2000). Attributing social meaning to ambiguous visual stimuli in higher-functioning autism and Asperger syndrome: The Social Attribution Task. Journal of Child Psychology and Psychiatry, 41(7), 831-846.
Ratajska, A., Brown, M.I. & Chabris, C.F. (2020). Attributing social meaning to animated shapes: A new experimental study of apparent behavior. The American Journal of Psychology, 133(3), 295-312.

BPS Members can discuss this article

Already a member? Or Create an account

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