How do 100 people walk a tightrope together? An experiment in large scale joint action

How do 100 people walk a tightrope together? An experiment in large scale joint action Daniel C. Richardson (dcr@eyethink.org) Cognitive, Perceptual & Brain sciences, University College London Gower Street, London WC1E 6BT, UK Rick Dale (radale@memphis.edu) Department of Psychology, The University of Memphis 202 Psychology Building, Memphis, TN 38152, USA John Rogers (john@delosis.com) James Ireland (james@delosis.com) Delosis, 8 Grosvenor Road, Twickenham, Middlesex TW1 4AE, United Kingdom Abstract A lecture hall full of people played a computer game together. Their goal was to keep a tightrope walker balanced. Each had a handset that could deliver a left or right nudge. The tightrope walker was also pelted by tomatoes which knocked him off balance. Across several games, the difficulty of the task was changed by the frequency of tomatoes and whether or not they were visible. After each game, the participants rated their own and the group’s performance. We analysed the button presses of individuals, and quantified how they related to the moment by moment action of the group and movement of the tightrope walker. On successful games, participants were able to anticipate the behaviour of the group and kept the tightrope walker in equilibrium. Keywords: joint action; wisdom of crowds; group behaviour, situated cognition Introduction There is wisdom and beauty in a crowd. Galton (1907) studied competitions to guess the weight of a cow, a common game at village fares. He noted that the average response of the crowd usually equalled or bettered any of the individual guesses. We now know that if the faces of all those villages were averaged too, they would beat any individual villager in a beauty contest (Langlois & Roggman, 1990). These principles have been extended into business decisions, analysing markets and predicting political events (Surowiecki, 2004). In each case, the claim is that the average of group’s response is superior to individual’s judgements, even when those individuals are thought to be experts. The same idea applies to a large number of judgements made by a single person: one’s own average estimate is better than any single guess (Vul & Pashler, 2010). One explanation is that the biases that distort individual judgements (or facial characteristics) are roughly randomly distributed. Polling a large number of people or decisions evens out these distortions. The principle is that if incompetence is normally distributed, then the average response will be wise. But is the superiority of crowds restricted to wisdom? In all these cases, single judgements or measurements are being made in response to static problems or criteria. What about governing continuous action, when a stream of decisions have to be made in time, in response to changing circumstances? In short, there may be wisdom in a crowd but what happens when they have to act together? Around the time of Galton, people were very interested in ‘the mob’, and the possibility of understanding a crowd as if it were an organism with a single mind (e.g., Le Bon, 1896, Freud, 1921). Analysis of the behaviour of large groups became the domain of sociology and political science, however, as psychology focused experimental tools on the individual. Social forces themselves are studied in social psychology of course, but perception and action are typically studied in their absence. The laboratory cubicle of a typical cognitive psychologist is a lonely place. More recently, that has been changing. A diverse set of researchers have come to the realisation that perception, action and cognition cannot be fully understood by investigating single individuals (e.g., Barsalou, Breazeal & Smith, 2007; Robbins, & Aydede, 2009; Sebanz, et al 2006). Studies of situated cognition show that cognition ‘in the wild’ is intimately linked not only to representations of the external world, but also to the cognitive processes of others. For example, Hutchins (1995) observed the ways that navy navigators would distribute cognitive processes between themselves by using external tools and representations, such as maps and notations. Knoblich and Jordan (2003) gave a detailed analysis of the way that two people coordinate their actions. To be successful, participants had to anticipate both the movements of the objects in the game and the actions of their partner. In our experiment, over a hundred people played a computer game together. Our first goal was to see if the ability of crowds to make good judgements (Surowiecki, 2004) also meant that they could successfully act together in a dynamic task. Our second goal was to take predictions about pairs of participants acting together (Knoblich & Jordan, 2003) and see if they scale up to much larger groups.